In allusion to the autonomous behavior of autonomous underwater vehicle (AUV), it is difficult to research its adaptability and reaction to the environment at present. According to the problems of motion planning in uncertain environment, behavior models were proposed and explained for AUV horizontal motion. Models for goal forward behavior, obstacle avoiding behavior and macro-behavior for horizontal plane were built through the selection of behavior variables. The simulation result in horizontal plane for AUV showed that AUV motion was accord with human behavior mode, at the same time the autonomous behavior reacted to unstructured environment fast, correctly and effectively, so the safety problem for AUV in complicated environment was solved.
[1]
Hyung Suck Cho,et al.
A sensor-based navigation for a mobile robot using fuzzy logic and reinforcement learning
,
1995,
IEEE Trans. Syst. Man Cybern..
[2]
Keith B. Hall,et al.
Correlated Q-Learning
,
2003,
ICML.
[3]
Joan Batlle,et al.
Efficient learning of reactive robot behaviors with a Neural-Q/spl I.bar/learning approach
,
2002,
IEEE/RSJ International Conference on Intelligent Robots and Systems.
[4]
Yoram Koren,et al.
Potential field methods and their inherent limitations for mobile robot navigation
,
1991,
Proceedings. 1991 IEEE International Conference on Robotics and Automation.