Launch and Recovery Systems For Unmanned Vehicles Onboard Ships. A Study and Initial Concepts

Launch and recovery systems for unmanned
vehicles onboard ships. A study and initial

concepts.
Centre for Naval

Architecture
MARCUS ERIKSSON
[email protected]
+46 703-925688
PATRICK RINGMAN
[email protected]
+46 703-5085073
Course: SD270X – Naval Architecture Master’s Thesis

Magnitude of work: 2×30 Cr.
Date: 4/10-2013
Version number: 1.0
Reviewed by: Per Thuvesson, Tobias Petersson
Intentionally left blank.
Abstract
This master’s thesis paper is an exploratory study along with conceptual design of launch and recovery systems
(LARS) for unmanned vehicles and RHIB:s, which has been conducted for ThyssenKrupp Marine System AB in
Karlskrona, Sweden. Two concepts have been developed, one for aerial vehicles (UAV:s) and one for surface
and underwater vehicles (USV, RHIB and UUV). The goal when designing the two LARS has been to meet the
growing demand within the world navies for greater off-board capabilities. The two concepts have been designed
to be an integrated solutions on a 90 m long naval ship and based on technology that will be proven in year
2015-2020. To meet the goal of using technology that will be proven in year 2015-2020, existing and future
possible solutions has been evaluated. From the evaluations one technique for each concept was chosen for
further development.
In the development of a LARS for aerial vehicles only fixed wing UAV:s have been considered. The concept
was made for a reference UAV based on the UAV Shadow 200B, which has a weight of 170 kg. The concept
that was developed is a parasail lifter that can both launch and recover the reference UAV effectively. In the
development of a system for surface and underwater vehicles only vehicle lengths in the span 1-12 m have been
considered. The concept that has been developed is a stern ramp that uses a sled to launch and recover all three
vehicle types. The two concepts that has been developed are in an early design state and the papers results should
therefore be seen as an estimation of what each system are capable of performing.
i
ii
Sammanfattning
Detta examensarbete är en utforskandestudie och konceptgenerering av hanteringssystem för obemannade
farkoster och RHIB-båtar. Hanteringssystemen som arbetet har behandlat har till uppgift att starta/återta
alternativt sjösätta/ta upp farkoster från ett fartyg. Arbetet har genomförts i samarbete med ThyssenKrupp
Marine System AB i Karlskrona. Två koncept har utvecklats, ett för flygande farkoster (UAV) och ett för yt- och
undervattens-farkoster (USV, RHIB och UUV). Målet med koncepten har varit att tillmötesgå det växande
behovet hos världens flottor att kunna hantera mindre farkoster ombord på fartyg. Koncepten har designats för
att vara integrerade lösningar ombord på ett fartyg med en skrovlängd på 90 m och basera sig på teknik som kan
vara beprövad år 2015-2020. För att tillmötesgå målet med att basera koncepten på teknik som kan vara
beprövad år 2015-2020 har existerande samt i framtiden möjliga lösningar utvärderats. Från utvärderingen har
sedan en lösning till varje koncept valts för vidare utveckling.
I utvecklingen av ett hanteringssystem för flygande farkoster har enbart farkoster med fasta vingar behandlats.
Konceptet som utvecklades formades kring en referens-UAV som är baserad på UAV Shadow 200B som har en
vikt på 170 kg. Konceptet som utvecklades består av en flygande skärm som bogseras efter fartyget och kan både
återta och starta referens UAV:n. I konceptutvecklingen av ett hanteringssystem för yt- och
undervattensfarkoster har farkost längder mellan 1-12 m behandlats. Konceptet som har utvecklats är en
akterramp som använder en släde för att sjösätta och ta upp alla tre farkosttyperna. De två koncepten som har
utvecklats befinner sig i ett tidigt designstadie och resultatet av arbetet ska ses som en uppskattning av vad
systemen kan prestera.
iii
iv
Preface
Acknowledgments
We would like to thank Per Thuvesson and Tobias Petersson for the support and feedback during the project. We
would also like to thank all the warm hearted people at the TKMS office in Karlskrona for their warm welcome
and making our stay in Karlskrona a pleasant one.
v
vi
Contents

Introduction ............................................................................................................................................................. 1
Background ......................................................................................................................................................... 1
Unmanned vehicles ............................................................................................................................................. 2
Unmanned Aerial Vehicle – UAV .................................................................................................................. 2
Unmanned Surface Vehicle – USV ................................................................................................................. 6
Unmanned Undersea Vehicle – UUV ............................................................................................................. 6
Unmanned Ground Vehicle – UGV ................................................................................................................ 8
Other off-board vehicles - RHIB ..................................................................................................................... 8
Scope ................................................................................................................................................................... 9
Delimitations ..................................................................................................................................................... 10
Methodology ......................................................................................................................................................... 11
Work process .................................................................................................................................................... 11
Analytical Hierarchy Process (AHP) ................................................................................................................ 12
Concept development ........................................................................................................................................ 14
Literature study and data collection .................................................................................................................. 15
Concept generation – UAV launch and recovery .................................................................................................. 16
Launch and recovery of a rotary winged UAV ............................................................................................. 16
Technology review - Launching a fixed-wing UAV ..................................................................................... 17
Technology review - Recovery of a fixed-wing UAV .................................................................................. 21
Evaluation of UAV-launching techniques .................................................................................................... 24
Evaluation of UAV-recovery techniques ...................................................................................................... 29
Evaluation results - UAV launch and recovery techniques ........................................................................... 34
Concept analysis – UAV launch and recovery ...................................................................................................... 35
Parasail lifter - concept description ................................................................................................................... 35
Input and Constraints ........................................................................................................................................ 36
Forces on the parasail ........................................................................................................................................ 37
Necessary altitude for launching ....................................................................................................................... 38
Towing line ....................................................................................................................................................... 39
Winch ................................................................................................................................................................ 40
Arresting mechanism/hook ............................................................................................................................... 41
Time scheme ..................................................................................................................................................... 41
Estimating operability ....................................................................................................................................... 44
Integration on a ship .......................................................................................................................................... 45
Concept generation – Surface & Underwater launch and recovery ...................................................................... 46
Identify needs .................................................................................................................................................... 46
Technology review – LARS for surface and & underwater vehicles ................................................................ 48
Evaluation of surface and underwater – launch and recovery techniques ..................................................... 50
Evaluation results – Surface & underwater launch and recovery techniques ................................................ 55
Generate Concept .............................................................................................................................................. 56
Sled design .................................................................................................................................................... 59
Concept analysis – Surface & Underwater launch and recovery .......................................................................... 60
Ramp slope........................................................................................................................................................ 60
Transom opening............................................................................................................................................... 62
Sill water depth ................................................................................................................................................. 64
Launch and recovery time and procedure ......................................................................................................... 66
Space needed ..................................................................................................................................................... 67
Concepts and solutions (results)............................................................................................................................ 68
Parasail UAV lifter – concept results ................................................................................................................ 68
Description of components – Parasail lifter .................................................................................................. 68
Operational procedure – Parasail lifter .......................................................................................................... 69
Discussion and conclusions – Parasail lifter ..................................................................................................... 70
Concluding remarks – Parasail lifter ............................................................................................................. 71
RHIB, UUV and USV –concepts ...................................................................................................................... 72
Operational procedure – Stern ramp system ................................................................................................. 72
Discussion – Stern ramp system.................................................................................................................... 73
Authors work and responsibility division ............................................................................................................. 75
References ............................................................................................................................................................. 76
Appendix A - Weighting of evaluation categories

Appendix B – Aerodynamic coefficients used in parasail calculations
Appendix C - Estimated operability charts for Parasail lifter
Appendix D - Beufort Wind Force Scale V.s. Pierson-Moskowitz Sea States
Abbreviations
CTOL – Conventional Take-Off and Landing
FRB – Fast Rescue Boat
HALE – High Altitude Long Endurance
Heli-deck – Short for “Helicopter deck”
IR – Infra-red
L&R – Launch and Recovery
LHLRS – Light Harpoon Landing Restraint System
MALE – Medium Altitude Long Endurance
MCM – Mine Counter Measure
NATO – North Atlantic Treaty Organization
OTS – Of the shelf (product availability)
PTOL – Point Take-Off and Landing
R&D – Research and Development
RAS – Replenishment at Sea
RATO – Rocket Assisted Take-Off
RHIB – Rigid Hull Inflatable Boat
SAR – Search and Rescue
SOLAS – Safety of Life At Sea (UN regulations)
SPARS - Ship Pioneer Arresting System
UAS – Unmanned Aircraft System
UAV – Unmanned Aerial Vehicle
USCG – U.S. Coast Guard
USV – Unmanned Surface Vehicle
UUV – Unmanned Underwater Vehicle
VTOL – Vertical Take-Off and Landing
VUAV – Vertical Take-off and Landing UAV (see UAV)
ix
Introduction
1
The use of marine based unmanned vehicles are at an unprecedented level of growth and with this a noticeable
trend is seen in the demand on ships capability to launch and recover these vehicles. Standardized interfaces are
not common as the variety in the vehicles is too large and the development of the vehicles is at a faster rate than
the life-span of a ship.
The complexity in the integration of unmanned vehicles is even more present when it comes to naval ships as the
technology requirements are usually over-all higher and space is very limited due to the many advanced systems
that are needed on board. Apart from the physical aspects of the integration it is sought after to have solutions
that are considered flexible, cost efficient and have a wide envelope of operability.
A disparate range of technical solutions are being developed to cope with the integration and allowing the launch
and recovery of unmanned vehicles. The solutions vary depending on the type of ship and of course, the type of
unmanned vehicle that it is designed to handle. A major aspect in a successful integration is to have launch and
recovery system in mind at an early point of the design of the ship.
This thesis is done for ThyssenKrupp Marine Systems AB (TKMS) as an exploratory study of systems for launch
and recovery of unmanned vehicles that are considered to be relevant for integration on board of littoral naval
ships. These include aerial, surface and underwater vehicles. The study leads to a preliminary development of
concepts that are considered to be possible for future integration.
Background
Coming from an era when blue water strategic capabilities was the major focus of the world’s prominent navies
the operational focus has changed. The focus is instead on a littoral strategy, where the navies act as a policing
force closer to the coast lines rather than an offensive force out on the oceans. With this a noticeable trend in the
demand in the outfitting of warships is seen. [1]
When shifting focus to the littoral zones the use of unmanned vehicles have increased. Naval ships are becoming
more of a multi-purpose platform that should be able to partake in a big variety of mission scenarios. Unmanned
vehicles offers a lot of advantages, they cannot only strengthen the ships awareness but also increase the crews
safety by allowing them to be kept at a safe distance in dangerous missions. By having the possibility to change
the vehicles carried by a ship its mission capabilities can be increased.
With the increasing use of unmanned vehicles comes an increase in demands on the ships handling systems for
these vehicles. As the number and types of unmanned vehicles increases, a need for an integrated system that can
handle a big verity of vehicle types and sizes have emerged. Such system would have to handle a wide spectrum
of different vehicles. The different types of unmanned vehicles that exist today are further explained in the
following section.
∙ 1. Introduction ∙ 1
Unmanned vehicles
Whether it is on land, in the air, on or under the surface there might be times when a task is too dangerous,
inconvenient or straight out impossible for humans to perform. It can be a hazardous environment, something
that takes a long time to do or just that the place is inaccessible that hinders a certain task to be done. For this
reason there have for a long time been a development of unmanned systems, or vehicles, that can do these tasks.
With advancement in technology these vehicles have become more efficient and more popular than ever.
Advanced programming, better components, sensors with high precision and better understanding of the
technology leads to unmanned vehicles not only being remote operated, but also gaining the capability of being
autonomous or semi-autonomous. Below follows a short description of each type of unmanned vehicle.
Unmanned Aerial Vehicle – UAV
”Unmanned Aircraft Systems are powered aerial vehicles that does not carry a human operator, uses
aerodynamic forces to provide lift, can fly autonomously or by remote piloting, can be expandable or
recoverable, and can carry a lethal or non-lethal payload. Ballistic or semi ballistic vehicles, missiles and
artillery projectiles are not considered unmanned aerial vehicles.” [2]
In this definition the term used is Unmanned Aircraft System (UAS) that is the conventional term used when
describing the whole system that is the vehicle, control station and surrounding equipment. Common alternative
names for these crafts are remotely piloted aircraft (RPA), remotely operated aircraft (ROA) or as they are often
called by the media, drone [3]. In this report the term that is used is unmanned aerial vehicle (UAV), which
refers to the vehicle itself and not the complete system.
As early as 1898 in the Spanish-American war UAV:s were used as means of military reconnaissance. Though
the technology used was not more advanced than a kite with a camera attached to it the contraption fulfilled the
purpose of being able to get a view of something that the soldiers on the ground could not see [4] . Much has
happened since then and the UAV:s of today are far more advanced. Development in the fields of electronics,
materials, propulsion and more has led to technical possibilities such as autonomous navigation, large operating
distance, long endurance and high resolution sensors (cameras) of different types. These are just a few of the
many features that UAV:s can have today.
Within military use UAV:s are considered as a great force multiplier as they offer many advantages. Commonly
these advantages are attained at a lower risk and a lower cost than if a corresponding piloted aircraft would do
the same task. Naval use is no exception. Typical uses for the navy include [5]:
 Shadowing enemy fleets.
 Decoying missiles by the emission of artificial signatures.
 Electronic intelligence.
 Relaying radio signals.
 Protection of ports from offshore attack.
 Placement and monitoring of sonar buoys and possibly other forms of anti-submarine warfare.
 Optical surveillance and reconnaissance.
Even though that the most common view of an unmanned aerial vehicle is considered to be a portable drone
carried by a soldier or a rather large pilot-less plane there are types within the whole span of these two. A great
variety with different technologies and purposes. There are many ideas on how they should be categorized.
Weight, maximum altitude, operational range or air time capabilities are some of the features that can
characterize an UAV. An overall classification is listed by NATO according to Table 1 below.
Table 1. NATO categorization of UAV:s. [6]

Class Category Operation altitude Examples
Strike / Combat 20,000m Predator (U.S.)
III (> 600kg) HALE 20,000m Global Hawk (U.S.)
MALE 14,000m Heron (U.S.)
II (150kg-600kg) Tactical 3,000m Örnen (SWE)
Small (15kg-150kg) 1,500m Scan Eagle
I (< 150kg) Mini (<15kg) 1,000m Falken (SWE)
Micro 60m Black Widow
This is a very simplified way of categorizing as it only takes into consideration the altitude and the size/mass of
the vehicle. More comprehensive lists exist, but in most cases these are government specific or too detailed to be
relevant for this report. Instead a short explanation of some of the general characteristic technologies is
presented.
There are different ways for UAV to navigate while in the air. One way is to have an operator controlling the
vehicle remotely with a steady link in between. This is often a downside if stealth properties are of importance
since radio signals can be detected. Another measure of controlling the UAV is to actually not control it. E.g. to
make use of autonomous navigation where the UAV is assigned objectives that it can carry out by itself. This
eliminates the problems with having radio signals that can be detected.
A major difference (at least for the subject of this report) is how the UAV take off and land. Here the systems are
generally divided into conventional take-off and landing (CTOL) [7] and vertical take-off and landing (VTOL).
For naval purposes there is often a limitation on available runway, especially for frigates and ships of smaller
size hence the term point take-off and landing (PTOL) [8] is evolving. In between these two is short take-off and
landing (STOL). These categories are in many cases connected to the type of aircraft. Fixed-wing UAV
normally fall under CTOL, STOL or PTOL -category while rotorcraft or tilt-rotor UAV fall under the VTOL -
category.
To get a better grasp of what the different categories mean in the design of UAV some examples are listed
below.
Skylark I
Skylark I or UAV02 Falken in Swedish use, have been used by the Nordic Battle Group and by Swedish ground
forces contribution in Afghanistan [9]. The size and weight of this system defines the UAV as a class I UAV
according to Table 1 and is considered as a mini-UAV. The complete system includes an operating station and
three planes and is estimated to weigh 45kg altogether [6]. The launch is easily carried out by throwing the plane
into the air or by “sling-shooting” it with a rubber band, hence it does not require any complicated launch
system. Recovery is not done by conventional means as it is designed to perform a deep-stall maneuver just
before it is about to land. With this procedure the runway needed is decreased to a near zero. This can be
combined together with a line stretched up in the air that catches the UAV. This solution helps to prevent the
plane to crash after the deep stall maneuver. These take-off and landing procedures fall under the PTOL
category. Sensors include a normal camera for visual feedback or IR-camera for night missions.
Skylark I is mentioned here because it has been used by the Swedish ground forces and the fact that it represents
the size and endurance level well, but it should be noted that it is being replaced by Wasp III and Puma, both
produced by AeroVironment.
Table 2. Data for Skylark I. [11]

Producer: Elbit Systems, Israel
Type: Fixed Wing
Wingspan: 2m
Max. Starting weight: 5 kg
Payload: 1.2 kg
Engine: Electric
Navigation: GPS
Max. Velocity: 70 km/h
Operational Range: 10 km
Endurance: 1.5 h
Launch scheme: By hand or rubber band
Recovery scheme: Deep-stall maneuver, line,
hook.
Figure 1. UAV02 Falken. Pic. from wikipedia.com.
Shadow 200B
In 2011 the Shadow 200B was taken into use by the Swedish armed forces under the name and thus replacing the
UAV01 Ugglan. Two systems were procured, each consisting of one operating station, four planes and a
launching ramp [10]. This is to allow for continuous surveillance 24h a day with no intermission. The launch
includes the use of a hydraulic ramp and the recovery is conventional, but can be combined together with a
line/hook. Sensors are, like the ones on the smaller UAV02, a normal camera and an IR-camera.
According to the NATO categorization in Table 1 Shadow 200B (Örnen) is a class II UAV. Because of its
tactical character with longer endurance this is probably a more interesting class of UAV than the smaller
Skylark I when it comes to naval use.
Table 3. Data for SHADOW 200B. [11] [12]
Producer: AAI Corp.
Type: Fixed Wing
Wingspan 4.3 m
Max. Starting Weight: 170 kg
Payload: 27.5 kg
Engine: Combustion Engine
(MOGAS)
Navigation: Autonomous GPS
and radio backup
Operational Range: 125 km
Endurance: Up to 9 h
Figure 2. UAV03 Örnen. Pic. from forsvarsmakten.se. Max. Altitude: 4,600 m
Launch scheme: Hydraulic ramp
Recovery scheme: Conventional
together with hook
Skeldar V-200 Maritime

Skeldar V-200 M (Maritime) is a rotorcraft in the shape of a helicopter developed by the SAAB group. It is
based on CybAero’s former platform APID 55. Due to the fact that it is a rotorcraft instead of a fixed wing it
supports vertical take-off and landing and is also able to hover at the same position in the air [6].This opens up
other mission capabilities than the conventional surveillance. Examples of other capabilities can be target
assessment by laser guidance and logistics support. To support different capabilities the platform is easily
accessible and allows simple switching of payload.
Table 4. Data for Skeldar. [13]

Producer: SAAB
Type: Rotorcraft (VTOL)
Rotor diameter: 4.7 m
Max. Starting weight: 230 kg
Payload: >30 kg
Engine: Heavy fuel, 2-cyl
625 cc
Navigation: Autonomous GPS,
Operator controlled
Operational Range: >100 km
Endurance: >6 h
Max. Altitude: >3,000 m
Launch scheme: VTOL
Recovery scheme: VTOL Figure 3. Skeldar. Pic. from SAAB group’s imagebank.
In this thesis VTOL UAV:s are not a part of the scope as the surrounding equipment needed for launch and
recovery is relatively minor. A brief description of launch and recovery of VTOL UAV:s is however included.
Unmanned Surface Vehicle – USV
Unmanned surface vehicles (USV) are any vehicles that operate without a human occupant on the water surface.
USV:s are used in both civil and naval applications. One advantage that USV:s have compared to manned
vehicles is a much longer endurance. One example of a civilian long endurance USV is the Wave Glider
produced by Liquid Robotics. The Wave Glider is propelled with wave- and solar- power, which enables it to
operate for years without refueling. A Wave Glider can, depending on model, travel with speeds of 0.5-1.7 kn
and are partly used to gather real time information about the oceans [14].
USV:s for naval applications can roughly be divided into two different categories, fast USV:s equipped with
weapons and low speed USV:s used for MCM [6]. One example of a naval USV is SAM made by the Swedish
company TKMS. SAM is a slow moving MCM-unit used for minesweeping in littoral zones and can be remotely
or autonomously operated. The use of unmanned vehicles in MCM enables personal and ships to be kept at a
safe distance from the sea mines. A fast going USV, made by TKMS, is Piraya. The Piraya can be used for
surveillance in coastal areas and several Pirayas can be controlled, at the same time, from one single tactical
operations console.
Table 5. Data for Piraya. [15]

Producer: TKMS
Length: 4m
Beam: 1.4 m
Displacement (full load): 400 kg
Speed: 20+ kn
Range: 60+ NM
Hull: Composite, V-shaped
Propulsion type: Outboard engine
Payload capacity: 100 kg
Figure 4. USV Piraya. Pic. from kockums.se
An ongoing trend at producers of USV:s is to develop multi-purpose vehicles [6]. The same USV:s should for
example be able to perform tasks such as harbors surveillance, coastal surveillance and protect surface vessels.
This trend demands a modular design with equipment that is easy to change.
Unmanned Undersea Vehicle – UUV

UUV’s are used in a big variety of purposes. In the offshore business UUV:s are scanning the seafloor in search
for oil but are also used to check pipelines for eventual damages. In naval applications UUV:s are used for MCM
and surveillance since they are very hard to detect.
The Swedish navy is using an AUV62 for MCM and rapid environmental assessment. AUV62:s ability to find
sea mines has proven to be more accurate than conventional methods [16]. It has found sea mines in areas that
earlier had been cleared by a conventional MCM-vessel. The AUV62 is also used when practicing anti-
submarine warfare. AUV62 can pretend to be a submarine by sending out the same noise as one. By using a
UUV instead of an actual submarine the cost for practicing anti-submarine warfare can be reduced.
There are different ways to categorize UUV:s. The one used in this report is the same as in [6] where UUV:s are
divided in to four different categories based on weight and endurance.
Table 6. The four categorize of UUV:s.

