Abstract：The ability of obstacle avoidance based on forward looking sonar is a basic requirement of AUV(autonomous underwater vehicle) during a long range travel in the unknown ocean environment.A real-time obstacle-avoiding expert system based on a multibeam forward looking sonar is presented.The specifications of sonar and the description of obstacles are introduced.Structure of the obstacle-avoiding expert system is given.The decision-making process of the obstacle-avoiding expert system is illustrated.Knowledge base of the expert system is designed based on the information of task execution.The inference engine is designed and implemented according to the images of sonar,which can send instructions of obstacle avoidance and path re-planning.Lake test validates the obstacle-avoiding expert system.
[1] Borenstein J,Koren Y.Real-time obstacle avoidance for fast mobile robots[J].IEEE Transactions on Robotics and Automation,1991,7(3):278-288.
[2] 张祺,杨宜民.基于改进人工势场法的足球机器人避碰控制[J].机器人,2002,24(1):12-15.
[3] Johann B,Yoram K.The vector field histogram-fast obstacle avoidance for mobile robots[J].IEEE Transactions on Robotics and Automation,1991,7(3):278-288.
[4] DeMuth G,Springsteen S.Obstacle avoidance using neural networks[A].Proceedings of the Symposium on Autonomous Underwater Vehicle Technology[C].Piscataway,NJ,USA:IEEE,1990.213-215.
[5] 张汝波,顾国昌.水下机器人模糊局部规划器设计[J].机器人,1996,18(3):258-162.
[6] 李智军,吕恬生.水下机器人模糊局部规划器设计[J].机械科学与技术,2000,19(4):547-549.
[7] Ying N,Eicher L,Xunzhang W.Real-time 3D path planning for sensor-based underwater robotics vehicles in unknown environment[A].Oceans Conference Record[C].Piscataway,NJ,USA:IEEE,2000.2051-2058.
