See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/221074461

AILA - design of an autonomous mobile dual-arm robot

Conference Paper  in  Proceedings - IEEE International Conference on Robotics and Automation · May 2011
DOI: 10.1109/ICRA.2011.5979775 · Source: DBLP

CITATIONS READS

19 693

9 authors, including:

Johannes Lemburg Markus Eich

Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research Queensland University of Technology
16 PUBLICATIONS   96 CITATIONS    35 PUBLICATIONS   282 CITATIONS   

SEE PROFILE SEE PROFILE

Peter Kampmann
Deutsches Forschungszentrum für Künstliche Intelligenz
19 PUBLICATIONS   177 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

MINOAS View project

Users' Body Experience and Human-Machine Interfaces in (Assistive) Robotics View project

All content following this page was uploaded by Johannes Lemburg on 02 June 2014.

The user has requested enhancement of the downloaded file.

 AILA - Design of an autonomous mobile dual-arm robot
Johannes Lemburg, José de Gea Fernández, Markus Eich, Dennis Mronga,
Peter Kampmann, Andreas Vogt, Achint Aggarwal, Yuping Shi, and Frank Kirchner

Abstract— This paper presents the design of the robot AILA,

a mobile dual-arm robot system developed as a research
platform for investigating aspects of the currently booming mul-
tidisciplinary area of mobile manipulation. The robot integrates
and allows in a single platform to perform research in most of
the areas involved in autonomous robotics: navigation, mobile
and dual-arm manipulation planning, active compliance and
force control strategies, object recognition, scene representation,
and semantic perception. AILA has 32 degrees of freedom,
including 7-DOF arms, 4-DOF torso, 2-DOF head, and a mobile
base equipped with six wheels, each of them with two degrees of
freedom. The primary design goal was to achieve a lightweight
arm construction with a payload-to-weight ratio greater than
one. Besides, an adjustable body should sustain the dual-arm
system providing an extended workspace. In addition, mobility
is provided by means of a wheel-based mobile base. As a result,
AILA’s arms can lift 8kg and weigh 5.5kg, thus achieving
a payload-to-weight ratio of 1.45. The paper will provide an
overview of the design, especially in the mechatronics area, as
well as of its realization, the sensors incorporated in the system,
and its control software.