Designation Characteristics
Man-Portable class Weight10-50 kg, Endurance 10-20 h
Light Weight Vehicle (LWV) class Weight <250 kg, Endurance 20-40 h
Heavy Weight Vehicle (HWV) class Weight <1,500 kg, Endurance 40-80 h
Large Vehicle class Weight ~10,000 kg, Endurance >400 h
To illustrate the big span of UUV:s two examples is presented. The first example Iver2 is a UUV from the man-
portable class and is a “of the shelf product”. Iver2 is developed for general survey work, research and
environmental monitoring in coastal areas [17].
Table 7. Data for Iver2. [17]

Producer: Ocean server
Class: Man-Portable
Length: 127 cm
Hull diameter: 14.7 cm
Displacement: 19 kg
Speed: 1-4 kn
Max. operating depth: 100 m
Endurance: 14 h @ 2.5 kn
Figure 5. UUV Iver2. Pic. from auvac.org.
The second example is a large class UUV, Theseus. Theseus was originally constructed to lay long fiber-optic
cables in the artic zone [18].
Table 8. Data for Theseus. [18]

Producer: International Submarine
Engineering Ltd.
Class: Large vehicle
Length: 10.7 m
Hull diameter: 127 cm
Displacement: 8600 kg
Speed: 4 kn
Max. operating depth: 2000 m
Endurance: >1360 km
Figure 6. UUV Theseus. Pic. from ise.bc.ca
Unmanned Ground Vehicle – UGV
These type of systems are not considered to be relevant for naval purposes, but are still mentioned to give the full
picture of what unmanned vehicles are about.
UGV:s are mainly used by ground troops as assistance in tasks that are often considered to be too dull, dirty or
dangerous, often referred to as triple D [19]. Like the aerial or water counterparts the mission objectives are
often surveillance or reconnaissance but the category can also be used for disarming mines and bombs or to carry
equipment that is too heavy for a person to carry.
After the earthquake that caused a nuclear catastrophe in Fukushima several different types of UGV:s have been
used to aid in the aftermath of the accident. The company iRobot re-outfitted the military used PackBot to be
able to survey and measure the environment inside the highly radioactive buildings. For more hands-on tasks
such as clean-up of rubble and debris the heavy-duty UGV BOBCAT from QinetiQ have been used also at
Fukushima [20].
As mentioned UGVs are not considered as necessary nor relevant for naval use and thus this category will not be
handled in this report.
Other off-board vehicles - RHIB

Although unmanned vehicles offers a lot of advantages the need for manned vehicles still exists. Most naval
ships today carries as rigid hull inflatable boat (RHIB), to be used in mission and serve as rescue boat.
A RHIB is a boat constructed with a solid hull and inflatable tubes as gunwale. RHIB:s are lightweight and
seaworthy, which makes them perfect as workboats on larger ships. “It has become a noticeable trend in warship
design to allow for the carriage of greater number of off board vehicles and boats” [1]. The size of boats carried
by naval ships is increasing and there is an “aspiration for 11 m or even 14 m RHIB:s in the future” [1]. The
number of RHIB:s on modern warships is also increasing. In one mission up to four RHIB:s can partake and at
least one boat has to stay aboard or close to the mother ship to support helicopter operations and man overboard
situations. Therefore the needs for manned vehicles have not yet been excluded from naval ships.
Scope
The aim of this thesis is to examine the impact that systems for launch and recovery of unmanned vehicles have
on naval ship design and derive concepts of solutions that are possible in the not so distant future (2015-2020).
These are to be fitted on a littoral naval vessel (90 m) and thus are subjected to the limitations that this brings.
This involves:
 Examining what unmanned vehicles are relevant for fitting on board this type of ship.
 Identify existing and possible future solutions that handle launch and recovery of these vehicles and
state issues and attributes of these.
 Establish design drivers for the integration of LARS and compare the different solutions performance in
these.
 Generate early stage conceptual designs of LARS for each category of unmanned vehicles.
 Analyze the concepts in terms of feasibility that can reconnect to the design drivers.
 Define the concepts to be able to reconnect with the design drivers.
The purpose of doing this is to get a view of what the future can hold when it comes to integration of LARS on
board ships. Especially naval ships. TKMS interest in this is continue having an up to date awareness of this. A
ship building company needs to consider this at an early stage in the ship design process as late re-designs or
retrofitting are time consuming, costly and likely to compromise the ships operational effectiveness.
Delimitations
In this section the delimitations made in the project is presented. Common to all work performed is that profound
analysis such as CFD, FEM are at the intended level of development considered unnecessary and thus refrained
from.
The delimitations made for the part of the project concerning aerial vehicles are:
 Only systems that are considered relevant for use on ships in the range of 90 m in length.
E.g. corvettes and frigates that are limited by not having a conventional runway.
 Only launch and recovery systems that handle fixed wing UAV:s are considered. VTOL
vehicles are refrained from.
 The type of UAV that is used as a reference in evaluating and analyzing the systems is
limited to TUAV:s and larger small UAV:s according to the NATO-categorization in Table
1.
The delimitations made for the part of the project concerning surface and underwater vehicles are:
 Only consider davits and stern ramps as possible solutions for LARS of surface and
underwater vehicles.
 Ship motions are not to be calculated. If needed the basic motion limiting criteria stated in
Table 12 should be used as worst-case scenario.
 Humans are considered to be the limiting factor in terms of accelerations; therefore the
acceleration criteria stated in Table 13 applies to all vehicles even unmanned.
 For surface vehicles V-shaped hulls are only considered and for underwater vehicles round
hull form are only considered.
Methodology
Solving a problem or obtaining new knowledge is the most common objective when doing the work that is a
2
master’s thesis. The tool to attain this new knowledge or solving a problem is called a method [21]. In this
chapter the method used in this thesis is explained so that the reader can determine if the “tools” are accurate and
relevant and thus if the information given is correct. The chapter is divided into subsections that describe: the
overall work process throughout the thesis, evaluation method used, the process of the concept development and
the literature study that is carried out.
The thesis was done in collaboration together with ThyssenKrupp Marine Systems AB (TKMS) in Karlskrona,
Sweden (formerly known as Kockums AB). It started out as an officially announced subject but it went through
some refinement and development after discussions. It can be seen as an exploratory study of which the purpose
is to gain further knowledge of the systems in question. At the same time it is a normative study as it is supposed
to work as a foundation or reference in future design work [22].
The aim of this thesis is to explore the field of technologies for launching and recovering unmanned vehicles that
are subjected to being used onboard ships. Included are also manned boats as they are considered to utilize the
same devices for launching and recovering as unmanned water based vehicles. From the explored systems
conceptual designs are developed that are meant to act as references and to be able to evaluate feasibility.
Work process
The work process of the thesis project can be described as the approach when conducting research. Several steps
need to be by past. These steps are condensed into the list that is shown below [23]. All steps are not in
chronological order as some are intertwining and this should be seen as an outline of the process.
- Attaining fundamental knowledge

Due to the inexperience within the specific branch that the thesis is carried out in the authors started by
attaining fundamental knowledge that was needed. This can be called a pre-study and can be seen as part of
the literature study that is explained later in this chapter.
- Determining the subject of study

Together with supervisors at TKMS and from the information gained from the pre-study the authors decided
on a subject of study.
- Planning the approach and methodology of the work

Before starting to look deeper into the subject of study the approach was planned. In this step it was
determined that the work should be divided into different segments. I.e. the literature study, the evaluation
and presentation of different technologies and the concept development.
- Implementation
This part is about the actual work of the thesis. The major parts in the implementation were literature study,
evaluation and concept development. This step intertwines with the Evaluation/Analysis -step since this was
a part of the concept development.
- Evaluation/Analysis
Firstly an evaluation is carried out between different types of technologies and solutions. These are
evaluated using Analytical Hierarchy Process (AHP) that is described in the section below. From the AHP
results concepts are developed and analysis of physical aspects are carried out to determine the concepts
feasibility.
- Present and report.

The thesis work is summarized into this report that is to be used as a part of the examination process at KTH
and also for TKMS to be used as pleased.
∙ 2. Methodology ∙ 11
Analytical Hierarchy Process (AHP)
Making a multi-criteria decision in categories that cannot be measured with the same scale can be problematic.
AHP is an analytical approach to aid in such decision making. The method proposes that pairwise comparisons
should be performed to create a hierarchy based on relative comparisons.
Here follows a brief introduction to the methods underlying math and the interface, which has been used in this
paper, when applying the method. A more thorough explanation of the method is presented in [24]. The
interface, that has been used, is developed in house at TKMS.
To illustrate how a multi-criteria decision is made, the decision is divided into three levels. Level 1, the decision
to be made, level 2, the criteria that the decision will be based on and level 3, the solutions of which one is to be
chosen.
Figure 7. The three decision levels.
The first step in employing the method is to set up the criteria in level 2 on which the decision should be based
on. The criteria should be formulated to describe the problem as thoroughly as possible, but not to an extent
where sensitivity is lost to changes in different elements [24]. This process is followed by a pairwise comparison
between the different criteria to determine their relative importance to each other. When each criterion
importance has been determine, the different solutions, between which the decision should be made, is pairwise
compared in each criteria. From these comparisons the most favorable solution is determine.
In this paper all pairwise comparisons were performed in a graphical interface where a bar is moved to define
each criterion or solutions importance in relation to another. In Figure 8 the graphical interface that has been
used is presented. The reason for showing the interface is to illustrate how the pairwise comparisons are
performed. No exact figures are used, instead the comparison reflect a person’s subjective opinion. No figures of
the ratio are shown to the user when moving a bar. This results in the interface being a completely visual tool,
since the only feedback the user gets is the visual impression of how far to one side the bar has been moved.
Figure 8. Graphical interface when making pairwise comparisons.
Each pairwise comparison results in a ratio. To illustrate how these ratios from each pairwise comparison are
processed an example given in [24] is used.
Assume that n stones A1,…,An has the weight w1,…,wn. By forming a matrix A of pairwise ratios, where each row
gives the ratio of the weight of each stone with respect to all others. The following matrix equation can be
formulated.
A1 A2  An
A1  W1 / W1 W1 / W2  W1 / Wn   W1   W1 
     
A2  W2 / W1 W2 / W2  W2 / Wn   W2   n  W2  (1)
           
     
An  Wn / W1 Wn / W2  Wn / Wn   Wn   Wn 
The formulated equation has given the solution to the hierarchy of the stones weight, the vector w. But a far-
reaching theoretical interpretation can be made. By multiplying the matrix A with a vector containing the stones
weight w the result wn is gained. If n is an eigenvalue of A, then w is an eigenvector of A. This means that the
matrix A has rank one, since every row is a constant multiple of the first and thus all its eigenvalues except one
are zero. The sum of a matrix eigenvalues is equal to the matrix trace, which in this case is equal to n. Therefore
n is equal to the non-zero eigenvalue.
The solution to Aw=nw is called the right eigenvector and is unique to within a multiplicative constant. To make
w unique it is normalized by the sum of all elements in w. It is clear that it is possible to from the comparison
matrix A to recreate the scale w.
When real life decisions are made it will not be possible to give the precise value to each ratio wn/wn, but only
estimates. Eigenvalue theory yields that a small deviation around a simple eigenvalue, as in n when A is
consistent, leads to a eigenvalue problem on the form Aw=λmaxw where λmax is the principle eigenvalue of A
where A may no longer be consistent but is still reciprocal [24].
If A is constructed based on the ratios obtained from a real life decision and the eigenvalue problem Aw=λmaxw is
solved the scale w can be obtained. But it is also possible to determine the level of inconsistences in the matrix.
By reconstructing the matrix based on w an approximation of A is gained that is consistent. Since the
approximation is consistent its trace equals its non-zero eigenvalue. This eigenvalue is used to form the
consistency ratio CR=λmax-n, which measures the deviation from the consistent approximation of A. The level of
consistency is used to determine whether or not the results from the process should be accepted.
When using AHP the first pairwise comparisons are between criteria. By forming the matrix A from this ratios
and then solving the eigenvalue problem, Aw=λmaxw, the vector w is obtained. The vector w represents each
criterions importance. By normalizing w, the sum of its entries becomes one. The vector contains each criterions
weight in relation to all others.
In the next step where solutions are pairwise compared in the different criteria, the same eigenvalue problem can
be solved. This results in a vector containing each solutions importance in that criterion. By multiplying the
vector with the respective criterions importance, the solutions local importance in that category is gained. This is
performed for every category. By summing the solutions local importance, its global importance is gained,
which can be compared with other solutions global importance to determine what solution is most favorable.
Concept development
Concept development can be regarded in many different ways. From simple sketches from a brainstorming
session to something that is almost ready for production. In this section the authors define what is meant by
concept development for this thesis and how far it goes. The concept development was a part of the whole work
process that is described above.
The concept development of a concept is seen as a seven-stage process according to [25].
1. Identify needs
2. Specify requirements
3. Generate product concepts
4. Choose concepts
5. Test concept
6. (Decide upon final specifications for concept)
7. (Plan for development)
In this thesis these stages were not considered as a recipe that is followed strictly but rather an outline that
guided in the work. Stages 6 and 7 are refrained from as the thesis is explorative and not meant to generate a
final design. Below follows a short description of how the first five stages are managed here.
Identify needs
First step in a concept development is to identify the needs. From discussions with TKMS and from the initial
pre-study it was identified that the subject and the need of the study would be to look at systems that handle the
launch and recovery of unmanned vehicles onboard ships. It was also decided that the study would result in
concepts that could be used as references and also to show a preliminary proof of feasibility.
Specify Requirements
Looking at requirements for the system has been done in two sequences. For the AHP evaluation process of
different technologies design drivers were identified. These were stated without numbers to be used subjectively
when weighing the technologies against each other. When continuing the work and refining a concept to analyze
the feasibility of it more specific requirements were derived. This includes what vehicles the system should
handle and what input and constraints a specific vehicle would mean. Also a generic type of ship was specified
to describe the surrounding constraints for the concept.
Generate concepts
Generating concepts was done through exploring existing technologies and studying what might be future
technologies in launch and recovery systems. This is the technology review/evaluation part of the thesis.
Choose concepts
The technologies that were gathered and evaluated were weighed against each other in the AHP-evaluation. The
results from this were used as a tool to make decisions on what concepts to continue the work with.
Testing concepts
The concepts that were chosen as the most suitable continued in development. These were dimensioned and
analyzed to see if they are feasible solutions. The concepts are not developed to a stage where they could be
tested physically for validation hence the analysis should be considered as the “testing” of the concepts.
Literature study and data collection
The thesis was initiated with a preliminary study where the authors built up a foundation of knowledge of the
subject and the adjacent areas of knowledge. At this stage of the study the information gathering was mainly
focused to be of a qualitative nature to get a better understanding of the field of the subject and also the design
mentality that exist within the production industry of naval ships.
As the subject of study is defined a deeper study is conducted to gain more comprehensive knowledge about the
systems and the vehicles they handle. Here a combination of qualitative and quantitative data are collected.
Primarily qualitative as the concept development as it is formulated above mainly requires collection of data that
is considered to be “soft data”. I.e. describing characteristics and functionality with words and comparison.
Secondarily a quantitative data collection is done to be able to do proper estimations when developing the
concepts further and also to be able to present a report that gives a better holistic view of the subject.
The thesis handles a topic where information is hard to come about. Partially because it is subjected to being
military secrecy classified but also because the information that exist is often in-house knowledge. Noticeable
sources of information are government agencies and companies within the private sector that are affiliated with
the Swedish defense. These are considered as reliable sources of information. Swedish defense research agency
(FOI) and Swedish defense material administration (FMV) are both examples of this. Also official reports from
other nation’s defense agencies have been used. Information has also been collected from research institutes that
are also considered to be highly reliable. Other sources are manufacturers of unmanned vehicles, launch and
recovery systems and surrounding equipment. There may be a question if values given by these are totally
accurate, but they are considered reliable enough for the scope of this thesis.
Concept generation – UAV launch and recovery
3
In this chapter different technologies or solutions for launching and recovering are presented together with
positive and negative attributes. These are evaluated in terms of design drivers to be able to decide on which type
of technology is chosen for further development and analysis.
The technologies are divided into technology reviews of launch devices and recovery devices. The technologies
presented here are initially chosen with the thought that they are or could in the near future be possible LARS for
UAV:s. This means that some solutions do not exist entirely as proven designs, but only on a conceptual level.
Another factor when choosing the technologies to examine is that they are considered as plausible solutions for
the category of UAV that it needs to handle. As stated in the Introduction the thesis is limited to fixed-wing
UAV:s in the range of TUAV according to the NATO-categorization class II.
Also stated in the Introduction is the type of ship that the system should be used on. It is declared that the
concept should be developed to be used on a littoral combat vessel that is 90 m long. In other words a corvette or
frigate.
Launch and recovery of a rotary winged UAV

