Fuzzy Controller Design Of AUV Based On Genetic Algorithms

Although fuzzy logic controllers have been successfully applied in many complex industrial processes, they experience a deficiency in knowledge acquisition and rely to a great extent on empirical and heuristic knowledge which, in many cases, cannot be objectively elicited. Among the problems to be resolved in fuzzy controller design are the determination of the linguistic state space, definition of the membership functions of each linguistic term and the derivation of the control rules. In recent years machine learning methods have been applied to this kind of intelligent control problems. One of the methods is the Genetic Algorithm(GA). The feasibility of designing fuzzy logic controller for Autonomous Underwater Vehicle(AUV) with the help of GA is discussed in this paper, and the testbed used here is an AUV which is developed by Harbin Engineering University.Based on the AUV's motion simulation system, the basic fuzzy logic controller is established using manual trial-and-error process at first. Then two approaches to discuss the feasibility of applying GA to AUV's fuzzy logic controller design are presented. One approach is to optimize membership function with GA based on a hand-designed fuzzy rule base. Another approach is to design membership function and fuzzy rule base with GA simultaneously. To check the effectiveness of each method, simulation research is done, which takes longitudinal-control and side-control as examples. The performance of the controller is discussed.The simulation results indicate that the designing method base on GA is feasible to AUV. The approaches discussed in the paper are practicable for the research of AUV's maneuverability and adaptive control.