I. INTRODUCTION
In recent years, autonomous mobile manipulation has
emerged as a new research area of robotics and has been
identified as critical for future robotic applications [1]. This
area comprises and groups together so far independent re-
search topics into a common ground. Mobility, perception,
manipulation and, primarily, the use of all these subareas
within a single system that is able to perceive and understand
Fig. 1. Robotic system AILA
its environment, move around, manipulate and learn about
objects is the goal of the newly popular research area. Ob-
viously, mobile manipulation shares many similarities with
’butler’ robot HERB [3] developed at Intel Research Labs in
the area of humanoid robotics but with a clear difference:
collaboration with Carnegie Mellon University, Rollin’ Justin
anthromorphism and, specifically, the use of legs is not
[4] developed at DLR Institute of Robotics and Mechatronics
mandatory. The focus lies on the combination of mobility
in Germany, and the UMan [5] from the Robotics and
and manipulation, that is, not only on the use of non-mobile
Biology Lab at the University of Massachusetts Amherst.
robot manipulators or not only on the use of rovers, but on the
PR-2 is is a dual-arm robot with an omni-directional mobile
combination of these systems into a single one with enhanced
base. It includes a variety of sensors as tilting laser scanner
capabilities to move around and modify the arrangement
on the head, and laser scanner on the mobile base, two pairs
of the environment. The area of mobile manipulation is
of stereo cameras, and an IMU located inside the body.
primarily concerned with dynamic and unstructured envi-
Currently, PR-2 is likely the most advanced autonomous
ronments that might be partially or completely unknown,
mobile manipulator, able to successfully perform complex
which requires the robot to acquire information and adapt
manipulation and navigation tasks. HERB is an autonomous
autonomously to the ongoing situation.
mobile manipulator that was designed for performing com-
Noteworthy examples within the area of mobile manipula-
plex manipulation tasks in home environments. The robot is
tors are the PR-2 robot [2] developed by Willow Garage, the
able to search, recognize, and store new objects as well as
The work presented in this paper is part of the SemProM project, funded manipulate door handles and objects and navigate in cluttered
by the German Federal Ministry of Education and Research (BMBF), grant environments. Rollin’ Justin is a progressive development
No. 01IA08002.
Alls authors are with the DFKI Robotics Innovation Center, Bremen, that builds on the well-known lightweight arms from DLR
Germany. Email: [email protected] (LWR-III) to build first a dual-arm robot (Justin) that fi-
nally has been equipped with a mobile base to enhance strong drives. These requirements have already been aimed at
the robot’s field of work. The UMan robot consists of a successfully by other robotic systems [6][7], but not with the
modified Nomadic XR4000 holonomic mobile base with limitation of a real human-size design space. The following
three degrees of freedom, a WAM seven degree-of-freedom requirements are axioms that had an important influence on
manipulator arm, and a four degree-of-freedom hand from the system’s concept:
Barret Technologies. • The payload-to-weight ratio of each arm shall be as high
as technically possible to enable fast reactions with high
II. SYSTEM OVERVIEW
accelerations.
AILA consists of an anthropomorphic upper body • The overall look-and-feel of the robot shall be anthro-
mounted on a wheeled mobile platform. The upper body pomorphic and aesthetical.
carries two arms, each of them with seven degrees of • The development of the arm drives shall be an improve-
freedom, a torso with four joints, and a head with two ment of existing components developed at our center for
degrees of freedom. The mobile platform (Fig. 5) consists previous robotics systems (Spaceclimber [8]) while, at
of six wheels with two degrees of freedom each, one for the same time, ensuring compatibility with them.
the steering axis and one for driving the wheel. The main • The working height of the arms and the center of gravity
focus of the design was on the development of the arms and of the upper limbs shall be adjustable with regard to the
the upper torso, being the mobile base a first solution to mobile base.
provide with mobility to the robot. Future developments will • The mobile base shall move holonomically on slightly
concentrate on new concepts of mobility/locomotion. The rough terrain.
robot’s hardware includes two Prosilica GC780C cameras
that create a stereo system unit in the head which is mounted IV. MECHATRONIC DESIGN
on a neck able to pan and tilt on an anthropomorphic
A. Mechanics
path. A periodically-tilting short-ranged Hokuyo URG Laser
scanner in the chest and a Mesa SR-4000 3D Time of For the design of the arms, a kinematic model of seven
Flight (TOF) camera in the robot’s stomach are combined for degrees of freedom was chosen with pair-wise grouping
object and scene recognition, as well as for pose estimation. of joints with intersecting axes at the wrist, and three
Two long-ranging Hokuyo UTM Laser scanners provide a intersecting axes at the shoulder (Fig. 2, Top). The grouping
circumferential view for the mobile base. These six different of joints lowers the weight of their combined housing and
visual systems allow to extract a multimodal view of the helps to reduce the moment of inertia of the limb by placing
environment. AILA is equipped with three computers. Two the relatively heavy motors near to the arm’s base. The joints
3,5-inches embedded PCs: one for motion control located in of the elbow and shoulder consist of brushless DC motors
the head and one for navigation located in the mobile base. in combination with harmonic drives and are independently
A mini-ITX board in combination with a dedicated graphics controlled by a stack of three PCBs housing power and
card for vision processing is located in the torso. The control electronics. The two degrees of freedom of the wrist
communication network consists of five independent CAN- are controlled by a parallel kinematic structure driven by two
lines for controlling the two arms, the torso, and the wheel DC linear motors, each of them including planetary gear,
modules of the base. GigaEthernet routed through two five- encoder, and its own controller.
port switches connect the head cameras, the three computers, The assembly of two joints in one housing (Fig. 2,
and the outside world. The motion computer communicates Bottom) is connected to the upper and lower arm structure by
through a dedicated RS-485 bus with the Skyetek M4 RFID four-point thin-section bearings. The housing itself consists
module which is integrated together with its antenna in the of machined aluminum parts joined together by carbon-
left hand. Further external sensors are two six-axis force- fiber reinforced plastic parts. Within each joint, a brushless
torque sensors at the robot’s wrists. Proprioceptive sensors motor drives a shaft that is supported on the one end by
include the hall-sensors for the measurement of the absolute a bearing and on the other directly on the wave generator
position of the joints, motor current measurements, and of the connected harmonic drive gear. Within the harmonic
incremental optical encoders for the motor commutation and drive, a spring-loaded electromagnetic safety-brake acts on
joint torque measurement. an axially-mounted brake disk. All necessary cabling goes
through the centre of the joint, guided by a tube that also
III. DESIGN CONSIDERATIONS acts for driving a magnet hovering over an absolute hall
In spite of AILA’s anthropomorphic appearance, its basic encoder on the PCB stack. The reason for a tube diameter of
purpose is to be a research platform for autonomous mobile 8mm is due to the opening of the smallest harmonic drives
dual-arm manipulation. Therefore, its underlying system con- flex-spline, used in the axial joint of the forearm. Optical
cept is a pair of arms mounted to a height-adjustable frame incremental encoders are integrated to measure the position
on a mobile base. To enable fast and precise coordinated of the rotor and commutate the stator accordingly.
dual-arm movements, the primary design goal is a high The conceptual considerations for the body and the head of
payload-to-weight ratio of the arms by featuring low masses, AILA follow a different approach than the structural concept
low moments of inertia, high stiffness of the limbs, and for the arms. The basic purpose of the body is to carry the
 Fig. 3. Body specification and sheet metal structure