Although not included in the scope of the thesis a short description of the procedure of landing and taking off
with rotary winged UAV:s as they are the alternative to having systems onboard for launching and recovering
fixed-wing UAV:s.
Launching and recovering a vertical take-off and landing UAV (VUAV) imposes little or no impact on the heli-
deck of the ship. Apart from being secured between operations the VUAV needs only to be positioned on the
deck before launching and retrieved after landing. Hence only a few personnel is needed. Smaller types of
VUAV:s can be retrieved by hand and larger types might need some sort of equipment to retrieve and stow the
VUAV. Examples of this can be to have the VUAV land on a cradle that is rail mounted onto the deck. The rail
allows for easier transport of the UAV from the heli-deck into a hangar. A system that does this in a similar way
for helicopters is Indal Technologies’ Aircraft Ship Integrated Secure and Traverse (ASIST) –system [26]. It
also expands the ability to conduct helicopter landings in rougher sea states as it has a securing system involving
a retractable probe dropped from the helicopter that connects to a latch. The helicopter is then winched onto the
deck where the mechanism secures it from toppling and sliding. This solution can of course be used for VUAV,
but is fairly impractical due to the complexity of the system and extra equipment that needs to be carried by the
VUAV. A more simple approach to securing the VUAV on deck is to use a grid that is integrated in the
helicopter deck. Northrop Grumman’s “Fire Scout” has a Light Harpoon Landing Restraint System (LHLRS)
mounted underneath its fuselage. When landed on the deck, standing on the grid, the harpoon is ejected into the
grid and mechanically locks into it. To launch again the lock is disengaged remotely by the operator just before
take-off [27]. The same type of system is used by the Super Lynx helicopters in naval operations.
∙ 3. Concept generation – UAV launch and recovery ∙ 16

Technology review - Launching a fixed-wing UAV
A fixed-wing UAV based on a ship is much more restricted in the launch procedure than the ones used on land.
Apart from the smaller types of UAV:s such as the Skylark I that can be launched by hand, UAV:s need means
to accelerate up to a certain speed to be able to take-off. Conventionally a runway is used where the UAV
accelerates by its own propulsion, but on a corvette or frigate the space is too limited to allow for this. Instead
measures have to be taken to assist in the take-off. This can be done in several different ways and here follows a
brief explanation of the four most common categories of launching systems employed today. The categories are:
rocket assisted take-off, bungee cord, hydraulic launchers and pneumatic launchers.
Rocket Assisted Take-Off

Often abbreviated RATO this launching system has its major advantages in being a small system that requires
near to zero take-off length. The basics of the system is that the UAV is placed on a stand with a rising angle. On
the UAV a rocket is attached in a reloadable motor system. In take-off the rocket is ignited to then accelerate the
UAV enough for its own engine to take over. The rocket is then released so it does not add to the weight during
the rest of the UAV’s operation. When released the rocket is usually dropped into the water. Either it is not a
favorable launch system from an environmental point of view or time is used to collect the rocket after launch.
The size and the amount of components needed for the launch makes it easy to carry onboard a ship and does not
require any huge alterations on the ship. However there can be an issue when it comes to stowing the rockets as
it is a pyrotechnic component and need to be housed as such. This is also an issue in manning as personnel is
probably needed for fire protection standby. During launch there is an interval where the UAV is airborne but
does not have enough velocity and airflow to produce the necessary aerodynamic forces to control the UAV.
This limits the RATO design to be used in calmer weather. The flames and the thrust generated by the rocket can
also be an issue if the deck is sensitive against this, e.g. if it is made out of fiber composites. The flames and
smoke that is caused by the rocket being fired can also have a negative effect if stealth characteristics is to be
maintained [28].
Figure 9. RATO launched Pioneer UAV. Pic. from science.howstuffworks.com.

+ Small deck footprint. - Neg. effects on visual, heat and aural signatures.
+ Small initial cost. - Rockets requiring special care.
+ Can be prepared in advance. - Safety considerations.
+ Minor installation needed on ship. - Significant recurring costs.

Bungee Cord
A catapult launching device that makes use of the energy that is stored in highly elastic bungee cords to launch
UAV:s. Common solutions include a metal rail that is positioned at a launch angle, a bungee cord (or multiple
cords) and a winch to be able to stretch the cord. Prior to launch the bungee cord is stretched into necessary
tension and secured with safety pins. These are then released to launch the UAV. Not considering an optional
motor for the winch the launch procedure is quiet in comparison. Due to the simplicity in the design the system,
not requiring many mechanical or electrical components, it is relatively light weight. Though it is restricted by
the characteristics of the cord and usually requires a long rail to be able to accelerate the UAV to the necessary
velocity. Bungee cord launchers are often limited to lightweight UAV:s. The initial release is usually without
any dampening and thus the result can be a “jerky” motion in the start of the launch. This can cause problems if
the UAV or its payload cannot withstand this type of acceleration. The solution is usually limited to support
UAV:s of smaller sizes. In its simplest design the bungee cord is probably the most “stealthy” technique
mentioned. It has no electrical components, it is relatively silent and the rail needed could be made out of low
signature material.
Figure 10. The German UAV LUNA launched from a bungee catapult. Pic. from miltechmag.com.
+ Simple concept. - Limited to smaller UAV:s
+ Low signature. - High initial accelerations.
+ Low cost. - Difficult to predict end velocity.
Hydraulic Launchers
Hydraulic launching devices are typically also designed as a catapult rail launcher. The fundamentals of the
system is that a two compartment cylinder is used. The compartments are separated by a piston that has a
compressible gas on one side of it and hydraulic oil on the other side. To store the energy needed for the launch
the oil is pumped so that it pushes the piston and thus compressing the gas on the other side. The side with the
hydraulic oil is connected via some sort of quick opening valve to a hydraulic motor. The motor powers a winch
that is attached to a rail-mounted cradle where the UAV is to be placed. To launch the UAV the mentioned valve
is opened. This lets the gas in the cylinder expand and move the piston that pushes the oil back, powering the
hydraulic motor. The motor turns the winch that pulls the cradle and accelerating the UAV to launch speed. The
hydraulic launch system is much more complex then bungee cord or rocket and requires power to operate.
Advantages are that it can produce relatively high launching velocities and the initial start can be softened to
avoid too high accelerations. This leads to the solution being more forgiving on the design of the UAV as it does
not need to support high accelerations as i.e. with bungee cord solution.

Figure 11. Shadow 200 hydraulic launch. Pic. from osimint.com.
+ Control over the whole launch phase. - High up-front cost.
+ Repeatable launch with quick reset. - Large deck footprint.
+ Adaptable to different UAV:s
+ Low recurring costs.
Pneumatic Launchers
The pneumatic launching systems are much like the hydraulic ones. Energy storage is a compressed gas and the
rails together with the cradle are of the same principle. Usually the gas that is used is air that is pressurized in
accumulator tanks. The launch is done by letting the compressed air out through a valve and accelerate a cradle
that holds the UAV. The force that the launch is done with can be adjusted by adjusting the pressure and by that
being able to support UAV:s of different mass. Some of the downsides with this technique is that a compressor
to pressurize the air needs to be powered and it can take time to refill the accumulator tanks [28]. This can of
course be solved by using the ships compressed air system if one exists, but that would involve the ship design
allowing outlets near the heli-deck.
Figure 12. Pneumatic launcher. Pic. from rpdefense.over-blog.com.

+ Control over the whole launch phase. - Long time for pressurization.
+ Proven design. - Large deck footprint
+ Low recurring cost - High initial cost.

Parasail (launch)
Looking much like the recreational parasails that tourists enjoy the parasail solution can be used for both launch
and recovery of UAV:s. Launching the UAV with a parasail includes letting the parasail take wind and carry the
UAV up to altitude. Once up at the necessary altitude the UAV is released into a dive in which it accelerates up
to the needed airspeed to perform a pull-up manoeuvre that puts it in the right level for a controlled flight. The
solution makes use of technology and sub-components that are proven and available [29]. As mentioned the
technique is used in recreational sports from small-sized speed boats that shows that the deck footprint is minor.
For fundamental use the only equipment that is needed on deck is a winch together with a towline. The parasail
is easy to stow as it can be folded and put into a bag. One concern that speaks against the solution is that in use it
denies other operations such as helicopter landings. The parasail is wind dependent which means that the
heading of the ship can be limited and that it might need to be windy as the speed of the ship can be under the
necessary limit. For more information, see the section Parasail (recovery) below.
Figure 13. Parasail lift of the UAV Exdrone.

+ Small deck footprint when stowed. - Large deck footprint when in use.
+ Proven solution (recreational parasailing). - Wind dependent.
+ Low recurring cost
+ Can be used for both launch and recovery.

Technology review - Recovery of a fixed-wing UAV
The simplest and most cost-efficient way to recover a fixed-wing UAV is the option that parallels the full-sized,
manned aircrafts, e.g. landing the conventional way on a runway. On smaller ships that only have a heli-deck
and not a full runway this is impossible. In this case there is a requirement on near zero-length recovery.
Net recovery
Using a net for recovery is a pretty straight forward solution. The technique is to have a vertically mounted net
that the UAV flies into for a quick landing. A simple and reliable technique that does not involve complex
components. A problem with net recovery is that it can damage the UAV as the deceleration is rather quick and
there is a risk of entanglement in the net that can damage propellers. However, many of the fixed-wing UAV:s
today have a pulse prop as propulsion that decreases the risk of damaging the propeller. Another disadvantage is
that a net system is logistically complex and labor-intensive, as it needs to be set up and taken down in between
operations [30]. As the footprint is fairly large it hinders helicopter operations if mounted on the heli-deck. For
larger types of UAV:s the net is not enough to absorb all the energy that a moving UAV has and thus a braking
system is required to complement the net [28]. During the Gulf War the U.S. Navy had UAV capabilities with
the Pioneer, which used the Ship Pioneer Arresting System (SPARS) vertical net system. These had good
statistics of capturing the UAV:s, but the vehicles often got damaged when being captured in the net [28].
Figure 14. Net recovery of a Pioneer UAV. Pic. from olive-drab.com.

+ Simple solution. - Requires a lot of manning.
+ “Zero-length” recovery. - Large deck footprint.
- Long setup time.
- Risk of damaging the UAV.
Arresting line
There are many different ways to employ an arresting line as a solution for recovery. One solution that is easy to
relate to is the one used on aircraft carriers when landing full-sized aircrafts. For an UAV the solution is of
course smaller. The system usually includes the arresting line and a braking system that are coupled together
with a stanchion and a boom. The boom is in some manner fitted against the ship. This can be on the deck or on
the side of the ship [31]. For this type of solution the UAV disadvantageously needs to be modified with a hook
in order to be able to latch on to the arresting line. Modifications can restrict the payload the UAV can carry as
well alter the aerodynamic characteristics of it.
+ Quick recovery. - Rough landing.

+ “Zero-length” recovery. - Commonly a fixed system on the ship.
- Suffers from ship motions.

Skyhook
A variant of having an arresting line the Skyhook recovery system uses a vertical wire instead. The system
includes an extendable crane and a catch wire. The recovery procedure is to have the UAV fly into the catching
wire with its leading wing tip. This causes the UAV to pivot down towards the frame of the crane, still in contact
with the wire. It slides down on the wire a bit until the wire gets wedged between the leading edge on the wing
and a hook on the wing tip [32]. The process is pretty “violent” as the deceleration is quick and the initial impact
of the wire hitting the UAV is concentrated on a small area. The footprint on the deck is small compared to the
net solution but the crane is larger when stowed. The solution exists today as a product from Insitu Inc. and is
also delivered in a complete system for the Scan Eagle UAV.
Figure 15. Skyhook recovery of ScanEagle UAV. Pic from thebahrainconspiracy.com.

+ Low manning necessary. - Large stowage.
+ “Zero-length” recovery. - Can suffer from ship motions.
+ Proven design. - Very rough recovery.
Windsock
The windsock recovery device has a low recovery time and a low degree of complexity. The design is conical
shaped windsock that at the large entry is much larger than the UAV:s wingspan. The smaller side of the
windsock is almost drawn shut. The recovery scheme is that the UAV flies into the large end of the windsock
and decelerated by the constricting nature of the windsock. Once it has come to a halt the UAV can be retrieved
easily by opening the narrow end of the windsock. Apart from the deck footprint that can hinder helicopter
landings the impact on the ships operations is minimal. The limitations on the UAV is that it cannot use a front
mounted propeller for propulsion as this involves risks of damaging both the UAV and the windsock [33].
Figure 16. Wind sock recovery solution visualized. Pic from navalengineers.org.
+ Quick recovery. - Limits types of UAV propulsion.
+ Low cost. - Large footprint.
+ Low weight. - Not an existing solution.

Parasail (recovery)
As mentioned in the Parasail (launch) section above, a parasail can be used for both launch and recovery of
UAV:s. It should be noted for clarification that this is not a parachute recovery that has an uncontrollable decent,
but instead uses a sail, or foil, that is coupled with the ship via a winch and tow line. The recovery method is for
the UAV to either latch on to a capture mechanism positioned close to the parasail or fly into a sort of a net that
is coupled with the tow line through secondary lines hanging from the parasail. The engines of the UAV can then
be turned off and the UAV is suspended in the air by the parasail. The parasail is then reeled down to the ship
with a winch, thus also bringing down the UAV. This solution allows for a relatively safe recovery as the
recovery point is up in the air, not close to any objects such as superstructure of the ship, that the UAV might run
into. It also dodges the issues with turbulent air wake effects caused by the ships superstructure. The recovery is,
compared to on-deck solutions, unaffected by the ships motions in the different degrees of freedom [29]. A big
disadvantage is that it hinders helicopter operations until all of the equipment is cleared from the deck.
+ Can be used for both launch and recovery. - Wind speed dependent
+ Smooth recovery. - Winch that might be permanent.
+ Recovery is independent of ship motions.
+ Safety risks moved away from the ship.

Evaluation of UAV-launching techniques
The listing of different techniques for launching fixed-wing UAV:s above is seen as the initial result of a concept
generation. These solutions are evaluated and compared against each other in terms of design considerations
criteria and how well the solutions perform. The higher the score the better the system performs in that criterion.
The results are then summarized and the criteria are weighed against each other in order to act as support in
deciding on a concept to work further with. The AHP evaluation method is described more in detail in the
Methodology chapter. The design criteria are as follows:
Impact on ship design.

How much the solution affects the design on the ship. If the system requires support from the ship, e.g.
power, this is considered to be a design alteration.
Safety issues.
Risks of damaging the ship, the UAV or any of the crew during launch operations.
Impact on ship operations.

How much the launch scheme affects the ships normal operational effectiveness.
Weight.
The weight of the launch system.
Deck footprint.
Area that the system uses. Both stowed and deployed mode are considered.
Cost.
Estimations on cost that derive from if parts and/or subsystems are commercially available or not.
Setup time.
The time it takes to set up the system and prepare it for launch.
Manning.
Amount of crew that is required to operate the system.

Impact on ship design
The catapult launchers have the biggest effect on ship design due to their size and weight. A large system, even
though mobile, needs the same volume as a TEU-container. The hydraulic and the pneumatic launchers require
either electrical power or being connected to the ships compressed air system. While the bungee cord solution
affects the ship design less than the pneumatic and hydraulic the rail needed for the UAV in question would be a
larger installation than the parasail. Using RATO might come off as the solution that affects the ship design the
least, but considering that the deck must withstand the burn from the rocket and that special care is needed as the
rockets are classified as flammable and explosive. For this reason they need to be housed in special contained
areas [34]. The parasail system requires a winch. Looking at the installations used on recreational boats for
parasailing the winch needed is a small installation that can even be made portable. The sail that is used is easily
folded and stowed away with ease.
Low impact
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High impact Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 17. AHP results - Impact on ship design.
Safety issues
Launching with rockets is obviously associated with some risks. Flames from the rockets and the phase right
after take-off, before the UAV has enough speed for aerodynamic stability makes the RATO system an
unfavorable solution. The catapult solutions differ slightly and this is because the solutions differ in amount of
control the operators have for the launch. In a bungee cord solution it is expected that there is no control during
the acceleration phase. With a hydraulic launcher this can be achieved with relatively easy measures such as just
controlling the amount of hydraulic pressure.
Low risk
Safety issues
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High risk Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 18. AHP results (launch) - Safety issues.

Impact on ship operations
As mentioned, the stability in the initial phase of rocket launch is a limiting factor for the RATO solution. For a
safe launch it requires low wind speeds and by that also low speed on the ship. In most cases the rocket, once
depleted, is dropped into the water and needs to be recovered that is an extra process that limits operations.
Using a parasail launch system can limit the heading of the ship depending on the speed and direction of the
wind.
Low impact
0,25
0,2
0,15
0,1
0,05
0
High impact Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 19. AHP results (launch) - Impact on ship operations.
Weight
This is another category that at first glance the RATO design should outperform parasail. The reason for this not
being the case is because to be able to launch multiple times the RATO solution requires new rockets that adds
to the extra weight. Launching with a rocket also requires a stand that is usually made out of some sort of metal.
The catapult launchers have their disadvantage in having heavy rails and both the pneumatic and the hydraulic
systems have machinery that adds to the weight. The pneumatic systems also have a tank for the compressed air
that also adds to the weight.
Light
Weight
0,6
0,5
0,4
0,3
0,2
0,1
0
Heavy
Bungee cord Hydraulic Pneumatic Parasail RATO
Figure 20. AHP results (launch) - Weight.

Deck footprint
Both the pneumatic and the hydraulic launcher systems are ungainly in relation to the other systems in this
evaluation. A catapult with a bungee cord does not involve as large equipment as the other catapult solutions.
The RATO system utilizes a stand during take-off that could be mobile and does not require a large area, but is
estimated to require a lot of area around it to be cleared out due to the rocket burst. The parasail deck footprint is
estimated to be fairly small as it only requires a winch on the deck. However, during the launch the parasail takes
up much space on the heli-deck as it is prepared before it is lifted by the wind.
Small footprint
Deck footprint
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Large footprint Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 21. AHP results (launch) - Deck footprint.
Cost
Even though the rocket system is probably the cheapest system installation wise (near zero installation) it is
deemed to be the solution that is the least cost efficient in the overall long run. This is due to the rockets used are
not reusable and thus a new rocket is used for every launch. The hydraulic and pneumatic launchers have
machinery that implies a higher cost. It can be argued that this is an increase in the UAV:s operational cost and
not isolated to the launch system. In this evaluation it is seen as the launching sequence cost. A parasail does not
involve much high-cost technology and the components that require the most maintenance are estimated to be
cheap to change as they are commercially available. The bungee cord solution is the most simple solution in
terms of technology and thus also probably the cheapest.
Low cost
Cost
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High cost Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 22. AHP results (launch) - Cost.

Setup time
A pneumatic catapult is probably the slowest of the evaluated systems to set up and launch. The launcher needs
to be placed and locked into position and the air pressure needs to be built up by a compressor before being able
to launch. The hydraulic system is almost the same, but it is estimated that a hydraulic motor can build up the
necessary stored energy faster than a pneumatic system. The bungee system requires the rail to be placed into
position and then preparing the launch by stretching the cord. Before being able to launch with a RATO launcher
the launch stand needs to be positioned and the actual rocket needs to be attached to the UAV. It is estimated that
the fastest setup time can be achieved with parasail as the winch that is needed is already fitted on the deck. The
preparation needed is to attach the UAV to the lifting device.
Quick setup
Setup time
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Slow setup Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 23. AHP results (launch) - Setup time.
Manning
The bungee cord launcher is quite simple in its handling and considered to not require a lot of manning. To place
it into position and prepare it for launch should require no more than two persons. A parasail system could
require more manning even though it is a lighter system it is slightly more complex than the bungee cord system.
The size of the sail is probably too large for one person to handle when unfolding and one person is needed to
operate the winch. The hydraulic and the pneumatic launcher are easy enough for one person to use once placed
into position, but requires more manning to actually position it. The actual launching with a rocket does not
require more than one person, but since there is a fire risk when using rockets it is estimated that a fire crew
could be necessary to be on stand-by during the launch.
Low manning
Manning
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High manning Bungee Hydraulic Pneumatic Parasail RATO
cord
Figure 24. AHP results (launch) - Manning.

Evaluation of UAV-recovery techniques
The same evaluation process that is used on the launch techniques is applied on the recovery techniques. The
criteria that are considered are also the same, but are instead evaluated for recovery solutions.

