Abstract：We propose a fuzzy adaptive controller for trajectory following of leg-wheeled robot. The initial fuzzy rules are extracted by using a traditional PD controller. Then expert experiences are used to complement the rules. Finally the error back-propagation algorithm is used for on-line adaptive parameter tuning. Simulation results verify that this method has better trajectory following precision and better rejection of disturbance.
 Gen Endo, Shigeo Hirose. Study on roller-walker: system integration and basic experiments [A]. Proc ICRA [C] Detroit, Michigan: IEEE, 1999. 2032-2037  Yu-jie Dai, Eiji Nakano. Motion control of leg-wheel robot for an unexplored outdoor environment [A]. Proc IROS [C] Tokyo Japan: IEEE, 1996. 402-409  Hironori Adachi, Noriho Koyachi. Development of a leg-wheel hybrid mobile robot and its step-passing algorithm [A]. Proc IROS [C] Maui Hawaii: IEEE, 2001. 728-733  Gen Endo, Shigeo Hirose. Study on roller-walker: multi-mode steering control and self-contained locomotion [A]. Proc ICRA [C]. San Francisco, CA:IEEE, 2000. 2808-2814  Shigeo Hirose, Hiroki Takeuchi. Study on roller-walk: basic characteristics and its control [A]. Proc ICRA [C]. Minneapolis, Minnesota: IEEE, 1996. 3265-3270  Craig J J. Introduction to robotics-mechanics and control [M]. Reading. Massachusetts: Addison-Wesley Publishing Company, Inc, 1986  陆文娟,戴民. 基于模糊神经网络的机械手自适应控制 [J].清华大学学报,1999,(5):24-27  孙增圻. 智能控制理论与技术 [M].北京:清华大学出版社,1997