B. Electronics and Power Supply

AILA’s electronics include various sensor modules, three
Fig. 2. Top: Arm specification. Bottom: Detail of the elbow housing computers, and three different kinds of motors. Therefore, it
containing two joints has to provide various voltages with noticeable currents for
the CPUs and the drives. The main system setup differen-
tiates between logic and power circuits. For safety reasons,
there is one power cut-off on the mobile base near the battery
arms and make their working space adjustable in height. The that mechanically breaks all circuits, and a second wireless
structural parts of the torso, leg, and head are sheet metal kill-switch affecting just the drive circuits for paralyzing
parts (Fig. 3), because additionally to being lightweight they the robot without interrupting the sensor and logic network.
also have to be exchangeable. Since AILA is a research Within the mobile base, the current of the 48V battery is
platform, it is foreseeable that there will be future changes split to a 1 kW 24V power supply, a 0,5 kW 12V power
to the sensory setup or the CPUs. The leg and the torso as supply, and direct supply of the wheel and arm drives. Fif-
main load-carrying structures have also as second function teen different sub-circuits are separated by automatic circuit
the housing of the vision CPU, a laser scanner, and a 3D breakers. Within each sub-circuit all damageable components
TOF camera for navigation, network components, and the are protected with fast-reacting fuses according to their
control board for its motors. nominal power consumption. Voltages below 12V, that are
The basic concept for designing lightweight structures is needed for the chest laser-scanner and the mini ITX board,
to achieve an evenly distributed stress near to the yield point are converted by a 120W CPU power supply placed in the
within all of the used material. Besides choosing a material torso.
with high specific Young modulus, the stiffness of a robotic The main wire harnesses are the power cables within
arm increases by a high second moment of area with its the mobile base, power, and communication cabling from
cross-section, which leads to choose a thin shelled tube with the base to the chest, and three branches to the arms and
as few openings as possible. Therefore, AILA’s arm structure the head. Within the mobile base, a noticeable amount of
principle design (Fig. 4) is a tube of high-modulus carbon design space is dedicated to the high-current power cables.
fiber along the force paths in combination with machined The arrangement of components was chosen to optimize the
high-strength aluminum where precision is needed or the routing of these stiff cables with large bending radii. Due to
loads are diverse. the main power converters and the circuit breakers placed in
 current, speed, and position sensors thus enabling local motor
control. The high-level commands are transmitted from the
embedded PCs via CAN messages. The arm joints and the
horizontal axes of the mobile platform use brushless DC mo-
tors from Robodrive. A similar control approach to the one
previously described has been used for these motors, which
has already been successfully integrated in the SpaceClimber
robot [8][9]. In this case, the in-house developed motor
electronics consists of a stack of three circular PCBs. These
PCBs are designed to be integrated directly behind the stator
of the Robodrive motors. The boards incorporate all sensors
which are necessary to monitor and control the motor. Three
motor current sensors are integrated in the low phases of the
three-phase H-bridges. Encoder wheels in front of the gear as
well as an absolute angular encoder behind the gear measure
the motor position for speed and position control purposes.
The stack is also equipped with two temperature sensors
and input voltage measurement. Additionally, one SD-card
Fig. 4. Hybrid joint and left arm structural parts module is used to log all sensor data at the highest possible
frequency. The interface to high-level control units is realised
via a CAN interface. Configuration data specific to each
motor - e.g. CAN identifier or position offsets - is written and
read to/from an EEPROM memory. All mentioned sensors as
well as current, speed, and position controllers are processed
by a Spartan3 FPGA from Xilinx.