A net recovery system requires a large heli-deck deck to be functional. The ship needs to have fittings for
vertically placed poles that stretches the net or have the poles permanently mounted. The Skyhook solution
involves having a crane that, if fixed on the ship, is a quite large impact on the design. It can however be put on
the ship as a more flexible solution by letting the crane be placed on a mobile platform, such as a trailer. The
system is smaller than the Skyhook and net solutions, but is limited in the same way as it involves either
permanently fixed poles or a trailer solution. Same goes for the windsock solution. The small amount of
installation required for the winch plus the bonus that it is both a launch and a recovery system makes the
parasail system favorable when it comes to the impact on ship design.
Low impact
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High impact Net Arresting Skyhook Windsock Parasail
line
Figure 25. AHP results (recovery) - Impact on ship design.

Safety issues
A fair amount of personnel is needed on deck during recovery operations with the net capture system which is
considered as a risk. There is also a high risk of entangling the UAV in the net and causing damage to it. The
arresting line and the Skyhook system are considered equally safe/dangerous as the principle of using a line
results in somewhat equally small recovery target area, i.e. the target area in which the UAV needs to fly into to
be recovered. Depending on the size, the windsock can have a larger target area than both of the line solutions.
These three suffer the most from ship motions as the target area moves with the ship. If the target is missed the
UAV needs to avoid the ship and try again. Missing the recovery target area is not only time consuming, but is
also accompanied by the risk of having the UAV crash into and damaging the ship structure. Letting the UAV be
recovered by a parasail this risk is moved away from the ship. The target area is estimated to be equal to the
arresting line and Skyhook, but it does not move with the ships motion and by that is easier for the UAV to dock
with.
Low risk
Safety issues
0,5
0,4
0,3
0,2
0,1
0
High risk Net Arresting Skyhook Windsock Parasail
line
Figure 26. AHP results (recovery) - Safety issues.

The net recovery scheme affects ship operations in many ways. It takes a lot of preparing and manning to have it
ready. As the net can be displaced by wind over the deck produced by the forward motion of the ship both the
course and speed could be limited. The arresting line and Skyhook systems is, as mentioned, heavily affected by
the ships motions and thus limits the ships heading and speed against the waves. The windsock suffers the same
from this, but with a larger possible target area it handles the motions better. The parasail requires wind to lift off
but the speed of the ship is estimated to be enough to generate this.
Low impact
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High impact Net Arresting Skyhook Windsock Parasail
line
Figure 27. AHP results (recovery) - Impact on ship operations.

Deck footprint
On predating solutions the net recovery systems uses almost the whole aft deck on the ship, hindering other
structures and operations. The Skyhook is a relatively large system as the crane has a deck footprint equal to
11×2.5 m when deployed and 3.75 ×1.1 m stowed [32]. The arresting line system requires somewhat less space
than the vertical Skyhook system since it does not involve as much structural parts. The windsock can be made
as a fairly compact solution. The sock itself is made out of textile or similar material that can be folded. The
stand that holds the sock is the only thing that takes up area on the deck. As a flexible alternative the stand can
be on a carriage that can easily be stowed away. The parasail is again favorable as the solution combines both
launch and recovery and thus taking up less space. The only component that has a big effect on the deck is the
winch.
Small footprint
Deck footprint
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Large footprint Net Arresting Skyhook Windsock Parasail
line
Figure 28. AHP results (recovery) - Deck footprint.
Cost
The initial cost for the net recovery system is estimated to be low, but since there is a high risk of the net and
UAV to be damaged the long-term costs is high. The simplicity of the Skyhook system speaks for a low initial
cost and a low cost for maintenance on the actual recovery system, but since it is limited to UAV:s that can
handle the violent recovery there is a risk of getting R&D –costs for the UAV instead. An arresting line system is
almost as simple as the Skyhook system, but advanced arresting gears can add to the cost. For the parasail
solution it is estimated that all parts are commercially available hence a low initial and maintenance cost. An
early stage trial on a parasail system conducted by the U.S Coast Guard used commercially available equipment
that had a total cost of 25 000 $ [35]. A windsock is also a very simple system with just a few components that
are considered to be available commercially.
Low cost
Cost
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
High cost Net Arresting Skyhook Windsock Parasail
line
Figure 29. AHP results (recovery) - Cost.

Weight
The crane that is included in the Skyhook system is estimated to be the heaviest of all the evaluated systems and
with a trailer to make it more flexible it is even heavier. The land version of the Skyhook system weighs
approximately 1700 kg in total [36]. The net itself might not be that heavy, but together with the subsystems it
adds up to a weighty solution. The arresting line includes stanchions, boom and an arresting gear that weighs a
bit. The parasail weight is negligible, but the winch adds some weight to the ship. The weight of textile for the
windsock is the same as the parasail, negligible. The majority of the weight derives from the stand that is,
relative to the others systems, considered light.
Light
Weight
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Heavy Net Arresting Skyhook Windsock Parasail
line
Figure 30. AHP results (recovery) - Weight.
Manning
To raise and mount the equipment in the net recovery system or bring it down and untangle the UAV requires
relatively many out of the crew to do it in a reasonable time. The Skyhook system could be quite labor-efficient
if a fixed system is employed that could be automated, but still manpower is required to get the UAV from the
line and onto the ship. The trailer solution of the Skyhook system requires more personnel as it needs to be
pulled out on the heli-deck. The parasail needs one person to operate the winch and one or two persons to hold
the sail before it is “inflated” by the wind and lifts off. It is estimated that the windsock recovery scheme only
requires one crew member to setup and dismantle.
Low manning
Manning
0,3
0,25
0,2
0,15
0,1
0,05
0
High manning Net Arresting Skyhook Windsock Parasail
line
Figure 31. AHP results (recovery) - Manning.

Setup time
The net recovery system is complicated to have prepared in advance. The poles must be mounted and the net
stretched up and since it is a relatively large system it takes some time. Hydraulics can be used for the Skyhook
system to enable a quick extension of the crane, but if the crane is mounted on a trailer that is stowed in a hangar
this needs to be brought out onto the heli-deck and prepared. Setting up the parasail recovery scheme is
estimated to take some time as the foil needs to be brought out of stowage and prepared for launch. The arresting
line solution does not need much preparation if it is a permanent solution on e.g. the side of the hull. Basically
just put it in recovery position. The windsock solution would probably fit on a light mobile platform that easily
rolls out of a hangar for quick setup.
Quick setup
Setup time
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Slow setup Net Arresting Skyhook Windsock Parasail
line
Figure 32. AHP results (recovery) - Setup time.

Evaluation results - UAV launch and recovery techniques
From all of the evaluated criteria a total grading is derived. This is based through a weighing between the
different criteria and using this together with the results from the evaluation of the techniques. In Figure 33 and
Figure 34 the total results for the launch and recovery techniques are seen. How the different categories are
weighed in the AHP-evaluation can be seen in Appendix A.
Launch techniques
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Parasail Bungee Hydraulic Pneumatic RATO
cord
Figure 33. AHP Total results for launch techniques.
Recovery techniques
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Parasail Windsock Arresting Skyhook Net
line
Figure 34. AHP Total results for recovery techniques.
With these results it can be determined on which of the techniques is most appropriate to continue the
development process with. In both results it can be seen that the solution with a parasail has the highest grading
in both the launch and the recovery scheme. From this it is decided that this technique is chosen for continued
development.
The parasail concept that is generated is based on a systems that has been tested for lighter types of UAV:s. In
[35] a LARS is tested to launch and also recover by capturing a small UAV from and to a RHIB by using a
parasail with a circular shape. The fundamentals of that concept are applied for the concept generated here. The
transition includes a change in the vessel the UAV is launched from, the UAV that is launched and recovered,
the components of the LARS and stowage of the system.
The parasail technique is further analyzed and defined in the next chapter, Concept analysis – UAV launch and
recovery, to determine if it is a feasible solution.

Concept analysis – UAV launch and recovery
In this chapter it is examined if the launch and recovery technique that raises an UAV in the air with a parasail is
feasible or not. In order to do this a simplified concept of the solution is developed. During the concept
4
development some key design considerations are analyzed to determine the feasibility of using a solution like
this. In other words the aim is not to design an optimal solution, but rather to explore the capabilities, features
and, of course, the limitations of a parasail system. A secondary goal is to see if the system can be designed with
commercially available components to maintain a low cost and be competitive to other solutions. This is
explored through finding and giving examples on commercially available components. After a design has been
defined it is analyzed in terms of time scheme and operability.
Parasail lifter - concept description

In this section the concept is further defined in terms of components and how the launch and recovery scheme is
thought to be accomplished. The idea is based on the same procedure explained in the concept generation
section. In Error! Reference source not found. the idea of the system is showed.
The main components in the system are:
 Winch contraption together with a boom that stabilizes the UAV when on deck.
 Towline(s).
 Mechanism that carries/suspends/latches the UAV.
 Parasail together with pilot chute.
Figure 35. Idea sketch of the solution. Full, sky view. Figure 36. Idea sketch of parasail and capturing device during UAV
inflight.
∙ 4. Concept analysis – UAV launch and recovery ∙ 35

The launch scheme, i.e. from stowage to having a UAV in the air is to first position the winch contraption in
place. In Figure 35 this is thought of as being placed on the heli-deck. The towline is wound on the drum of the
winch in advance. The mechanism that carries the UAV is coupled with the towline and the parasail. The UAV
is attached to this with a grappling hook that can be released. With everything put together one or two crew
members unfolds and holds the parasail as it catches wind. Either the winch is controlled by a remote or a third
crewmember operating the winch starts to reel out the sail. Depending on the wind speed the reeling out needs to
be controlled with a breaking system on the winch. Once the parasail together with the UAV has reached a high
enough altitude the UAV releases the grappling hook and starts a free fall acceleration with engines on. When it
has reached the minimum stall speed it starts to level out into a controlled flight and can perform the intended
mission.
To recover the UAV again the parasail is raised in to the air in the same manner as the launch scheme. At the
altitude of recovery the UAV starts an inflight towards the capturing mechanism (same as the one used for the
launch and can be seen in Figure 36). The inflight can be guided by fitting the capturing mechanism with light-
or IR-beacons. This is advantageously done since UAV:s are normally fitted with sensors of that sort. As the
UAV makes contact and couples with the capturing mechanism it turns off the engines and is decelerated by the
drag and lifting force of the parasail. Due to the fact that the parasail is not a stiff structure the deceleration of the
UAV is dampened.
Input and Constraints

For this analysis the UAV that the launch and recovery system should handle is set to be a TUAV as this is the
type that is probably the most relevant for the use on the type of ship that is considered in this thesis. Smaller
types of UAV:s (SUAV) do normally not require launch and recovery devices and they do not have the same
endurance nor capabilities as their larger counterparts. Larger UAV:s (MALE, HALE) are considered to be too
heavy and too large to even be considered for launch and recover on a corvette/frigate and thus not handled in
this thesis. As a design example the Shadow 200B is used and the reason for this is, like stated in the chapter
Introduction, that it is familiar to the Swedish military since it is already used, but launched and recovered on
land. The UAV is included in the NATO-categorization class II (tactical) that this thesis is limited to.
With a “design UAV” chosen a weight of the UAV is acquired to act as an input value of what the parasail must
be able to lift. Also the stall speed is available which can be used when estimating a necessary launch speed.
Summarized the relevant inputs from UAV03 are:
 UAV Weight 170 kg
 Stall Speed 130 km/h (~36 m/s)

Forces on the parasail
The forces acting on the sail are aerodynamic forces, the force from the tow line that pulls the parasail and the
gravitational forces from the mass of the UAV that it carries (the mass of the sail itself is neglected). In Figure
37 below the forces acting on the parasail are shown.
Figure 37. The forces acting on the parasail defined in 2D (side view). FL and FD are the aerodynamic forces lift and drag. Vwind is the
velocity of the encountering wind. The gravitational force of the load (the UAV) is denoted FUAV. Ftowline is the pulling force of the
towline. α and β are angle of attack and bridle angle respectively.
The aerodynamic forces are calculated by approximating the parasail as a thin plate with a small angle of attack
against the wind. This allows for estimations of the lift and drag coefficients. The aerodynamic forces are split
into lift and drag that are defined in their fundamental state as
1 2
FL  CL  vwind Ax (2)
2
and
1 2
FD  C D  vwind Ay . (3)
2
CL and CD are lift and drag coefficients, ρ is the density of air, v is the encountering wind speed and Ay and Ax are
the projected areas on the noted axis. For basic approximations the aerodynamic coefficients can be kept
constant, but here they are determined as angle of attack and velocity dependent. The calculations for the
coefficients can be seen in Appendix B. It can be seen as overworking the problem as the accuracy of the
modelling is at an estimation level, but the calculations are done to attain all components that are velocity
dependent.
To be able to carry the load that is the UAV a parasail design is chosen with a 10 m span and 3 m long. In the
thin plate calculations the sail is estimated to have a constant angle of attack of 5°.

Necessary altitude for launching
In the launch scheme the UAV is lifted up into the air by the parasail. The altitude that the UAV needs to be
lifted to is determined by the minimum velocity that the UAV needs to maintain the aerodynamic forces, i.e. to
generate enough lift for it to stay in the air. The minimum velocity is often referred to as the stall speed and
having this as a requirement a necessary free fall altitude can be calculated. Using Newton’s second law together
with the definition of drag force the free fall can be expressed as the differential equation
dv CD Av2
 g  . (4)
dt 2m
Here g is the acceleration due to gravity, CD is the total drag coefficient, A the frontal area and m is the mass of
the UAV. No approximated or empirical based value of the drag coefficient was found for the UAV in question
so a value of 0.7 is set as a constant. This is based on a CD value for a typical model rocket [37] that is
considered to be the same shape as the UAV fuselage, excluding the wings that are estimated to have a
negligible contribution to the drag. Solving the equation numerically for the position and the velocity the free fall
phase can be described as in Figure 38 below.
Figure 38. Free fall phase during launch.
Figure 38 shows how that the UAV requires a free fall of 80 m for it to be able to accelerate up to the minimum
stall speed. In addition to this there is a loss in altitude when the UAV levels out from the free fall. To correctly
model the altitude loss during the leveling out phase is a task that is not possible without having much more
information about the UAV and its performance. One way to go around this is to estimate turn curvature derived
from maximum allowed accelerations on the fuselage of the UAV. However these would produce false results as
the leveling out path does not follow a radius, but rather an arc. Instead a value for the altitude in the leveling out
is set to 70 m. It should be noted that this is not a qualified estimation, but it is necessary to continue the scope of
this thesis.

Towing line
The force on the towing line is approximated by using the same coordinate system as the sail force calculations.
Assuming a static case where the sail is in a stable condition the sum of all forces acting on the bridle point,
where the towline is attached to the sail, is equal to zero. The vertical and horizontal components of the tension
acting on the line are then
Tv =FL -W (5)
and
Th =FD . (6)
The absolute tension acting in the direction of the line is then
T  Tv2  Th2 . (7)
This is the dimensioning force that the towline should withstand. The tension is examined for different
encountering wind speeds in Figure 39. As payload would decrease the vertical component in the tension the
case is considered most critical without any payload. The increase the payload would have on the drag is
neglected. The curve shows that for an encountering wind speed of 25 m/s the necessary strength is 6191 N.
Figure 39. Tension requirement for different wind speeds.
No suitable single line was found that alone handles this force and at the same time is practical. Instead the
concept is chosen to utilize two lines that are of a type that is found among similar applications. The Paracord
type IV (750) is a line that is used by civilian skydivers and by the U.S military. The data that is relevant for this
thesis is taken from the listing of the manufacturer [38] and seen in Table 9 below.
Table 9. Draft from Paracord IV (military grade) specification by 5COL.

Min. Break Strength: 750 lbs (349 kg)
Elongation: 30%
Outer diameter: 3/16” (4.7625 mm)
Length per mass 165 ft/lbs (110.9 m/kg)

To calculate the required length of the towing line the assumption that the line is kept straight along the path to
the ship. The point where the towing lines connect to the sail is called the bridle point. In the vicinity of this
point the inclination angle can be expressed as
tan   (Tv / Th ) . (8)
Assuming that this is the angle that is kept through the whole length of the lines (no slack) and the required
altitude calculated a minimum line length is determined to 176 m.
Winch
The force on the towing line calculated above is assumed to be the same force that the winch must be able to
counteract in order to retract the parasail to the ship again. The power output required by the winch is defined as
P  T  vwinch , (9)
assuming that both the force and the speed of the winch is kept constant over time. Through calculating the drag
and lift force on the parasail and by that the tension line force the required winch effect is approximated. In
Figure 40 the required winch effect is shown as a function of reel in speed. The required effect is shown for the
encountering wind speeds of 5 to 15 m/s with an increment of 1 m/s per curve. The critical load is estimated to
be with the sail not carrying any load, i.e. no UAV. The drag force that a carried UAV would contribute to the
horizontal force component is negligible in comparison to the decrease in vertical force that the mass of the
UAV would do.
Figure 40. Minimum required winch effect for different reel in speeds. Each curve correspond to an encountering wind speed from 5
to 15 m/s.
As an example an encountering wind speed of 15 m/s and a reel in speed of 1.6 m/s results in a required winch
effect of 4.35 kW. Same output effect would be able to reel in at 1.0 m/s if the encountering wind would instead
be 20 m/s.
If the winch performance results in loads on the towlines that is higher than the breaking strength it is probably
over dimensioned. In Figure 41 the force on the towline is seen and compared to the breaking strength, FT,max. A
note is the negative slope in the beginning for the lower encountering wind speeds. This is due to the drag being
the larger component, but moving to being the lesser with higher reel in speeds.

Figure 41. Towline force as a function of reel in speed. Each curve corresponds to an encountering wind velocity. The curves are
drawn with a decreasing increment of 0.5 m/s. The dashed, constant line is the maximum towline strength.
It is seen that the limit is well above for all the winch speeds and reel in speeds that are practical and realistic. In
this thesis the strength of the sail is not considered since a very general solution is examined. The towline is
usually much stronger than the sail hence if developing the concept fully the sail should be examined also.
The size of the whole winch subsystem is approximated through looking at winches that are used for recreational
parasailing. A winch system that has specifications estimated to be similar to the requirements of the designed
system here is found from the company Custom Chutes Inc. Its dimensions are 19H×32W×15D ´´ (0.5
×0.81×0.38 m) [39]. This is a single drum winch without a motor fitted. For the concept here a dual drum design
would be necessary since two tow lines are used. The example system that was found is however reasonable in
size as the drum capacity is much higher than the capacity that the concept system requires. A dual drum for the
concept is thus estimated to be of the same size as the example system.
The winch can be powered by hydraulic motors or electric motors. A hydraulic motor is usually smaller than a
corresponding electrical motor, but requires surrounding equipment such as oil tank, pump and hoses. It has
advantages in less complex design and high reliability. For a mobile system that is meant to move to and from
stowage it would be more feasible to use electric motors since they do not require any surrounding equipment.
With a minimum total effect of 4.35 kW it is roughly estimated that two engines of 2.2 kW would be able to
handle the task since it is determined that a dual line and dual drum design is applied. Malmbergs offers 2.2 kW
electrical engines that would increase the width of the winch system by about 0.3 m on each side. The weight of
one of these engines is 26 kg per piece [40]. Further dimensioning of the motors and winch system is refrained
from as it is outside the scope of this thesis.
Arresting mechanism/hook
The U.S. Coast Guard have done experiments with a system that is similar to the concept here. In the
experiments two subsystems are used to launch and recover respectively. To launch, the UAV is carried up by
the parasail with a hook-like device that has a remote electronic control carried by the sail. In recovery a net is
used that is hanging down from the tow line, parallel with the towing direction. Even though it is considered a
successful method of recovery in there are some issues documented. The propeller of the UAV got entangled in
the net and got cracked. It is also stated that during one of the trials one of the wings partially separated from the
fuselage due to the system being too stiff and thus causing to high accelerations in recovery. Here follows an
argumentative description on a solution that could handle these issues.
For this concept the subsystem that suspends the UAV in the air and is lifted by the parasail is thought of as an
arresting mechanism fitted close to the airborne end of the towlines, on a stanchion close to being underneath the
parasail. The UAV is coupled with the arresting mechanism with a hook that is locked before launch, during lift
up in the air and after recovery when the UAV is “captured” by the arresting mechanism. The hook is controlled
by the control system in the UAV and is opened up to release the UAV from the arresting system when initiating
the dive sequence of the launch.

