关键词: CMAC神经网络, 辨识, 块对角控制系统, 变结构控制

Abstract: A combined control algorithm is proposed to overcome the drawbacks of the inverse solution for the block diagonal control system(BDCS) of the missile. A few important aerodynamic parameters are identified by using cerebellum model arithmetic controller(CMAC) neural network. Different control methods, such as CMAC neural network based variable structure control(VSC), fuzzy control theory, analytical inverse solution and VSC, are used to design the controller in order to improve the performance of the system. By using adaptive law to estimate the error of estimation, the uncertainty of the system is reduced. The variable structure gain is tuned by the fuzzy logic. For this scheme, stable updated laws are determined by Lyapunov theory, and the boundedness of the estimated error, the weight error and all signals in the closed loop system is guaranteed. No prior offline training phase is necessary. The simulation results verify the efficiency of the proposed approach.

Key words: CMAC neura l networ k, identifica tion, BDCS, var iable st ructure contr ol