B. Control of the mobile base

The control system consists of two different control units,
one micro controller and one FPGA (Fig. 6), each of
them communicating over a CAN bus with the navigation
computer. Each of these controller-pairs controls one wheel,
which consists of two motors which are coupled over a
common base point at the ground of the wheel. The vertical
axis steers the wheel in a range of 180 degrees by using a
Faulhaber DC motor. The horizontal axis controls the wheel
Fig. 5. Wheeled mobile base used to provide mobility to the first platform. speed of the RoboDrive brushless DC motor. All sensors
Future developments will focus on new concepts of mobility/locomotion for needed for monitoring the state of the motors as well as
AILA. processing modules for current, speed and position control
as well as data-logging are directly integrated in these control
units. The described arrangement allows linear, lateral, and
the mobile base, the number of cables routed to the torso
rotary movements and hence offers the rover an almost omni-
increases. All these cables are bundled to one harness that
directional movability. The wheel control system sends status
follows a route similar to the placement of a spine with
messages such as the actual speeds, angles, temperatures,
minor deviations in the hip, the shoulders, and the head. To
and power consumption to the navigation computer, whereas
provide enough slack for the head-movement, the spine-like
desired wheel speed and steering angles are the possible
harness goes with a large bending radius from the neck into
messages received from the wheel controllers.
the design space of the tongue. For the ease of maintenance,
mainly pre-tailored cables are used. Excessive cable-length is
stored in the right hemisphere of the head, all communication C. Control of the arms and torso
and debugging interfaces are pointing to the left. The overall architecture of the manipulation software is
shown in Figure 7. The coordinator is implemented as a
V. CONTROL SOFTWARE
hierarchical state machine and controls the robot at task-
A. Control of the motor joints level. It makes use of the behavior base, which represents a
The torso joints as well as the vertical axes of the mobile collection of basic robot functionalities (e.g. Plan Trajectory,
platform use Faulhaber DC Motors which are controlled by Tilt Head, ...) that can be combined to achieve more complex
an in-house developed power electronics board controlled behaviors. The behaviors themselves are implemented in
by a STM32 microcontroller. The board is equipped with different modules and can be triggered by action calls of the
 the upper body for 3D shape recovery. Figure 8 shows a
typical manipulation scenario, i.e., a shelf containing several
objects for manipulation. In order to execute high-level
commands like “Take the yellow box from the shelf” a
semantic interpretation of the sensor data is mandatory. The
robot needs to identify the shelf from the raw 3D data
in order to approach it. Currently, we work predominantly
on the detection of spatial entities like tables, doors, and
shelves and how they can be described using spatial feature
descriptors [11].
In our previous work ([12],[13] and [11]) we described
in detail how the bridging between the robot’s perception
and the symbolic description can be achieved. A tilting
laser range finder generates points in a geometric coordinate
system. In our semantic perception approach, we use the 3D
point cloud data from the tilting laser in order to recover
structural information about the robot’s environment. Using
a modified region-growing approach [13] which is based
on Rabbani’s Algorithm [14], planes are detected in the
Fig. 6. Control scheme of the mobile base unorganized point cloud. Each detected plane is analyzed
for the structural information contained. In order to further
process the detected planes, the convex shape of each single
coordinator. Thereby, the motion planner entails the func- plane is calculated using alpha shape recovery [15]. Once
tionality for trajectory planning, whilst the motion controller the shapes have been recovered from the unorganized point
contains routines for trajectory execution and other hardware- cloud, the goal is to classify the structure the robot perceives
related features. The world model collects information about and to label the structure with semantics. To make semantic
the robot environment (currently, mainly from the vision labeling possible in indoor environments, we make use of
module) and the robot’s current configuration. Moreover, some basic assumptions. If we look around in a typical
upon request this module supplies this information to any indoor environment like a household environment or an office
other modules. For interprocess communication and as inte- environment, it is clear that most structures are of rectangular
grating software platform, we use the open-source framework shape and mostly parallel or orthogonal to each other. The
ROS [10]. robot has to extract a vector of feature descriptors from the
spatial entities in order to compare them with the semantic
knowledge database. In a first approach, we define a set
of vectors which are able to describe spatial entities of an
environment.