The necessary span of the mechanism, and thus also the length of the arresting line, is dependent on the accuracy
of how well the UAV can target the beacons and hold its course towards the arresting line. In this thesis the
dimensioning of the span is not included.
Figure 42. Arresting mechanism.
In Figure 42 a visualization of the idea of the arresting mechanism is shown. It is thought of to be equipped with
IR-beacons that allow the UAV to recognize and navigate to the arresting mechanism. These are placed in a
triangular pattern to allow for a more accurate recognition (triangulation). The major advantage with using IR-
beacons is that they are not as constrained by weather and other sight-limiting factors as normal light beacons
are. As a complement to this the arresting system should be equipped with a GPS-transmitter so that the UAV
can recognize the recovery position before the IR-beacons are visible. The Shadow 200B UAV uses a system
called tactical automated landing system (TALS) that employs a direct link with microwaves to communicate
and guide the UAV in approach for the landing. TALS is used for landing on ground and is a system that is
considered too heavy to be equipped on the parasail [41]. However, a system like this could be relevant to have a
direct communication between the arresting mechanism and the UAV. This would minimize the delay in data
such as GPS-coordinates in comparison to relaying via the ships control interface.
As mentioned the arresting mechanism is mounted underneath the parasail on a boom. The boom can be seen as
the hub that couples the parasail’s smaller bridle lines, the tow lines and the arresting mechanism. In Figure 43.
Parasail with the arresting mechanism on the boom. below the idea of the principle is visualized.
Figure 43. Parasail with the arresting mechanism on the boom.
A note on Figure 43 is that the bridles lines and the two tow lines are attached to the ends of the boom. The idea
with this is to allow the parasail to be somewhat controlled in yaw motions by pulling one of the tow lines. The
principle resembles the controlling of a kite. This is not analyzed as it is not included in the scope, but mentioned
here as an idea.

Time scheme
The time scheme is obviously dependent on how well drilled the crew handling the system are. Here some
simple estimations are done for the less time consuming operations. The operation where the parasail is lifted to
the altitude of launch or recovery is assumed to be the phase that takes the longest time. This operation is not
necessarily crew dependent but instead dependent on how the system performs. To approximate the time it takes
to reel out the towline up to the required altitude the winch is considered to be totally declutched. The lifting
force that is needed in order to lift the UAV is set as lower limit. The heading that the ship has against the wind
is kept at 0°. The actual wind speed and the ship speed are summarized into only an encountering wind speed. In
Figure 44 below the blue lines each represent one encountering wind speed, starting at 10 m/s and going up to 20
m/s with an increment of 0.5 m/s.
The following is an example to determine the reel out speed. If the minimum mass that needs to be lifted is 170
kg (1667.7 N), an encountering wind speed of 15 m/s, an altitude of 150 m, a resulting inclination angle of 58.9°
(line length of 203.3 m) a maximum reel out speed is estimated to 1.9 m/s. This reel out speed is the maximum
while still generating enough lift. The time it takes to reel out is then 106 s.
Figure 44. Lift force as a function of reel out speed. Each curve corresponds to an encountering wind velocity. The curves are drawn
with a decreasing increment of 0.5 m/s. The dashed, constant line is the minimum lift required.
To be able to evaluate the reel in time the winch is used as the limiting factor. This is due to the fact that it is
already checked if the line holds for the effect that the winch performs and the sail strength is not included in this
analysis. As established in the Winch section above, a winch output of 4.35 kW and encountering winds of 15
m/s results in a maximum reel in speed of 1.6 m/s. With a line length of 203.3 m this means a reel in time of 127
s.
In Table 10 below the whole time scheme is presented with estimations for each operation. The operations that
are mainly man handled are given subjective estimations since they mainly depend on how well the crew is
trained. The calculated values are derived above and estimated for encountering winds of 15 m/s and a total
winch power output of 4.35kW.
Table 10. Estimated time scheme.

Task Time [min]
Retrieve system from stowage 2
Place system on helideck 1
Systems check 2
Raise the parasail to launch/recovery altitude 2
Launch (drop)/Recover (catch) >1
Reel in 2.2
Fold parasail 3
Stow system again 2
Total time: 15.2

Estimating operability
For systems on board of ships the operability is often explained in terms of sea states. I.e. different wave
conditions that implies different accelerations that affect the system. To be able to determine those accelerations
methods of analyzing wave data together with a ships response is usually applied. For this thesis the scope is to
look at the concepts in a more general sense hence there is neither a fully defined ship nor wave data available
for analysis. Instead the operability is analyzed in terms of wind and ship speed together with the ships heading
against the wind. Checking the operability with respect to wind speeds is a more suitable approach as the
solution is more affected by the wind than by the motions and accelerations on the ship.
The calculations used here are just a compilation of the calculations done in the preceding sections in this
chapter. In Figure 45 an example of the operability is examined for wind speeds of 12 m/s. The force on the
towline is compared against the maximum breaking strength and as with the analysis of the towline the mass of
the UAV is excluded to have the largest towline force during the launch and recovery procedure. Figure 46
shows a comparison between the generated lift and the minimum lift required. The diagram is also shown for
wind speeds of 12 m/s.
Figure 45. Operability polar-diagram. Towline Figure 46. Operability diagram. Generated lift
tension [N] for different ship speeds and headings [N] is plotted for different ship speeds and
against the wind [°] (blue lines). Maximum headings against the wind [°] (blue lines).
towline force (Tmax) is presented as the thick, red Minimum lift required (FL,min) is shown as the
line. The diagram is shown for wind speeds of 12 thick red line. The diagram is shown for wind
m/s. speeds of 12 m/s.
In Appendix more charts are presented that represent different wind speeds. These charts define minimum and
maximum speeds and headings against the wind for the ship. As a rough comparison the wind speeds are
checked in the Beufort Wind Force Scale. 12 m/s is equivalent to a 6 on the Beufort Scale that can be roughly
translated to Sea States. Even though the Beufort Scale and Sea State scale have different measurements they can
be converted. In Appendix a conversion table it is seen that a 6 on the Beufort Scale corresponds to Sea State 4
to 5.

Integration on a ship
The size of the winch is estimated to approximately 1.2W×0.5L×0.5D m. This approximation is done by using
the example sub-components suggested in the Winch section and assembling in a CAD-model as a complete
design. Using steel as material for the drum and the structural parts a full weight for the winch is determined to
be 200 kg.
The launch and recovery system itself is with its relatively small size considered to fit in a standardized TEU-
container. The complete system including a UAV and a control station could be fitted in the same TEU-
container. The UAV could be configured with foldable or detachable wings for easier stowage. To get the winch
out of the stowage to an operations position a telescopic rail system is suggested. This would also act as locking
mechanism to hold the winch in place during operations. In Figure 47 the idea is shown together with the
container in a hangar.
Figure 47. An example of how the system can be integrated in a containerized solution.
Having a containerized system is a flexible solution that quickly allows a change in the ships mission
configuration. If a UAV is not needed on board it is easy to remove the whole system if it is fitted in a container.
It is also advantageous when looking at it from a logistical perspective. If the ship is not stationed close to its
homeport and a change in the ships mission configuration to having a UAV on board is necessary it is easily
transported where the ship’s location by standardized logistical means.

Concept generation – Surface & Underwater launch
and recovery
5
In this section a concept on a LARS for surface and underwater vehicles is develop. First the needs on the system
are identified followed of a presentation and evaluation of two launch and recovery (L&R) –techniques. One of
the techniques is chosen and further analyzed.
Identify needs
In this section the objective is to identify what is needed of a LARS for surface and underwater vehicles in year
2015-2020. In order to identify the needs, similar projects has been studied and also what criterion there is on
such system today.
In the past each off-board vehicle had its own LARS. On a modern naval ship this would mean a great number of
LARS’s that need space, add weight, need to be maintained etc. Because of this problem there is an effort
towards a LARS that can handle several types and sizes of vehicles [1]. Another aspect to be considered is that
naval ships last longer than off-board vehicles [42]. This implies that a LARS will have to handle vehicles that
do not exist today.
One example of a project that intends to develop a combined LARS for surface and underwater vehicles is
LAURA managed by MARIN in Netherland. LAURA (launch and recovery of any small navy craft) aims at
designing a LARS that can handle a big variety of vehicles (50 kg – 12 tons) in moderate sea state (4-5 possibly
up to 6). [43]
LAURA is a joint industry project with navies and shipyards as members. Because LURA:s members consists of
possible users and developers of a LARS for surface and underwater vehicles the projects vehicle size objectives
is adopted in this paper. The vehicle sizes that the system should be capable of handling are presented in Table
11.
Table 11. Vehicle size objectives, LARS for surface and underwater vehicles. [42]
Min Max
Length of manned craft and USV 4m 12m
Beam of manned craft and USV 1m 3.5m
Height of manned craft and USV 1m 3m
Weight of manned craft and USV 200kg 12,000kg
Length of UUV 1m 10m
Beam of UUV 0.3m 2.5m
Height of UUV 0.3m 3m
Weight of UUV 35kg 9,000kg
To limit the mother ships movement where the system is located a set of basic motion-limiting criteria are
adopted. The criteria are the same for side-davits as for stern ramps. The basic motion-limiting criteria were
developed during full-scale trials in the Bering Sea in 1995. Since the motion criteria can be applied to both side
davits and stern ramps they are adopted as general limiting criteria for all LARS for surface and underwater
vehicles and presented below. [44]
Table 12. Basic motion-limiting criteria. [44]

Criterion Limits
Roll < 8.0o SSA
Pitch < 2.5o SSA
Vertical acceleration < 0.2g SSA at boat station
Lateral acceleration < 0.2g SSA at boat station
∙ 5. Concept analysis – Surface & Underwater launch and recovery ∙ 46

A boat that is launched or recovered with a LARS will be subjected to various accelerations. To limit these
accelerations the same criterion as adopted by the IMO Code of Safety for High-Speed Craft should apply to
boats launched or recovered via a stern ramp [21]. One of the delimitations is that persons are the limiting factor
in terms of accelerations. Therefor these rules apply to all vehicles launched and recovered, regardless of L&R -
technique. These acceleration criteria are shown in Table 13.
Table 13. Criteria for ship deployed vehicles, maximum acceleration not to be exceeded. [44]
Safety level Vertical Transverse Longitudinal
1. Minor risk standing persons 0.60g 0.25g 0.25g
2. Minor risk sitting persons 0.80g 0.35g 0.35g
3. Risk of injury sitting persons 1.00g 0.50g 0.50g
To set an objective on the conditions in which the LARS should be able to operate the operational objectives
used by LAURA is adopted. The reason for adopting LAURA:s operational objectives is the same as the reason
for adopting its vehicle size objectives. These operational objectives are formulated as a threshold value and a
target value. The threshold is what has to be exceeded for the system to be effective. The target value is what is
desirable.
Table 14 operational requirements, same as for LAURA [42].

Threshold Target
Speed during L&R [kn] 4 12
Sea state (all headings) 4 6
Sea state (optimal heading) 5 6
Another trend that has been identified within warship design is to reduce ships signature. A reduction of a ships
signature can give the ship strategic advantages. This is something that should be kept in mind when designing a
LARS. Therefore if possible decisions should be made to contribute to a reduction of the systems signature.

Technology review – LARS for surface and & underwater vehicles
The purpose of this section is to present the two most common techniques to launch and recovery vehicles today.
Launching and recovering vehicles at sea can be a real challenge. In high sea states an operation, which is simple
in calm weather, can get a lot more difficult. The challenge with L&R –operations in high sea states is primarily
due to relative -movement and –speed between off-board vehicle and mother ship. Recovery operations are
usually more difficult than launch operations. The recovery operation can be divided into two steps. Where in the
first step the vehicles is connected to the recovering vessel and in the second brought onboard. How these steps
is performed depends on the vehicle type and what L&R -technique that is used.
Davits
On warships davits are most commonly mounted on starboard and portside to allow flexibility in the side on to
which to deploy. They are commonly used for launching and recovering of the fast rescue boat (FRB) which
usually also serves as ship boat. Davits exist in a big variety of designs. The most commonly used on naval ships
are slewing davits, A-type frame davits and overhead telescopic davits. [1] Below is a listing of the
characteristics of the different techniques.
 Slewing davit, have a deck-mounted pedestal in front of the boats stowage with the boom resting over
the boat. When launching or recovering the boom swings in the horizontal plane (see Figure 48).
 A-type frame davit with an integrated stowing cradle. The davit swings in the vertical plane when
launching or recovering (see Figure 49).
 Overhead telescopic davits are usually used when vehicles are stowed in hull. Deployment and
recovering can be performed through a hatch that protects the system when not in use (see Figure 50).
Figure 48. Slewing davit (Pic. from Figure 49. A-type frame davit (Pic. from Figure 50. Overhead telescopic davit (Pic.
palfingerneddeck.com). palfingerneddeck.com). from palfingerneddeck.com).
Davits usually makes a relatively small footprint on the ship design, they are compact and easily retrofitted. They
can even be delivered as “plug and play units”, which more or less just have to be bolted to the deck. One
negative aspect with slewing davit and A-type frame davit is that they are most likely to be mounted on weather
deck, which increases the ships signature. Overhead telescopic davits on the other hand are primarily used when
vehicles are stored inside the hull. Hatches can be used to cover the davit and stowing place when the system is
at rest. By covering the davit the signature is reduced compared to a davit mounted on weather deck. The hatch
also serves as protection against the weather. All types of davits presented are of the shelf products (OTS) and
complies with the rules stated by SOLAS for FRB.
When recovering with a davit the vehicle has to be connected to the system in some way. How this is performed
depends on if the vehicle is manned or not. If the vehicle is manned it approaches the vessel and one of the
crewmembers on the small boat catches a line from the davit and connects it to the boat. When the boat is
connected step two begins and the boat is raised out from the water. If the vehicle is unmanned, the vessel and
vehicle positions themselves besides each other and the vehicle releases a line. A crewmember on the vessel
captures the line by using a rope with a hook in one end. When the line is brought onboard it is connected to the
davit. The davit can then raise the vehicle out of the water.

The interface between vehicle and davit is most commonly a quick release hook connected to a lifting eyebolt
located on top of the vehicles center of gravity. This simple but effective interface makes it possible to handle a
big variety of vehicle sizes and types. When launching or recovering with a davit the hook is connected to the
vehicle together with a line connected to the bow (see Figure 51). The purpose with having a line connected to
the bow is to control the longitudinal movement of the vehicle. When a vehicle is raised or lowered, crew is
needed to operate both the bow line and the davit. This results in davits requiring a lot of personnel when in
operation.
Figure 51. Launch/Recovery with davit.
The two factors that primarily limit davit systems operability is pendulum effects and hoisting speed. A too slow
hoist speed can result in snatching or the vehicle overturning if caught by a wave, too fast and crew will
experience high accelerations. The pendulum effect can cause the vehicle to collide with the recovering vessel.
How much the vehicle will swing back and forth depends on the ships movement and how far up the davit is
located.
Stern ramp
Stern ramps are primarily used to launch and recover RHIB’s and other surface vehicles but can also be used for
underwater vehicles. A stern ramp is what the name suggests, a fix or movable ramp in the stern of a ship. The
ramp is used to slide vehicles up and down the water. When launching, the vehicle can either be pushed down
the ramp or slide due to its own weight. When recovering, the vehicle is propelled by its own machine up the
ramp and hooked into a winch that pulls the vehicle to its rest. Both the launch and recovery operation can be
performed while the mother ship is traveling forward. Two advantages that stern ramp has compared to davit is
the short launch and recovery time and that it generally requires less personnel. The Swedish coast guard vessel
KBV201 can launch its FRB in 10 s and recover it in 10-12 s [44]. One negative aspect with a stern ramp is that it
makes a huge impact on the ship design. The ramp creates a big hole in the stern that might lead to stability
problems. To avoid this a stern ramp installation should be thought of in an early ship design stage.
When recovering a vehicle with a stern ramp the procedure varies depending on vehicle type. Surface vehicles
can drive up the ramp but this is not the case for underwater vehicles. When an underwater vehicle is to be
recovered, the vehicle and vessel positions themselves beside each other. The underwater vehicle releases a line,
which is caught by the crew in the same way as when using a davit. The line is then connected to a winch that
hauls the vehicle up the ramp. This procedure is very similar to when a davit is used except that the vehicle never
has to be lifted.

The factor that primarily limits a stern ramp’s operability is the sill water depth. The sill water depth is the water
depth over the stern ramps aft end (see Figure 52).
Figure 52. Sill water depth.
When the ships stern moves relative the water, the sill depth varies and the sill can even be exposed. If a vehicle
is recovered when the sill depth is to shallow it can experience high impact accelerations. To avoid this from
happening the system or coxswain controlling the vehicle has to watch the ramps movement and embark the
ramp at the right moment. An aspect that limits the number of vehicles a stern ramp can handle is the ramps
supporting structure. During launch and recover vehicles need to be supported by the ramp as they are moved in
and out of the water. The supporting structure needs to be tailored to the specific vehicle it is handling. The need
for a tailored supporting structure limits the number of vehicles a stern ramp can handle.
Evaluation of surface and underwater – launch and recovery techniques

To determine the most suitable launch and recovery technique to base a concept on that fulfills the needs stated
an evaluation is performed. The evaluation method presented in the section Analytical Hierarchy Process (AHP)
is used. The two launch and recovery techniques presented, davit and stern ramp, are evaluated. The evaluation
is performed in seven categories, which are further explained below. The categories are of different significance
and are therefore weighted against each other according to the method (AHP). The result from the weighting is
presented in Appendix A. All seven categories are listed below in order of importance.
Interoperability
Interoperability is the ability of different systems to cooperate. For LARS it is about the ability
to handle a variety of vehicles. One scenario is that an off-board vehicle is launched from one
ship and recovered by another. A high level of interoperability increases the number of missions
that the LARS can be used in.
Operability
LARS:s operability is defined as the sea state in which it is possible to launch and recover
vehicles with acceptable risks of damaging ship, vehicle or personnel. Thereby a high
operability translates into the ability to operate in a high sea state.
Low need for personnel

The personnel needed to operate a LARS are defined as the number of persons required
operating the actual LARS. The personnel operating the ship and other systems are thereby not
included.
Short launch and recovery time

The launch time is defined as the time from when the launch operation starts to the point where
the vehicle is propelled by its own propulsion system. The recovery time is defined as the time
from when the recovery operation starts to the point when the vehicle is at rest. The recovery
operation starts when the coxswain or control system approaches the LARS with the intension to
embark. The recovery time does thereby not include the time needed for the vehicle to find the
right moment to embark.

Availability as OTS-product
OTS-products are products that can be purchased on the market and have no need for
development. Some reasons for using OTS-products are a desire to reduce the overall
development of a system and the long term maintenance cost. A whole system is maybe not
available as an OTS-product, but instead its main components could be.
Low impact on ship design

A LARS:s impact on the ship design affects many aspects, e.g. the possibility to retrofit the
system or make changes in its design. Systems that have substantial impact on the ship design
have to be included in the design process at an early state to avoid expensive changes of the
design in a late stage.
Low cost
Cost is always a consideration aspect in developing new systems for a ship.
Interoperability
The aspect that primarily limit a LARS:s interoperability is the interface between vehicle and system. The two
techniques have two different interfaces. Davits use a quick release hook that is connected to a lifting eyebolt on
the vehicle as explained. This is a much simpler interface than stern ramps that need to have a supporting
structure tailored to the vehicle that is handled in addition to a quick release hook that can secure the vehicle on
the ramp. Davits “simple” interface allows them to be used with a big variety of vehicles, which translates into a
high interoperability.
Interoperability
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 53. AHP results – Interoperability.
Operability
Operating a davit involves lifting and having a davit that allows lifting in high sea states ensures a high
operability. Lifting could cause problems due to pendulum effects and the risk of damaging personnel or vehicle
is high during the whole operation. For stern ramps the critical moment is during recovery when the vehicle
embarks the ramp. Vehicles can experience high acceleration at this point both due to the relative speed and
relative movement between vehicle and mother ship. But when the vehicle has reached the point where it is fully
supported by the ramp it will follow the mother ships movement and the risk of damaged is decreased. The time
during operation where the vehicle is at high risk of being damage is much shorter for stern ramps than for
davits. Stern ramps have also proven to have a higher operability than traditional cranes and davits on U.S Coast
Guard (USCG) patrol boats [45].

