Autonomous Vehicle Navigation: Homework
Autonomous Vehicle Navigation: Homework
Autonomous Vehicle Navigation: Homework
by
Several algorithms can be used for self- driving vehicles in the environments where
they exist. These algorithms can be categorized as follows:
Artificial intelligence, bug algorithms, artificial potential field and nearness diagram
algorithm. The selection among the above-mentioned algorithms depends on many
factors, whereas the three key ones are calculation time, path calculation and the type
of obstacle deal with. The calculation time is the time required by the algorithm to
specify the optimum path of several possible ones to take the vehicle from its current
position to the desired one, which in fact considered an important factor in autonomous
vehicles. The path calculation is how to determine the distance between the start point
and the destination one. The type of the obstacle to be avoided is also another
important factor where many algorithm pre-defined to deal with known environment so
the mobile obstacle are not considered in these algorithms. Here, we are going to
focus our explanation on one of those algorithm which is bug algorithm specifically
bug 1 algorithm and bug 2 algorithm.
1
1.2- Bug 2 algorithm
When we look at this bug 2 algorithm, we find that it is an update for the bug 1
algorithm, which was developed to reduce the time and distance necessary to be
moved from the beginning point till the end point. This development is achieved
through specifying a straight line between the start point and the target one as a
reference for the vehicle path. Then the vehicle to go from the start position the desired
one it must to follow that reference straight line and when an obstacle confronts the
vehicle path, the vehicle becomes forced to turn left or right according to its pre-defined
instructions following the edge or perimeter of the obstacle till it reaches the straight
line on the other side of the obstacle then it complete its path following the straight line
towards the target without any need to go around the whole perimeter of the obstacle
to specify the closest point to the target position such as in bug 1 algorithm
2
2.1 -autonomous vehicle navigation system in unknown environment
In fact, this system is considered very important for the sake of human being where it
makes the vehicles capable of reaching difficult locations or performing dangerous
tasks instead of a human.
This system allows the vehicle to freely move from its current position to any desired
position the human wants without any need to human continuous interference. For
vehicle to autonomously move to the desired position, an optimum path must be
specified among several ones. And this is done by intelligence control system as a
main controller, GPS system, GPRS modem, compass and internal and external
sensors. The navigation process can be divided into parts as follow:
3
destination longitude and latitude on Google map using GPRS that communicates
vehicle system with the internet.
Then the location system stores that destination and enable GPS module to connect
the satellite in order to determine the vehicle current position, these two positions are
employed by the algorithm which in turns calculates the distance as the difference
between the start and end positions data and the angle by taking tangent inverse for
the longitudes and latitudes, after that the algorithm does many computations on
several paths to extract the best path. When the best path is selected the main
controller interprets all the previous information and sends commands to the vehicle
to start motion through drivers which interprets these command into the output
actuators "may be electric motors", during the motion information about velocity and
torques is continuously required by the main controller to be capable correct any
deviation away from the selected path. Also laser or camera feedbacks are used to
sense obstacles in vehicle environment.
2.2- Navigation system in known environment
This navigation system depends on vision systems and external sensor with the help
of pre-defined paths such as material handling in industries where the required motion
is from a known point to another through also a known paths, sensors only used to
avoid any instantaneous obstacles. These to do the repetitive cycles which the human
feels bore because no new action.
References
1. Dhanasingaraja R, Kalaimagal S, Muralidharan G "Autonomous Vehicle Navigation and
Mapping System". International Conference on Innovations in Engineering and Technology
(ICIET’14). Vol. 3, Special Issue 3, March 2014.
2. Vayeda Anshav Bhavesh"Comparison of Various Obstacle Avoidance Algorithms".
International Journal of Engineering Research & Technology (IJERT), ISSN: 2278-0181, Vol. 4
Issue 12, December-2015.
3. Justin E. Manley. " Unmanned Surface Vehicles, 15 Years of Development". Battelle Applied
Coastal and Environmental Services 397 Washington St. Duxbury, MA 02332
4. Jacoby Larson*, Michael Bruch, and John Ebken " AUTONOMOUS NAVIGATION AND
OBSTACLE AVOIDANCE FOR UNMANNED SURFACE VEHICLES" Space and Naval Warfare
Systems Center, San Diego, 53560 Hull St., San Diego, CA 92152-5001
5. Hui-Zhong Zhuang, Shu-Xin Du, Tie-Jun Wu. ''REAL-TIME Path Planning For Mobile Robots''. In Pro
ceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou,
18-21 August 2005.
4
6. C. M. Johnson And J. O. Gray. “Hierarchical Path-Planning for a Mobile Robot”. Eng. Applic. Artif. lntell.
Vol.7, No. 2, pp. 137-149, Elsevier Science Ltd, 1994.
7. Rodney A. Brooks, . “Solving the find Find Path Problem by Representing Free Space Generalized
Cones”. Technical Report A.I MEMO No. 674, Massachusetts Institute Technology, May 1980.
8. Ray Jarvis. Distance transform based path planning for robot navigation," Robotics and Intelligent
Systems, Vol. 11 of, World Scientific, 1993.
9. Dijkstra, E. W. ''A Note On Two Problems In Connexion With Graph''. Numerische Mathematik . Vol.
1: pp. 269–271,1959.
10. P. Hart, N. Nilsson and B. Raphael, “A Formal Basis for the Heuristic Determination of Minimum Cost
Paths,” IEEE Transactions on Systems Science and Cybernetics, Vol. 4, No. 2, pp. 100- 107,1968.
11. Stentz, A. “Optimal and efficient path planing for unknown and dynamic environments”. Technical
Report, Carnegie Mellon Robotics Institute Technical Report, CMU-RI-TR-93-20,1993.
12. Oussama Khatib. “Real-time obstacle avoidance for manipulators and mobile robots," International
Journal of Robotics Research, Vol. 5, No. 1, pp. 90-98, 1986.
13. Nearchou A. “Path planning of a mobile robot using genetic heuristics”, Robotica, Vol. 16, No. 5, pp.
575-588, ISSN 0263- 5747,1998.
14. M. Dorigo et L.M. Gambardella, “AntColony System : A Cooperative Learning Approach to the
Traveling Salesman Problem”, IEEE Transactions on Evolutionary Computation, Vol.1, No.1, pp 53-66,
1997.