Fig. 8. Left: The robot AILA perceives the shape of the shelf by the
segmentation of the point cloud, generated by a tilting 2D laser range finder
in the upper part of the body. Right: The shelf as it is perceived by the robot
Fig. 7. Architecture of the manipulation framework
In our current implementation, we again consider a like-
lihood function in order to deal with uncertainties. For
D. Semantic Perception instance, two shapes can be parallel with the certainty of
The robot AILA perceives the 3D environment through 0.9 due to noise and rounding differences in the extraction
a variety of sensor modalities. A stereo vision system is process. Using a model based on Spatial Ontologies, which
integrated into the head for depth and texture perception. can be expressed with the language OWL-DL , detectable
Additionally, a tilting laser range finder is integrated into features are thereafter matched using a similarity function.
For a detailed description of the semantic labeling approach, plans. While the constraints are incorporated during the
refer to [11]. planning phase, the criteria can be used to select an optimal
path from a set of paths or ask the planner to replan if the
E. Vision and Manipulation
criteria are not met.
The stereo vision system, which is integrated in the
head of the robot, consists of two Prosilica color video
cameras which are used for the perception of AILA’s nearby
operating area, including the arms. On the software side, the
vision module contains algorithms for object recognition and
tracking, scene interpretation, self-perception, and hand-eye
calibration.
For object recognition, we generate textured 3D mod-
els of the objects in AILA’s environment, extract global
properties, as well as local, texture-based features (e.g.
SIFT features [16]), and store this information in a model
database. While in operation, the vision framework consis-
tently matches the features extracted from the observed scene
with the database and computes the pose of the recognized
objects using 2D-3D point correspondences (see Figure 9).
Fig. 10. OpenRAVE-based simulation of AILA performing a dual-arm
manipulation task