Operability
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 54. AHP results – Operability.

The need for personnel when launching and recovering manned vehicles varies depending on the L&R –
technique. Launch operations with both davits and stern ramps can be remotely performed by the coxswain
onboard the boat. The extra personnel needed to operate the system are thereby zero. At recovery the needed
number of personnel differs between the two techniques. Using a davit requires one extra person onboard the
vehicle in addition to the coxswain to connect a line to the bow and the quick release hook while the boat at the
same time has to be maneuvered. Recovering with a stern ramp requires no extra personnel besides the coxswain
because the boat can be driven up the ramp and automatically connected to a winch.
Launching a USV can for both techniques be remotely performed from the USV:s control station in the same
way as manned vehicles by the coxswain. At recovery the procedure differ as well as the need for personnel.
Using a davit requires one person that catches the line released by the USV and connects it to the davit and one
that operates the davit. It is possible for the same person that catches the line to also operate the davit but when
the USV is close to the ship it is at high risk of being damaged. By having one extra person that operates the
davit in addition to the person that catches the line the time needed for the operation can be reduced. Recovery
operations with a stern ramp can be performed in the same way as for RHIB:s and the need for personnel is
thereby zero.
Both launch and recovery with a davit of UUV:s can be performed in the same way as for USV:s and the need
for personnel is thereby the same. Launching a UUV with a stern ramp is performed in the same way as with a
USV or a RHIB but recovery differs. Recovering a UUV with a stern ramp is very similar to the recovery
operation with a davit. A line released by the UUV has to be connected to a winch that hauls the UUV up on the
ramp. This operation requires the same amount of personnel as when using a davit i.e. 2 persons.
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 55. AHP results – Low need for personnel.

Stern ramps have the shortest launch and recovery time since the vehicle can slide down the ramp due to its own
weight and drive on to it with its own propulsion. Lifting a vehicle in and out of the water is a much slower
process.
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 56. AHP results – Short launch and recovery time.
Availability as OTS-product
Davits are available OTS as explained earlier, but this is not the case for stern ramps. Some of stern ramps sub-
systems can be purchased OTS, e.g. capturing mechanism, but this is far from the whole system. This is because
a stern ramps design needs to be tailored to the specific ship it will be used on to a much higher extent than
davits. This of course affects the cost of the system but also the possibility to use the same design on several
ships. Therefore the solution using a davit is considered the most favorable in this category.
Availability as OTS‐product
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 57. AHP results – Availability as OTS-product.

A davits impact on the ship design is small compared to a stern ramp. Since davits can be purchased as OTS-
products they more or less just have to be mounted on the ship. Stern ramps, on the other hand, create a space in
the aft where waves can enter. This can cause stability problems and stern ramps should therefore be included in
the design in an early stage. Stern ramps also require more space than davits and are therefore more likely to
interfere with other systems onboard.
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 58. AHP results – Low impact on ship design.
Low cost
In this part of the evaluation real figures have not been used because they are hard to find and it is not specific
products that are compared, but instead the two L&R -techniques. Since davits can be purchased as OTS-
products they are of course cheaper than stern ramps that have to be adjusted to the specific vessel it will be used
on. Stern ramps big impact on a ships design also results in more work in the design stage that adds to the high
cost.
Low cost
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Davit Stern ramp
Figure 59. AHP results – Low cost.

Evaluation results – Surface & underwater launch and recovery techniques
When the results from each category are compiled stern ramp gets 53% against davits 47%. The result shows
that the difference between the two system types is small when compared in these categories and that there is no
obvious choice of the most suitable LARS. The result also suppresses the importance of weighting the categories
against each other. One way to improve the evaluation could be to perform it with a specific ship in mind that the
system will be mounted on. Despite the uncertainty in the result stern ramp is chosen for further development
because more potential improvements are seen in this solution.
Results
0,54
0,52
0,5
0,48
0,46
0,44
0,42
Davit Stern ramp
Figure 60. AHP results – Results from evaluation of concept.

Generate Concept
From the evaluation stern ramp was chosen as launch and recovery technique for the concept. In this section the
concept that is developed is presented. In order to meet the requirement that the concept should be based on
techniques that will be proven in year 2015-2020 others work within the area is studied. Inspired by the
techniques and ideas found a concept is generated.
A stern ramp deployment system can be divided into ten major elements that affect the systems performance
[44]. These elements are:
 Ramp slope
 Overhead clearance
 Ramp sill depth
 Ramp surface
 Capture mechanism
 Shape of ramp opening
 Stern door or gate configuration
 Water management system
 Stern flap/wedge
Most of these elements can to some extent be examined without any information of the specific ship the system
will mounted on. These elements have been used as design aspects when generating a concept. The elements that
are considered to be too ship dependent or out of reach for this paper are water management system and stern
flap/wedge. The water management system is considered to be a too complex problem and stern flap/wedge too
ship dependent.
The two aspects that are considered to primarily limit stern ramps operability are the relative movement and
relative speed between off-board vehicle and LARS. The relative speed is especially a problem for UUV:s,
which is the slowest vehicle type. UUV:s can in average travel with a speed of 4 kn, meaning that the mother
ship will have to travel with an even lower speed at recovery. In rough weather the mother ship will most likely
have problems maintaining course at such low speed.
One way to address the problem with relative speed between vessel and vehicle is to employ a capturing
mechanism that uses a sled. The sled is released from the stern ramp and the wires that connect the sled to the
stern ramp are kept slack so that the relative speed is reduced. When the off-board vehicle is connected to the
sled they can be hauled onboard as one unit. The use of a sled that is deployed from the ramp does not only
address the problem with relative speed but also relative movement. UUV:s travel underwater where the
disturbances from surface waves are less. Allowing for the sled to dive when recovering UUV:s the connection
between sled and UUV can be performed underwater. Allowing the sled to dive is of course only useful at
recovery of underwater vehicles. Recovering surface vehicles such as USV and RHIB a sled that is towed on the
surface is used. Having a sled that is towed on the surface will also reduce the relative movement between
vehicle and ship, since the vehicles properties are more similar to the sled then to a 90 m long ship they will
behave more similar in the waves.
The recovery procedure described above where the sled is deployed from the ramp results in a longer recovery
time then for a fixed stern ramp where the off-board vehicle can drive directly up on the ramp. This is not
possible for UUV:s due to their low speed. But faster vehicles such as USV:s and RHIB:s has the required speed
for such recovery procedure. It is therefore desirable to allow for this type of recovery. By allowing the sleds aft
end to be lowered into the water while still mounted to the ramp the required sill depth can be achieved and thus
allow the ramp to be used as a fixed stern ramp. This type of system would still have a longer recovery time then
a fixed ramp, since the sled need to be lowered and hauled back up. But the recovery time would be reduced
compared to deploying the whole sled. By combining these two recovery procedures all three types of vehicles
can effectively be recovered. When the relative movement between off-board vehicle and mother ship is too big
to recover vehicles with the sled mounted to the ramp. The recovery procedure where the sled is deployed can be
used and thereby increase the systems operability.

Regardless of whether the sled is deployed or mounted to the ramp the vehicle need to connect to it at recovery.
The connection need to be automated for unmanned vehicles and preferably for manned too. On manned
vehicles an automated connection means that the crew can be seated during the whole recovery operation. Such
system exists on the market and has proven to be successful on Dutch navy patrol vessels [46]. The system
employs three poles that hold a metal triangle in position (see Figure 61). The two poles in the back hold the
triangle with springs. The vehicle is equipped with a hook in the bow. When the vehicles hook catches the
triangle the two springs absorbs the kinematic energy from the vehicle. When the vehicle has stopped, the third
pole ensures that the vehicle does not slide of the sled. By mounting the capturing mechanism on the sled the
same sled can be used for both types of recovery procedures.
Figure 61. Three catchpole capturing mechanism.
A probe and a receiver build up the capturing mechanism that connects an UUV to the sled. The probe is
mounted in the UUV:s front end and the receiver on the sled. At recovery the UUV enters the sled and is guided
to the receiver by a funnel (see Figure 62). When the UUV is in the sled and connected, the sled can be hauled
back onboard. The probe is retractable and can be hidden away to not interfere with the UUV:s hydrodynamic
performance.
Figure 62. Capturing mechanism for UUV.
When vehicles are being launched there is no need to deploy the sled, but the procedure differs depending on the
vehicles engine. If the engine can be started dry there is no need to lower the sled. Otherwise the sled need to be
lowered so that the vehicles stern gets in contact with the water to provide cooling water to the engine. When the
engine is started the vehicle is released from the catch mechanism and slides down the ramp into the water.
When the sled is deployed to recover it is attached to the stern ramp with two wires (see Figure 63). By
connecting the sled to the stern ramp with two wires that are spread apart on the stern ramp. The sled will be
forced to line up with the stern ramp when hauled in.
Figure 63. Sleds connection with stern ramp.

When a vehicle is hauled up on the stern ramp it will be subjected to various forces. Large UUV:s are typically
not designed to withstand lateral or shear forces [47]. Therefore the sled needs to be designed to absorb these
forces to protect the vehicle. This imposes that the supportive structure on the sled is tailored to the specific
vehicle that is handled. This will result in one sled for each vehicle that needs space. The sled should therefore
not only serve as capturing mechanism but also as stowing place. This can be performed by the use of skids on
the sled. Skids would allow the sled to be hauled up and down the ramp but also to be placed on a flat surface.
Such a solution will save a lot of space because vehicle and sled can be stored as one.
Another benefit with the possibility to tow a sled behind the vessel is that replenishment at sea (RAS) can be
included in the concept. One of the big advantages with unmanned vehicles compared to manned is the much
longer endurance. Therefore it is very likely that during longer missions the only reason to recover an unmanned
vehicle is to refuel. Regardless of how the vehicle is recovered, the operation involves a lot of risks. By
integrating the possibility to RAS into the system the need to fully recover unmanned vehicles can be reduced. A
special sled can be used to provide the needed interface towards the vehicle. One example of such interface is
shown below in Figure 64.
Figure 64. System to refuel at sea.
The RAS –system uses a sled equipped with a receiver that is towed behind the vessel. The USV is equipped
with a probe in the bow. When the system is used the USV drives into the sled so that the probe connects to the
receiver, allowing fuel to be transferred from the vessel to the USV. This system has been demonstrated at sea in
tests by [48].
When the stern ramp is not in use the opening in the transom can be closed to protect the system and limit the
amount of water that enters the ramp area. The stern door configuration consists of two doors that when opened
up creates a funnel that help guide the vehicle or sled onto the ramp.
Figure 65. Stern door configuration.

Sled design
The success of a launch and recovery operation with the designed system is to a big extent dependent on the
sleds design. Each vehicle needs its own sled that is tailored to the specific vehicle but the main different in the
sled design is if it will be used for surface vehicles or UUV:s. To further define the concept designs on one sled
to be used with UUV:s and one for surface vehicles has been developed.
The sled for UUV:s needs to automatically dive and maintain a constant depth so that the UUV can enter. The
capturing mechanism is as explained built up by a retractable probe in the front end of the UUV that connects to
a receiver on the sled. The receiver is located in the sleds front end (see Figure 66) As the UUV enters the sled it
is guided to the receiver by the structure that also support the UUV as it is hauled upon the ramp. The sleds two
side hulls are fitted with one wing each that can maintain a constant depth on the sled. Power supply and control
unit to the two wings is fitted within the two side hulls. By placing the control unit and power supply in the two
side hulls the sled only has to be connected to the stern ramp with the two towing cables.
Figure 66. Sled to be used at launch and recovery of UUV:s.
A catamaran hull builds up the sled for surface vehicles. The two side hulls are joined together by a structure that
also supports the vehicle at launch and recovery (see Figure 67). As the vehicle is driven up on the sled it is
guided towards the capturing mechanism by the two side hulls. Side hulls and support help maintain the vehicle
in position as the sled is hauled back onboard. All components on the sled is fully mechanical, which means that
the sled only have to be connected to the stern ramp with the two towing cables.
Figure 67. Sled to be used at launch and recovery of surface vehicles.

Concept analysis – Surface & Underwater launch
and recovery
To further define and evaluate the developed concept an analysis is performed. The analysis is divided in to five
6
different sections. The goal has been to set values or define the different elements in order to evaluate the
concepts performance and size. The analysis is performed in a general form to not tailor the system to a specific
ship but instead to a ship length (90 m). The analysis has been performed by investigating others work within the
area and by performing estimations.
Ramp slope
The ramp slope affects primarily two aspects, the possibility to self-launch and the relative speed at which a
vehicle can embark. The launch time is drastically reduced if it is possible for a vehicle to self-launch by sliding
down the ramp due to its own weight. In an investigation performed on several ships worldwide it is determined
that a minimum ramp slope of 12o is required for a vehicle to overcome the static friction and slide down the
ramp [44]. Whether the static friction will be overcome or not is of course dependent on which material the
vehicle and the ramp are made of. But a minimum ramp slope of 12o is considered to be the minimum needed
ramp slope for vehicles to effectively self-launch.
The ramp slope also affects the induced impact accelerations on the embarking vehicle during recovery. When a
vehicle is operating aft of the mother ship it will be affected by the propeller wash and waves. In order to
maintain directional stability as it approaches the stern ramp it will have to travel with a relatively high speed.
The approaching speed is usually twice the mother ships [44]. It is of course possible for the approaching vehicle
to reduce the speed just before getting in contact with the ramp but the speed need to be sufficient for the vehicle
to slide up the ramp to the point where it is connected to the capturing mechanism. The induced impact
accelerations that the embarking vehicle will experience are primarily dependent on the relative speed and the
ramp slope. An investigation has been performed to evaluate the induced impact acceleration as a function of
relative speed and ramp slope. The acceleration that has been calculated is the mean value over the time it takes
for the embarking vehicles keel to reach the same inclination as the ramp (see Figure 68).
Figure 68. The acceleration is calculated as the mean value over the time it takes the vehicles keel to reach the same inclination as the
ramp.
In the estimation it has been assumed that the boat enters the ramp horizontally and maintain its absolute velocity
until the keel has the same inclination as the ramp. The acceleration has been calculated in the boats center of
gravity in a horizontal- and vertical -direction. To the induced impact acceleration the basic motion criteria in
Table 12 has been added. The reason for adding the basic motion criteria is to create a worst-case scenario where
the mother ships stern moves as much as is allowed. The vertical and horizontal acceleration has then been
plotted for different speeds as a function of ramp slope. In the plots the safety levels presented in Table 13 are
shown to determine at which speed and ramp slope they are exceeded.

A boat that is launched or recovered with a LARS will be subjected to various accelerations. To limit these
accelerations the same criterion as adopted by the IMO Code of Safety for High-Speed Craft should apply to
boats launched or recovered via a stern ramp [21]. One of the delimitations is that persons are the limiting factor
in terms of accelerations. Therefor these rules apply to all vehicles launched and recovered, regardless of L&R -
technique. These acceleration criteria are shown in Table 13.
Figure 69. Horizontal induced impact acceleration for different speeds as a function of ramp slope.
Figure 70. Vertical induced impact acceleration for different speeds as a function of ramp slope.
Since it has been concluded that a minimum ramp slope of 12o is required for a vehicle to self-launch, ramp
slopes from 12o and above are of primary interest. With a relative speed of five knots and a ramp slope of 12o
none of the accelerations exceeds any safety levels. When the relative speed increases to 7.5 kn and the ramp
slope is kept at 12o the vertical acceleration exceeds safety level 2 (minor risk of injury to sitting persons). Since
manned vehicles will be recovered with an automated capturing mechanism the whole crew can be seated. This
means that safety level 1 can be exceeded but preferably not safety level 2. With a relative speed of 7.4 kn the
vertical acceleration reaches just up to safety level 2 and the horizontally acceleration is well below.
To allow for vehicles to self-launch and at the same time enter the ramp with a high relative speed the obvious
choice of ramp slope is 120. This will allow for vehicles to enter with a relative speed of 7.4 kn if the crew is
seated onboard the embarking vehicle.

Transom opening
The need for overhead clearance in the transom opening is dependent on the embarking vehicles relative
entrance speed. If the overhead clearance is too small it could significantly reduce the effectiveness of the stern
ramp for manned vehicles. When a manned vehicle approaches the ramp it is difficult for the crew to determine
the distance between the vehicle and the overhead obstacle. The crew can have the perception that the overhead
obstacle is closer than it actually is. This phenomenon can create apprehension for the crew when entering the
ramp, which leads to a reduced effectiveness. The needed overhead clearance can be estimated with a method
where the overhead clearance is divided into a static and dynamic allowance. [44]
The static overhead clearance is dependent on the entrance speed, i.e. higher entrance speed requires larger
vertical clearance. The method proposes that for zero relative speed the clearance should be 0.2 m and linearly
increasing to 0.4 m at 10 kn. The needed static overhead clearance is based on a civil engineering approach used
for land constructions.
The dynamic overhead clearance is added due to the relative movement between the small boat and the vessel. In
the method it is assumed that the small boats movement relative the vessel correlates to the wave elevation in the
stern. The dynamic overhead clearance is defined as the maximum of the relative movement (single amplitude).
In [44] the dynamic vertical clearance is given for varying ship lengths and significant wave height. For a 91.5m
medium endurance cutter the dynamic vertical clearance varies as follow.
Table 15. Need for dynamic vertical clearance on a 91.5 m medium endurance cutter. [44]
Significant wave height Dynamic vertical clearance
3m 2.2m
4m 2.56m
5m 2.63m
6m 3.0m
The total required height on the opening in the transom, from the calm water line is formulated as
H ramp  H boat  Tboat  Cstatic  Cdynamic (10)
Where Hboat is the total height of the embarking boat, Tboat is the draft of the embarking boat, Cstatic is the static
vertical clearance and Cdynamic is the dynamic vertical clearance. The required overhead clearance has been
determine for a manned vehicle with a height of 3 m, which is the maximum vehicle height the LARS is meant
to handle. The vehicle is assumed to enter the ramp with a relative speed of 7.4 kn, which has been determine to
be the maximum allowed relative speed with a ramp slope of 12o. The draft of the embarking vehicle has
conservatively been assumed to be 0.5 m. The overhead clearance has been calculated as a function of significant
wave height and plotted.

Figure 71. Overhead clearance on a 91.5m medium endurance cutter for a vehicle height of 3 m and a relative speed of 7.4 kn.
Significant wave height 3-4 m is within the span of sea state five and 4-6 m is within the span of sea state six
[49]. It can be seen that in order to operate in sea state five the required overhead clearance is 5.0 m and in sea
state six 5.8 m. Since the goal is to operate the stern ramp in up to sea state six the needed overhead clearance is
5.8 m.

