Title :
Autonomous Vehicle Control and Obstacle Avoidance Concepts Oriented to Meet the Challenging Requirements of Realistic Missions
Author :
Shmaglit, Anna ; Rinat, Keren ; Brand, Zev ; Fischler, Amit ; Velger, Mordekhai
Author_Institution :
ELOP Electro-Opt. Industries Ltd., Rehovot
Abstract :
This paper presents the control and obstacle avoidance concepts for off-road autonomous vehicle. The hierarchical control system includes three levels which mimic driver´s actions: activating the actuators, maintaining desired off-road path, that is bounded by natural obstacles. The main achievement of the proposed control system, proven in extensive field-tests, is operating at high speeds in harsh terrains with accuracy better than 1 meter, while maintaining vehicle dynamic stability during all maneuvers. The high performance is preserved also when driving on sparse waypoints database and in obstacle-dense environments. The obstacle avoidance algorithm was designed with emphasis on efficiency: avoidance path is constructed in real-time without slowing the vehicle down. In the Qualification-Inspection-Demonstration phase of the DARPA Grand Challenge Race 2004 the vehicle avoided all 13 obstacles in a record time of 6:45 minutes
Keywords :
collision avoidance; hierarchical systems; mobile robots; robot dynamics; stability; actuators; avoidance path; hierarchical control system; obstacle avoidance; off-road autonomous vehicle; off-road path; path following; vehicle dynamic stability; Algorithm design and analysis; Control systems; Databases; Mathematical model; Mobile robots; Path planning; Remotely operated vehicles; Stability; Testing; Vehicle dynamics; autonomous vehicles; control; dynamics; obstacle avoidance; path following;
Conference_Titel :
Control, Automation, Robotics and Vision, 2006. ICARCV '06. 9th International Conference on
Conference_Location :
Singapore
Print_ISBN :
1-4244-0341-3
Electronic_ISBN :
1-4214-042-1
DOI :
10.1109/ICARCV.2006.345219