VI. EXPERIMENTS
First experiments have been performed in which AILA
made use of its key capabilites: semantic perception for
recognising objects in an office environment, autonomous
navigation, object recognition and pose estimation, and au-
tonomous dual-arm manipulation. Figure 11 shows snap-
shots of the sequence for grasping an object using two
arms in a laboratory setup. The robot recognised the table,
Fig. 9. Feature matching and 3D pose estimation navigated towards it, recognised the object on the table as
well as its 3D pose, and planned the manipulation strategy
For motion planning of the upper body to grasp and to grasp and manipulate the object using two arms.
manipulate objects detected by the vision system, the kine-
matic relationships are defined between the 4 torso joints, 14 VII. CONCLUSIONS/FUTURE WORK
arm joints and 2 head joints. For tasks involving dual-arm Robots have been and still are great tools for studying
grasping and manipulation of homogeneous objects (Fig. 10), artificial intelligence. While in the earlier days of AI systems
motion plans are generated for 18 DOFs (arms and torso). like Shakey [19] were designed to study aspects of planning,
Incorporating the torso joints in the motion planning allows plan execution monitoring, and navigation, their primary
the utilization of the entire reachability space of the robot achievements where to carry the sensors (mainly cameras)
which is necessary for constrained dual-arm manipulation through the environment that was to be explored. Today we
tasks. OpenRAVE [17] is used as the common platform understand that the central question of AI research can only
for integrating the various modules of motion planning be answered if the tools we use - the robots - provide a larger
like kinematics, workcell representation, planning, collision amount of robotic capabilities. Internal representation of a
avoidance, trajectory smoothing, and simulation. The plan- robot exploring an environment can be massively enriched
ning algorithms are based on the Bi-directional Rapidly- if along with a camera and other optical data (laser scanner-
exploring Random Trees approach [18] and additional task- based point clouds), internal and proprioceptive data is used.
based constraints and criteria are incorporated for dual- This data can come e.g. from the motors in the kinematic
arm manipulation. The constraints include maintaining the chain of a robot arm that handles objects in the environment,
relative end-effector (EEF) configurations between the two for instance, opening a door. The higher the disposition of
arms which ensures that the object grasped is not lost while it robots to interact (actively) with the environment, the better
is being moved. For tasks involving open-top or liquid-filled will be the database for environmental representation and
object manipulation, additional constraints are imposed on the enhancement of their strategies for control, navigation,
the orientation of the object while it is being manipulated. and planning. In this paper, we presented a robotic platform
Further, additional criteria like minimization of the lengths that incorporates multiple degrees of freedom as well as
of the EEF paths are used for selecting the optimum motion integrates a great deal of sensors to allow a self-evaluation of
 Fig. 11. Preliminary experiments of AILA performing autonomous dual-arm manipulation