Sill water depth
The sill water depth affects the stern ramps availability in higher sea states. The needed sill water depth is
dependent on the relative water levels movement in the stern of the mother ship. To better understand the relative
water level in the stern, data from a model test on an 80 m patrol vessel is analyzed. The model had a simulated
depth of 3 m and bilge keels as the only roll damping action. In Figure 72 and Figure 73 the probability of
exceedance as a function of relative water level in different sea states and headings are presented. Heading 180o
is head sea and 0o is following sea. The data are measured with a probe in the stern of the model. Rayleigh
distributions have then been formulated based on the collected data before handed to the author.
Figure 72. Relative water level in stern on an 80 m patrol vessel in sea state five.
Figure 73. Relative water level in stern on an 80 m patrol vessel in sea state seven.
With a sill water depth of 0.5 m the lowest probability of the sill being completely exposed on the model ship in
sea state five is 65 %. If the sill water depth is increased to 1 m the probability of complete exposure is reduced
to 40 %. This is a significant improvement that would increase the stern ramps effectiveness in sea state five.
The same comparison is performed on the model data from sea state seven. With a sill water depth of 0.5 m the
probability of complete exposure is 76 % and the probability is 67 % of a 1 m sill water depth. The much smaller
improvement in sea state seven shows that other means might be necessary to operate effectively in sea state
seven. The goal for the project is to develop a system that can operate in sea states six. Unfortunately data is not
available from sea state six and the test in sea state five and seven where not performed at the same speed. It is

therefore hard to say anything about the recovery operation where the sled remains on the ramp from this data.
But the data shows the amplitude of the water levels variation in the stern. This can be used to evaluate the
recovery operation when the sled is deployed from the ramp. The sill depth needed when the sled remains on the
ramp is highly ship dependent and is therefore hard to evaluate in general terms. But must be greater than 0.5-
0.75 m depending on arrangement in order to be effective [44]. The Canadian coast guard ship Gordon Reids
with an overall length of 50 m has a stern ramp that can be used in up to sea state six [44]. This ship length is
much shorter then the 90 m ship that is the objective for this report. Despite the difference its stern ramps sill
depth of 0.86 m can be seen as an indication of what is needed from a 90m ship. Because a 50m ship is more
similar to the intended vehicles then a 90 m ship they will move more similar at sea. It can therefore be assumed
that a 90 m ship needs additional sill water depth compared to the 0.86 m on Gordon Reid. It is concluded that a
sill water depth of 1m will give the system if mounted on a 90 m ship an operability of possibly up to sea state
six. To create this sill water depth without allowing large volumes of water to enter the ramp area the sled can be
partly lowered so that its aft end creates the needed sill dept.
By allowing the sled to be lowered to provide additional sill depth to the actual stern ramps, the stern ramps sill
depth is dependent on the recovery of the sled. It is thereby possible to employ a sill that is located at or above
the calm water line. This would reduce the amount of water that enters the ramp. When recovering vehicles with
the sled deployed the vehicle can experience high levels of trim. If the water level in the stern decreases when
the sled is about to enter the ramp the vehicle can trim to a degree where its transom gets submersed. The angle
of trim that a vehicle can be exposed to without the transom submersing is of course dependent on the vehicles
properties. To determine at which level it is reasonable to believe that the transom will get submersed, a static
geometric problem has been formulated. A vehicle from the smallest surface going category has been chosen for
the investigation. The vehicle is the Piraya that is presented in Table 5.
The trim will occur when the sleds front end is lifted upon the ramp sill. The scenario is illustrated in Figure 74
and the investigation has been formulated as a static geometric problem. The height of the transom is set to
0.38 m, which is the smallest shaft length on a standardized outboard engine [50].
Figure 74. Forces used in trim estimation.
The bending moment has been calculated around the point P. The bouncy force has been formulated as
FB  water gV (11)
Where V is the submersed volume, ρwater is the density of seawater and g the gravity. The submerged volume has
been determined by multiplying the submerged triangular area with half of the boats width to take account of the
hulls V-shape.

With a relative water level of -2.35 m the water reaches just up to the top of the transom and forces the vehicle to
trim 30o. It has to be mansion that the sleds contribution to the bouncy has not been taken into account. If the
stern ramps sill where to be placed at the same height as the calm water level the sill water depth would not
make any contribution to the relative water level. To determine whether an allowed relative water level of -2.35
m is enough for the system to operate in sea state seven data has been extracted from the model test. The data
that has been extracted is the probability of the relative water level to reach below -2.35m
Table 16.Probability of exceeding -2.35 m in sea states five with Table 17. Probability of exceeding -2.35 m in sea states seven
a speed of 8 kn. with a speed of 10 kn.
Probability Heading Probability Heading
0% 180o 11.4% 180o
0% 135o 5.8% 135o
0.6% 90o 39% 900
0.5% 0o 6.25% 00
The conclusion is drawn that the sled can be released and hauled onboard in any heading in sea state five. In sea
state seven the probability of exceedance is highly dependent on heading. This causes a higher impact on the
ships operation but by adjusting the heading the risk of the transom being submerged can significantly be
reduced. This shows that by positioning the stern ramps sill at the same height as the calm water level the system
can be operated in up to sea state seven with respect to the relative water level. How a ship behaves is highly
ship dependent. The data used can therefore not be seen as general. Instead it should be viewed upon as a
prediction of what is reasonable to expect.
Launch and recovery time and procedure

To determine the systems launch and recovery time the operational procedure has to be defined. How a launch or
recovery operation is performed is dependent on what vehicle that is handled and in what sea state. The
operational procedure can be divided into two types. The first one is when vehicles are launched and recovered
without the sled leaving the ramp and the second one when the sled is deployed from the ramp.
The first type of operation when the sled remains on the ramp, the launch procedure depends on the vehicles
engine. If the engine can be started dry there is no need to lower the sled before disconnecting the vehicle.
Otherwise the sled has to be lowered to a point where the vehicles stern gets in contact with the water. When the
launch time is estimated the latter is conservatively used. When recovering, the sled is lowered into the water to
create a sufficient sill water depth for the embarking vehicle. In the time estimation it is assumed that the sill
water depth needs to be 1 m. When the sled is in position the vehicle can embark. With the vehicle secured on
the sled they can be hauled up us one unit.
In the second type of operational procedure the sled is lowered and released from the stern ramp and held by two
wires. The sled is then positioned behind the mother ship by slacken the wires. How far behind the sled need to
be is dependent on how disturbed the water is aft of the mother ship. In the time estimation 20 m has been used.
When the sled is in position the vehicle can approach and drive on to the sled. When the vehicle is connected to
the sled they can be hauled onboard.
To estimate the time needed for these operations is not easy since human factors are involved. Therefore ships
equipped with similar stern ramps have been studied. To estimate the launch and recovery time for the
operational procedure when the sled is kept on the ramp the German sea rescue service vessel, Vormann Steffens
is studied. Vormann Steffens is equipped with a stern ramp that employs an automatic capturing mechanism and
can recover a RHIB in 10-15 s [44]. Vormann Steffens does not use a sled, which take time to lower and haul up.
Therefore time need to be added. 5 s is conservatively been added, which results in a recovery time of 15-20 s.
The launch time has been estimated by looking at the USCG patrol boat, Hammerhead. The stern ramp on
Hammerhead has a ramp slope of 120, which is the same as for the developed concept. The launch time on
Hammerhead is 7sec [44]. Again time need to be added to lower and haul up the sled. 5 s are added resulting in a
launch time of 12 s.
For the recovery process when the sled is released from the stern ramp no data has been available. Therefore the
process has been divided into steps. The time needed for each step has then been estimated in an attempt to
predict the time needed for the operation. In the first step the mother ship position itself in front of the vehicle
(see Figure 75). In the second step the sled is deployed. In the third step the vehicle approaches the sled and
connects. In the last step the sled and vehicle is hauled onboard as one unit.

Figure 75. Recovery operation with sled capturing mechanism.
The recovery time is counted from step two where the sled is deployed. It is assumed that the mother ship can
deploy the sled 50 m in front of the vehicle. The slowest vehicle type is UUV:s. An average UUV can travel at a
speed of 4 kn. To control the depth of the sled intended for UUV:s the sled has to move forward with a speed of
1 kn, which means that it will take the UUV 33 s to reach the sled. When the UUV is connected to the sled it can
be hauled in (step three). If the mother ship travels with a speed of 6 kn it will have reached 84 m during the 33 s
it takes the UUV to reach the sled. A reasonable winch speed is 2 m/s, which means that it will take 42 s to
winch the sled onboard (step four). The total time for this procedure is 75 s.
Space needed
The space that is needed for the stern ramp system is to a great extent dependent on the largest vehicle that the
system will handle. The biggest vehicle has the measurement 12L×3.5B×3H m. The sled is assumed to have the
same length as the longest vehicle resulting in a ramp length of 12 m. The minimum width of the ramp is defined
by the maximum vehicle width. But to this width an addition is needed to give the coxswain confidence to enter
the ramp, but not too much that the vehicle will come to rest out of position [44]. Since a sled is used it will help
guide the vehicle during recovery, which the ramp side does on a stern ramp without a sled. In an investigation
of several stern ramps the side clearance varied between 10 cm to 45 cm [44]. A side clearance of 45 cm is
assumed to be sufficient in order not to underestimate the space needed. The side clearance is measured from the
vehicles side to the side of the stern ramp. The total ramp width is thereby 4.4 m. In front of the ramp space is
needed for the two winches connected to the sled. Each winch is assumed to need one cubic meter of space each,
which is added to the total length of the system.

Concepts and solutions (results)
7
In this section the resulting concepts are presented. One preliminary concept has been developed for fixed-wing
UAV:s and one concept has been developed for water based vehicles. These include UUV:s, USV:s and RHIB:s.
Parasail UAV lifter – concept results

The early stage concept that was developed was a parasail lifter that can act as a device for both launching and
recovering of fixed wing UAV:s. The concept is developed to be able to do an initial analysis of feasibility and
to generate approximated values on particulars. The main particulars and attributes are shown in Table 18 below.
Table 18. Particulars for parasail lifter concept.

Parasail size: 10×3 m (W×L)
Towlines: 2×Paracord IV lines ∅ 4.8 mm
Winch powering: 2×2.2 kW electrical motors
Winch size: 1.2×0.5×0.5 m (W×L×D)
Winch weight: ~200 kg
Min. encountering wind speed: 14.5 m/s
Max. encountering wind speed: 26 m/s
Suggested stowage: TEU-container
Launch time: 8 min
Cycle time: 16 min
Recovery guide: IR-beacons (triangulation)
The concept was designed to be utilized onboard a littoral combat ship. This implies short take-off and landing
(STOL) or point take-off and landing (PTOL). The concept conforms to this as it requires no runway onboard the
ship.
The concept was made for a reference UAV based of the Shadow 200B. The UAV has a starting weight of 170
kg and a launch speed of 36 m/s. The design is considered to be able to both launch and recover this reference
UAV.
Operability is determined in terms of wind speeds. For charts concerning this see Appendix C.
Major components in the system were specified though calculations and examples of commercially available
components that match the specifications were given.
Description of components – Parasail lifter

Here follows a short description of the components and how they are joint together. In Figure 76 below the
components and how they are placed is shown. The winch is fitted on a telescopic rail system that can be
extended out through one of the opened sides of the container. The rails are in the level of the floor of the
container and enables the winch to be rolled out, still fixed by the rails, on the heli-deck. The winch is split into
two drums that each handle their own tow line. I.e. two tow lines are used to tow the parasail. The lines are in
one end attached to the drums, going through a support/guiding stanchion. On the other end the lines are attached
to ends of a boom that is the hub of the parasail and arresting mechanism. The arresting mechanism is joint with
the boom in the middle of it so that it hangs down in between the points of where the tow lines are attached. The
arresting mechanism is the sub system that carries the UAV up to launch and also captures the UAV in recovery.
It consists of a triangle frame, an arresting line and IR-beacons. The arresting line is where the UAV hooks on to
in both launch and recovery procedures. The IR-beacons are placed in a triangular pattern to allow for the UAV
to recognize and accurately navigate to the arresting mechanism for recovery. Bridle lines are used between the
boom and the parasail and are also attached to the ends of the boom.
∙ 7. Concepts and solutions (results) ∙ 68

Figure 76. Components of the parasail concept.
Operational procedure – Parasail lifter

The procedure of using the solution is here described as if it were to be used on a ship that has a heli-deck and
capabilities to carry at least one TEU-container on or adjacent to the heli-deck. The whole concept is stowed in
this container. The winch with the tow lines wounded on the drums is rolled out on the rails. The tow lines are
attached to the hub boom. The boom can in beforehand be prepared with the bridle line attachment to the
parasail so that it is on standby. The boom is placed on the guiding stanchions on the winch and the parasail is
placed on the deck not yet taking wind. The hook on the UAV is locked on to the arresting mechanism. The
smaller pilot chute is released into the air and by the time it has taken wind the main parasail is released also. As
it catches wind it ascends to the necessary height. The ascent is controlled by the breaks and motors of the
winch.
In launch it carries the UAV up to necessary launch altitude. For the reference UAV in this report this is
approximated to 150 m. At that altitude the hook that is connected on to the arresting mechanism is unlocked and
thus releasing the UAV into a deep dive. With engines on the UAV accelerates in the dive to a velocity that is
necessary for the UAV to have aerodynamic control (generate lift). The minimum velocity of the reference UAV
is 36 m/s and when it has reached this it starts to level out and can commence its mission.
In recovery the UAV is assumed to navigate to the vicinity of the ship and the parasail lifter. When in range it
can locate and recognize with IR-sensors the parasail and the arresting mechanism’s IR-beacons. The UAV the
starts an inflight from behind the parasail at minimum speed. The hook on the top of the UAV is set into
recovery position (upright). The UAV targets a flight path directly underneath the arresting mechanism to allow
the hook to capture the arresting line. When the hook engages the arresting line the engines of the UAV is shut
off and a quick deceleration halts the UAV. The deceleration is softened by the dampening nature of the parasail
being lifted by air flow.

Discussion and conclusions – Parasail lifter
The above stated results show that the parasail lifter concept is a solution that works to launch and recover a
tactical fixed wing UAV. Some conclusions on the system can be drawn that motivates the feasibility.
- The approximated size of the parasail used is deemed to be in line with what is needed to
lift the reference UAV up to the necessary height and at the same time not be too
impractical to use on a ship. The solution is thought of as using a smaller pilot chute to
initially lift the main parasail of the heli-deck. This allows for a safer handling on deck as
handling the main parasail directly could be hazardous for the crew working with the
system.
- Forces on the towlines were approximated and a suggestions on lines that can withstand
those forces was given.
- A winch contraption was calculated and a preliminary sizing was set. This size was
determined to be feasible since it fits in a flexible solutions such as a TEU-container.
- The launch time is feasible. Comparing to other systems the concept performs in similar
times when looking at the time from stowage to having an UAV up in the air.
Apart from the conclusions that argue for the concepts feasibility some other conclusions are made:
- The design is in no ways optimized. The concept was only generated to explore the
solutions feasibility.
- Preliminary operability charts were generated. The minimum wind required was determined
to 14.5 m/s. In still winds this would mean that the ship would have to travel in 28 kn. A
high speed that can be considered impractical. A larger or more effective design on the
parasail could be the solution to this.
- The parasail is modelled as a thin plate with a small angle of attack against the wind. The
model is 2-D and excludes the rotation of the parasail. At this stage of the concept
development these types of estimations can be seen as relevant, but for continued
development one should implement a model that analyses the parasail in 3-D and includes
rotation of the parasail.
- During the course of the development it has been clearer that a system for launch and
recovery is seldom considered as an individual product, but rather a part of the larger
system that includes UAV:s, control station and the launch and recovery device. To
consider the launch and recovery device as it is done in this thesis, as an individual system,
is relatively impractical since one generic system is not possible. The development of the
system should be more coherent with the development of the UAV.

- The parasail system fits in a TEU-container, but not the reference UAV used in the thesis.
Even though the scope of the thesis was focused on the launch and recovery device one
should consider this during the design process.
- The phase where the UAV levels out after a free fall in the launch scheme was considered
too time consuming and required more input than available to model accurately. This was
refrained from and only an estimate was given for the altitude loss during this phase.
- No data was found on accuracy on the flight of UAV:s. This was considered to be the main
factor when determining the necessary span of the arresting mechanism.
- Snow, ice and rain are factors that have been left out in this report. These are, of course,
important design considerations and should be included in the evaluation of the different
technologies.
- The parasail could be used for purposes other than the launch and recovery of UAV:s that
are not mentioned in this report. E.g. relaying radio communications or acting as a
signal/radar decoy.
Concluding remarks – Parasail lifter

The development of this concept has not been focused on presenting an optimized design, but rather on
exploring the field of systems for launching and recovering fixed wing UAV:s and creating a concept to evaluate
if it is a feasible solution or not. The results show that it could be a feasible method of launching and recovering
a UAV that is of the same type as the reference UAV in this report. However, it should be noted that evaluations
have been done subjectively based on the authors gained knowledge during the course of the thesis. In analyzing
the concept the calculations are based on simplified models that does not describe the reality truly.
In continuing the LARS of UAV:s development it is recommended that the results of this report are not to be
used as design parameters, but rather a foundation in another iteration in an evaluation process of different
technologies. I.e. being a support in development decisions.

RHIB, UUV and USV –concepts
The concept that has been developed for surface and underwater vehicles is a stern ramp that employs a sled.
The sled is connected to the ramp with two wires, which are spread apart in the ramp end. By using two wires
instead of one the sled will be forced to line up with the stern ramp when hauled in. By employing a sled both
the problem with relative movement and relative speed is addressed. The system is based on a stern ramp, which
is a proven technique for launching and recovering off-board vehicles. The concept that has been developed can
launch and recover UUV:s, USV:s and RHIB:s. The L&R–technique employed in the concept is based on
proven designs and designs that is under development. Therefore the concept has potential of being a proven
design in year 2015-2020. The concepts main particulars and visualization is presented below.
Figure 77. Visualization of LARS for surface and underwater vehicles.
Table 19. Main particulars for stern ramp system.

Ramp slope 12o
Ramp length 12m
Ramp width 4.4m
Total length of system 12.7m
Total height system 3.5m
Overhead clearance in transom opening 5.8m
Ramp height over calm water line (aft end) 0m
Sill water depth with sled lowered 0-1m
Relative entrance speed 7.4knop
Launch time 12sec
Recovery time with sled onboard 15-20sec
Recovery time with sled deployed 75sec
Operational procedure – Stern ramp system

The systems launch procedure varies depending on vehicle engine. If the engine can be started dry i.e. without
access to seawater. The engine is started and the vehicle slides down the ramp due to its own weight. If the
engine cannot be started dry the sled is lowered so that the cooling water inlet is submersed. The engine is started
and the vehicle slides in to the water in the same way as when the engine is started dry.
The recovery process can be performed in three different ways depending on vehicle type and sea condition. If
the sea condition allows for the system to be used as a fixed ramp, the sled is lowered into the water to provide
the needed sill water depth. The vehicle drives up the ramp and connects to the sled. When the vehicle is
connected to the sled it is hauled back up on the ramp. This recovery procedure can be used for USV:s and
RHIB:s.
Figure 78. Recovery process with sled remaining on the ramp.

When the relative movement between embarking vehicle and vessel is too big for the system to be used as a
fixed stern ramp. The sled is released and towed behind the vessel. The vehicle can then approach the sled and
connect. The sled and vehicle is then hauled back up on the ramp as one unit. The reason for deploying the sled
is because the sled will move more similar to the vehicle then the ship due to size differences. This recovery
procedure can be used for USV:s and RHIB:s.
Figure 79. Recovery process with sled deployed from ramp.
When UUV:s is to be recovered the sled is released from the stern ramp. The two winches that connect the sled
to the stern ramp are kept slack to reduce the speed between UUV and sled. The sled is equipped with two wings
that enable the sled to dive and maintain a constant depth. The UUV approaches the sled and connects. When the
UUV is connected to the sled it can be hauled back up on the ramp.
Figure 80. Recovery process for UUV.
Discussion – Stern ramp system

When the concept has been developed focus has been put on solving the problem that occur due to relative -
movement and –speed between embarking vehicle and mother ship. These problems have been addressed with a
design that meets the needs stated, in this paper, on a LARS for surface and underwater vehicles. To verify the
developed concepts feasibility an analysis has been performed. The analysis is a preliminary evaluation of the
system, performed by comparisons with existing stern ramps and estimations. The evaluation of the concept
should be seen as a prediction of what the system are capable of performing.
Some conclusions can be drawn about the systems feasibility.
- The concept has potential of being a proven design in year 2015-2020, since most of the systems
subcomponents are already proven designs and the fundamental part of the system (a stern ramp) is
widely used.
- When developing the concept focus has been on the actual LARS. In order for the system to work
effectively another system is needed that can move vehicles between the ramp area and stowing place.
This system has not been included in the design and further investigation is therefore needed in this
area.
- A positive aspect with this technique, that has not been thoroughly investigated, is the possibility to
RAS. The possibility has been verified by looking at others work, but how such a system would impact
the ship design has not been investigated.
- The use of sleds instead of fixed mounted supporting structure on the ramp enables the ramp to be used
with a big variety of vehicles. The sled cannot only be used in launch and recovery operation but also
for stowing of the vehicle.

Conclusion about the analysis limitation:
- All estimations performed have been on static problems. Since the behavior of system and vehicle is
dynamic this approach implies limitations. Therefore further analysis is required on the dynamic
behavior to verify the systems feasibility. Difficulties could arise due to relative movement between
sled and vessel. This problem has been addressed to some extent by employing two wires that will help
guide the sled into position. But since the sleds behavior when towed, is a complex problem further
analysis has not been possible within the time frame of this project. It is therefore suggested that the
dynamic behavior of the towed sled is further analyzed.

Authors work and responsibility division
Both authors have contributed to the work equally, but with different subjects in the report. Below is a listing of
how the authors divided the writing and work in the report.
Marcus Eriksson: Patrick Ringman:

Abstract Ch. 1, Introduction:
Ch. 1. Introduction: Introduction
Unmanned Surface Vehicle – USV Background
Unmanned Undersea Vehicle – UUV Unmanned Vehicles
Other off-board vehicles – RHIB Unmanned Aerial Vehicle – UAV
Ch. 2, Methodology: Unmanned Ground Vehicle - UGV
Analytical Hierarchy Process (AHP) Launch and recovery system for aerial vehicles.
Ch. 5, Concept generation – Surface & Scope
Underwater launch and recovery Ch. 2, Methodology:
Ch. 6, Concept analysis – Surface & Underwater Introduction
launch and recovery Work process
Ch. 7, Concepts and solutions (results): Concept development
RHIB, UUV and USV –concepts & discussions Literature study and data collection
Ch. 3, Concept generation – UAV launch and
recovery
Ch. 4, Concept analysis – UAV launch and
recovery
Ch. 7, Concepts and solutions (results):
Parasail results and discussions
Report layout and structure

References
8
[1] A. Kimber, "Boat Launch and Recovery – A Key Enabling Technology For Flexible Warships," Sydney,
Australia, 2012.
[2] U.S. Department o Defense, "Dictionary of Military and Associated Terms," 2010.
[3] "Wikipedia," [Online]. Available: http://en.wikipedia.org/wiki/Unmanned_surface_vehicle visited on
2013-04-11. [Accessed 11 04 2013].
[4] "HowStuffWorks," [Online]. Available: http://www.howstuffworks.com/reaper1.htm. [Accessed 11 04
2013].
[5] R. Austin, Unmanned Aircraft Systems: UAVS Design, Development ad deployment, John Wiley & Sons
Ltd, 2010.
[6] P. Ögren, P. Bull, P. Grahn, G. Hillerström, P. Johansson, M. Jändel, A. Karlholm, R. Karlsson, L.
Lundgren, L. Läfgren, T. Mårtensson, P. Nilsson, F. Näsström, A. Rensfelt, J. Robinson, J. Schubert, M.
Sparf, P. Svenmarck, P. Thoren and M. Ulvklo, "Förstudie obemannade farkoster," FOI, FHS, 2012.
[7] "Wikipedia," [Online]. Available: http://en.wikipedia.org/wiki/CTOL. [Accessed 22 04 2013].
[8] "Wikipedia," [Online]. Available: http://en.wikipedia.org/wiki/PTOL. [Accessed 22 04 2013].
[9] L. A. Karlberg, Falken rekar och räddar liv, Ny Teknik, 2010.
[10] "fmv.se," [Online]. Available: http://www.fmv.se/sv/Projekt/Ornen---Taktisk-UAV-03. [Accessed 24 04
2013].
[11] "forsvarsmakten.se," [Online]. Available: http://www.forsvarsmakten.se/sv/Materiel-och-
teknik/Flyg/UAV-03-Ornen. [Accessed 23 04 2013].
[12] "Wikipedia," [Online]. Available: [13] http://en.wikipedia.org/wiki/Shadow_200. [Accessed 23 04 2013].
[13] S. Group, "Skeldar Naval Datasheet," [Online]. Available:
http://www.saabgroup.com/Global/Documents%20and%20Images/Air/Airborne%20Solutions/Unmanned
%20systems/marin/SAAB_Skeldar_Naval_DataSheet.pdf. [Accessed 23 04 2013].
[14] "liquidr.com," [Online]. Available: http://liquidr.com/. [Accessed 29 04 2013].
[15] ThyssenKrupp Marine System AB, [Online]. Available: http://www.kockums.se/en/products-
services/naval-surface-ships/coast-guard-security/piraya-usv/. [Accessed 11 09 2013].
[16] "fmv.se," [Online]. Available: http://www.fmv.se/sv/Nyheter-och-press/Nyheter-fran-FMV/Effektivare-
ubatsjakt-med-obemannad-farkost/. [Accessed 03 05 2013].
[17] "iver-auv.com," [Online]. Available: http://www.iver-auv.com/. [Accessed 02 05 2013].
[18] "ise.bc.ca," [Online]. Available: www.ise.bc.ca. [Accessed 02 05 2013].
[19] "Wikiepedia," 27 04 2013. [Online]. Available: http://en.wikipedia.org/wiki/Unmanned_ground_vehicle.
[Accessed 29 04 2013].
[20] "Popular Mechanics," 18 04 2011. [Online]. Available:
http://www.popularmechanics.com/technology/military/robots/how-battle-tested-robots-are-helping-out-at-
fukushima-5586925. [Accessed 29 04 2013].
[21] I. M. Holmen and B. K. Solvang, Om kvalitativa och kvantitativa metoder, Studentlitteratur, 1997.
[22] G. Wallén, Vetenskapsteori och forskningsmetodik, Studentlitteratur, 1996.
[23] R. Patel and B. Davidsson, Forskningsmetodikens grunder: att planera, genomföra och rapportera en
undersökning, Studentlitteratur, 2003.
[24] T. J. Saaty, "How to make a decision: The Analytic Hierarchy Process," European Journal of Operational
Research 48, p. 18, 1990.
[25] K. T. Ulrich and S. D. Eppinger, Product design and development, McGraw-Hill, 2004.
[26] A. R. Feldman and R. G. Langlois, "Autonomous straightening and traversing of shipboard helicopters,"
Journal of Field Robotics, vol. 23, pp. 123-139, 2006.
[27] "kell-strom.com," Kell-Strom, [Online]. Available: http://www.kell-strom.com/tools/dcn/dcnpg03.htm.
∙ 8. References ∙ 76
[Accessed 11 06 2013].
[28] P. Fahlstrom and T. J. Gleason, Introduction to UAV Systems - 4th ed., John Wiley & Sons Ltd., 2012.
[29] W. R. McDonnel, "Launch and recovery system for unmanned aerial vehicles". U.S.A. Patent US7097137
B2, 29 08 2006.
[30] P. McGillivary, "Design Considerations for launch & recovery of autonomous systems from ships,
including coast guard icebreakers.," in Launch & Recovery 2010, Arlington, 2010.
[31] M. D. Adamski, R. G. Root Jr. and A. M. Watts, "UAV recovery system". U.S.A. Patent US7219856 B2,
22 05 2007.
[32] "Insitu - company webpage," [Online]. Available: http://www.insitu.com/systems/launch-and-
recovery/recovery-systems. [Accessed 07 06 2013].
[33] M. Goodman and R. Mortimer, "UAV integration Aboard U.S. Navy Ships," 2010.
[34] P. v. Blyenburgh, "UAVs - Current Situation and considerations for the way forward," European Vehicle
Systems Association, Paris, 1999.
[35] M. Walz, "Parasail Launch and Recovery of Fixed Wing UAVs," Unmanned Systems, Jan/Feb 2002.
[36] "www.insitu.com," Insitu Inc., [Online]. Available: http://www.insitu.com/systems/launch-and-
recovery/recovery-systems. [Accessed 31 07 2013].
[37] "Wikipedia - Drag Coefficient," [Online]. Available: http://en.wikipedia.org/wiki/Drag_coefficient.
[Accessed 31 08 2013].
[38] "www.5col.net," 5COL Survival Supply , [Online]. Available: http://www.5col.net/750-Parachute-Cord-
MIL-C-5040H-PIA-C-5040/dp/B009B1PJ0A. [Accessed 06 08 2013].
[39] "customchutes.com," Custom Chutes Inc., [Online]. Available:
http://www.customchutes.com/index.php/winches/new-winches. [Accessed 01 09 2013].
[40] Malmbergs, 3-fas motorer ECOL, IE2, Malmbergs.
[41] L. Newcome, "New approaches to launch and recovery," Launch and recovery, 03 2003.
[42] S. Knight and M. Walker, "LAURA Joint Industry Project," in Wageningen 2012, 2012.
[43] "marin.nl," 08 04 2011. [Online]. Available: http://www.marin.nl/web/News/News-items/LAURA-JIP-
launches.htm. [Accessed 03 07 2013].
[44] R. Sheinberg, T. G. Beukema, P. V. Minnick, W. Kauczynski, A. L. Silver and C. Cleary, "Stern Boat
Deployment Systems and Operability," Society of Naval Architects and Marine Engineers, New York.
[45] M. Willis P.E., H. A. M. Schwender and B. Sherman, "An Operational Perspective of Launch & Recovery
for USCG Patrol Craft".
[46] "palfingerneddeck.com," [Online]. Available: http://www.palfingerneddeck.com/en/navy-and-coast-
guard/rhib/. [Accessed 09 08 2013].
[47] K. Sharp, D. Cronin, D. Small, R. Swanson and T. Augustus, "A Cocoon-Based Shipboard Launch and
Recovery System for Large Autonomous Underwater Vehicles," in MTS/IEEE Oceans 2001, Nov. 2001.
[48] B. Galway and B. Harris, "Unmanned Sea Surface Vehicle (USSV) Motion Data and Refueling," Naval
Surface Warfare Center Carderock Div Suffolk Va Detachment Norfolk, 2010.
[49] "Wikipedia.com," [Online]. Available: http://en.wikipedia.org/wiki/Sea_state. [Accessed 15 08 2013].
[50] "Wikipedia.org," [Online]. Available: http://en.wikipedia.org/wiki/Outboard_motor. [Accessed 22 08
2013].
∙ 8. References ∙ 77
Appendix A
Weighting of evaluation categories
Here follows two charts, each containing the weighting between the categories used in the evaluation process
between launch and recovery techniques. In Figure A- 1 the categories used in the evaluation of aerial vehicles
are presented and in Figure A- 2 the evaluation categories for surface and underwater vehicles. In each
respective chart the categories importance in relation to all other categories in that chart are presented as a
number, which is used in the evaluation method AHP.
Weighting ‐ UAV
0,3
0,25
0,2
0,15
0,1
0,05
0
Impact on Safety Impact on Weight Manning Setup Deck Cost
ship issues ship time footprint
design operations
Figure A- 1. Weighting of evaluation categories, used in evaluation of LAR – techniques for UAV.
Weighting ‐ UUV, USV and RHIB
0,4
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
Figure A- 2. Weighting of evaluation categories, used in evaluation of LAR – techniques for UUV, USV and RHIB.
A-1
A-2
Appendix B
Aerodynamic coefficients used in parasail calculations.
Equation Section (Next)
This appendix is a brief explanation of how the aerodynamic coefficients are derived.
Area of the sail (rectangular shape):

A  sc (1)
The span of the sail is denoted s and c is the length.
Aspect ratio of the sail is given by:

AR  s 2 / A (2)
Lift
The lift coefficient, CL is defined as:
CL 0
CL  (3)
CL0
(1  )
  AR
CL0 is the theoretical lift coefficient at zero speed and given by:
CL 0  2 (4)
In eq.(4), α is the angle of attack in the vertical direction.
Drag
The drag coefficient is defined as a function of the theoretical drag coefficient at zero speed (CD0) and the
induced drag coefficient (CDi) defined as
CD  CD 0  CDi (5)
The definition of induced drag coefficient used here is:
C Di  C 2L / ( AR) (6)
B-1
The density of air is here denoted ρ. η is a shape factor. For a rectangular shape this is chosen as 0.7. AR and CL
are defined in eq. (2) and (3) respectively.
The theoretical drag coefficient at zero speed is here defined as:
CF A  FF
CD 0  (7)
Aproj
CF is the skin friction contribution to the drag in a dimensionless form. Aproj is the projected area “behind” the
sail. I.e. when looking at the sail from the front. FF is a form factor. Here a definition for a planar surface is used
and given as1:
4
t t
FF  1  1.8    50   (8)
c c
As seen in eq. (7) the friction has a contribution to the drag. Here an empirically derived formulation of the skin
friction coefficient is used for planar surfaces1.
0.455
CF  (9)
log(Re) 2.58
1
R. Hendrickson, D. Roman, D. Rajkovic ”Drag: An Introduction” 1997
B-2
Appendix C
Estimated operability charts for Parasail lifter
The following charts (Figure C- 1 to Figure C- 4) show the estimated operability when considering the minimum
lift required. The lifting force is increasing with the radius. The heading of the ship against the wind is the
angular change in the chart. The thinner, blue curves represent the generated lift at different ship speeds. Starting
at 0 kn, going to 28 kn with an increment of 7 kn/curve. The thicker, red curve represents the minimum required
lift of approximately 1670 N.
Figure C- 1. Wind: 6 m/s. Figure C- 2. Wind: 8 m/s.
Figure C- 3. Wind: 10 m/s. Figure C- 4. Wind: 12 m/s.
C-1
Figure C- 5 to Figure C- 7 show the operability when considering the maximum force in the towlines. The
heading of the ship against the wind is the angular change in the chart. The thin, blue curves represent the
tension force in the towlines at different ship speeds. The thick red curve is the maximum allowed tension in the
lines of approximately 7000 N. The tension is increasing with the radius.
Figure C- 5. Wind: 12 m/s. Figure C- 6.Wind: 14 m/s.
Figure C- 7. Wind: 16 m/s.
C-2
Appendix D
Beufort Wind Force Scale V.s. Pierson-Moskowitz Sea States
Table D- 1. Wind force Vs. Sea states. From http://www.globalsecurity.org/military/systems/ship/weather.htm.

NUMBER DESCRIPTION DEFINITION WINDS(KN) AVG WAVE SEA STATE
HEIGHT (FT) EQUIVILANT
(NOTE 1)
1 LIGHT AIRS RIPPLES W/APPEARANCE OF 1-3 0.05 0
SCALES.
2 LIGHT BREEZE SMALL WAVELETS, GLASSY 4-6 0.18 0-1
APPEARANCE.
3 GENTLE BREEZE LARGE WAVELETS, CRESTS 7-10 0.6-0.88 1-2
BEGIN TO BREAK.
4 MODERATE SMALL WAVES BECOMING 11-16 1.40-2.90 2-3
BREEZE LARGE WAVES, WHITE CAPS
APPEAR.
5 FRESH BREEZE MANY WHITE CAPS, CHANCE 17-21 3.80-5.00 3-4
OF SEA SPRAY.
6 STRONG LARGE WAVES BEGIN TO 22-27 6.40-9.60 4-5
BREEZE FORM FOAM CRESTS
EXTENSIVE PROBABLY
SPRAY.
7 MODERATE SEA HEAPS UP, WHITE FOAM 28-33 11-16 5-6
GALE BLOWS IN STEAKS
SPINDRIFT IS SEEN.
8 FRESH GALE MODERATELY HIGH WAVES 34-40 19-28 6-7
OF GREATER
LENGTH. FOAM IS BLOWN
SPARY AFFECTS VISIBILITY.
9 STRONG GALE HIGH WAVES, DENSE FOAM 41-47 31-40 7-8
STREAKS, SEA BEGINS TO
ROLL.
D-1

Launch and Recovery Systems For Unmanned Vehicles Onboard Ships. A Study and Initial Concepts

Uploaded by

Copyright:

Available Formats

Launch and Recovery Systems For Unmanned Vehicles Onboard Ships. A Study and Initial Concepts

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Launch and Recovery Systems For Unmanned Vehicles Onboard Ships. A Study and Initial Concepts

Uploaded by

Copyright:

Available Formats

Launch and recovery systems for unmanned

vehicles onboard ships. A study and initial

Centre for Naval

Course: SD270X – Naval Architecture Master’s Thesis

Appendix A - Weighting of evaluation categories

Unmanned Aerial Vehicle – UAV

 Shadowing enemy fleets.

 Decoying missiles by the emission of artificial signatures.

 Relaying radio signals.

 Protection of ports from offshore attack.

 Optical surveillance and reconnaissance.

Table 1. NATO categorization of UAV:s. [6]

Table 2. Data for Skylark I. [11]

Figure 1. UAV02 Falken. Pic. from wikipedia.com.

Skeldar V-200 Maritime

Table 4. Data for Skeldar. [13]

Table 5. Data for Piraya. [15]

Figure 4. USV Piraya. Pic. from kockums.se

Unmanned Undersea Vehicle – UUV

Table 6. The four categorize of UUV:s.

Table 7. Data for Iver2. [17]

Figure 5. UUV Iver2. Pic. from auvac.org.

Table 8. Data for Theseus. [18]

Figure 6. UUV Theseus. Pic. from ise.bc.ca

Other off-board vehicles - RHIB

 Define the concepts to be able to reconnect with the design drivers.

- Attaining fundamental knowledge

- Determining the subject of study

- Planning the approach and methodology of the work

- Present and report.

Figure 7. The three decision levels.

Figure 8. Graphical interface when making pairwise comparisons.

The concept development of a concept is seen as a seven-stage process according to [25].

3. Generate product concepts

6. (Decide upon final specifications for concept)

7. (Plan for development)

Launch and recovery of a rotary winged UAV

∙ 3. Concept generation – UAV launch and recovery ∙ 16

Rocket Assisted Take-Off

Figure 9. RATO launched Pioneer UAV. Pic. from science.howstuffworks.com.

∙ 3. Concept generation – UAV launch and recovery ∙ 17

∙ 3. Concept generation – UAV launch and recovery ∙ 18

Figure 12. Pneumatic launcher. Pic. from rpdefense.over-blog.com.

∙ 3. Concept generation – UAV launch and recovery ∙ 19

Figure 13. Parasail lift of the UAV Exdrone.

∙ 3. Concept generation – UAV launch and recovery ∙ 20

Figure 14. Net recovery of a Pioneer UAV. Pic. from olive-drab.com.

+ Quick recovery. - Rough landing.

∙ 3. Concept generation – UAV launch and recovery ∙ 21

Figure 15. Skyhook recovery of ScanEagle UAV. Pic from thebahrainconspiracy.com.

∙ 3. Concept generation – UAV launch and recovery ∙ 22

∙ 3. Concept generation – UAV launch and recovery ∙ 23

Impact on ship design.

Impact on ship operations.

∙ 3. Concept generation – UAV launch and recovery ∙ 24

Figure 17. AHP results - Impact on ship design.

Figure 18. AHP results (launch) - Safety issues.

∙ 3. Concept generation – UAV launch and recovery ∙ 25

Figure 19. AHP results (launch) - Impact on ship operations.