the current robot’s state and scenario. Such platform should [9] S. Bartsch, T. Birnschein, F. Cordes, D. Kühn, P. Kampmann,
permit the emergence of learning capabilities by exploiting J. Hilljegerdes, S. Planthaber, M. Römmermann, and F. Kirchner,
“SpaceClimber: Development of a six-legged climbing robot for space
its immersion in human environments with its own body exploration,” in Proceedings for the Joint Conference of ISR 2010
resources and limitations. (41st International Symposium on Robotics) and ROBOTIK 2010 (6th
German Conference on Robotics). VDE Verlag GmbH, June 2010.
R EFERENCES [10] M. Quigley, B. Gerkey, K. Conley, J. Faust, T. Foote, J. Leibs,
E. Berger, R. Wheeler, and A. Ng, “ROS: an open-source robot
[1] O. Brock and R. Grupen, “Nsf/nasa workshop on autonomous mobile operating system,” in In Proc. of IEEE International Conference on
manipulation (amm),” http://robotics.cs.umass.edu/amm, March 2005, Robotics and Automation (ICRA), 2009.
houston, USA. [11] M. Eich, M. Dabrowska, and F. Kirchner, “Semantic labeling: Clas-
[2] S. Chitta, B. Cohen, and M. Likhachev, “Planning for autonomous sification of 3D entities based on spatial feature descriptors,” in
door opening with a mobile manipulator,” in In Proc. of IEEE Workshop Best Practice Algorithms in 3D Perception and Modeling
International Conference on Robotics and Automation (ICRA), 2010. for Mobile Manipulation, IEEE International Conference on Robotics
[3] S. Srinivasa, D. Ferguson, C. Helfrich, D. Berenson, A. Collet, R. Di- and Automation, (ICRA-10), May 2010, Anchorage, 2010.
ankov, G. Gallagher, G. Hollinger, J. Kuffner, and M. VandeWeghe, [12] M. Eich and F. Kirchner, “Reasoning about geometry: An approach us-
“HERB: a home exploring robotic butler,” Autonomous Robots, 2009. ing spatial-descriptive ontologies,” in Workshop AILog, 19th European
[4] M. Fuchs, C. Borst, P. R. Giordano, A. Baumann, E. Kraemer, Conference on Artificial Intelligence, (ECAI-10), 16.8.-16.8.2010, Lis-
J. Langwald, R. Gruber, N. Seitz, G. Plank, K. Kunze, R. Burger, bon, 2010.
F. Schmidt, T. Wimboeck, and G. Hirzinger, “Rollin’ justin - design [13] M. Eich, M. Dabrowska, and F. Kirchner, “3D scene recovery and
considerations and realization of a mobile platform for a humanoid spatial scene analysis for unorganized point clouds,” in In Proceedings
upper body,” in ICRA’09: Proceedings of the 2009 IEEE international of 13th International Conference on Climbing and Walking Robots and
conference on Robotics and Automation. Piscataway, NJ, USA: IEEE the Support Technologies for Mobile Machines, (CLAWAR-10), 31.8.-
Press, 2009, pp. 1789–1795. 03.9.2010, Nagoya, Japan, 2010.
[5] D. Katz, E. Horrell, O. Yang, B. Burns, T. Buckley, A. Grishkan, [14] T. Rabbani, F. van Den Heuvel, and G. Vosselmann, “Segmentation of
V. Zhylkovskyy, O. Brock, and E. Learned-Miller, “The UMass point clouds using smoothness constraint,” in International Archives
mobile manipulator UMan: An experimental platform for autonomous of Photogrammetry, Remote Sensing and Spatial Information Sciences,
mobile manipulation,” in In Workshop on Manipulation in Human vol. 36, no. 5. Citeseer, 2006, pp. 248–253.
Environments at Robotics: Science and Systems, 2006. [15] W. Shen, “Building boundary extraction based on lidar point clouds
[6] G. Hirzinger, N. Sporer, A. Albu-Schaffer, M. Hahnle, R. Krenn, data,” in ISPRS08. ISPRS, 2008, p. 157.
A. Pascucci, and M. Schedl, “DLR’s torque-controlled light weight [16] D. G. Lowe, “Object recognition from local scale-invariant fatures,” in
robot III - are we reaching the technological limits now?” in IEEE The Proceedings of the Seventh International Conference on Computer
International Conference on Robotics and Automation (ICRA)., vol. 2, Vision, vol. 2, 1999, pp. 1150–1157.
2002, pp. 1710–1716. [17] R. Diankov and J. Kuffner, “OpenRAVE: A planning
[7] C. Ott, O. Eiberger, W. Friedl, B. Bauml, U. Hillenbrand, C. Borst, architecture for autonomous robotics,” Robotics Institute,
A. Albu-Schaffer, B. Brunner, H. Hirschmuller, S. Kielhofer, R. Koni- Tech. Rep. CMU-RI-TR-08-34, July 2008. [Online]. Available:
etschke, M. Suppa, T. Wimbock, F. Zacharias, and G. Hirzinger, “A http://openrave.programmingvision.com
humanoid two-arm system for dexterous manipulation,” in 6th IEEE- [18] J. Kuffner and S. LaValle, “RRT-connect: An efficient approach to
RAS International Conference on Humanoid Robots, Dec. 2006, pp. single-query path planning,” in Proc. IEEE International Conference
276–283. on Robotics and Automation (ICRA), April 2000, pp. 995–1001.
[8] J. Hilljegerdes, P. Kampmann, S. Bosse, and F. Kirchner, “Develop- [19] N. J. Nilsson, C. A. Rosen, B. Raphael, G. Forsen, L. Chaitin, and
ment of an intelligent joint actuator prototype for climbing and walking S. Wahlstrom, “Application of intelligent automata to reconnaissance,”
robots,” in International Conference on Climbing and Walking Robots Stanford Research Institute, Tech. Rep., December 1968, project 5953
(CLAWAR-09), 2009, pp. 942–949. Final Report, From the Nilsson archives, SHAKEY papers.

View publication stats