Adaptive back-stepping control based on recurrent neural network for BLDCM EMA

The performances of brushless DC motors (BLDCM) will directly restrict the performances of electro-mechanical actuator (EMA). To improve the performances of the BLDCM, a dynamic mathematical model with taking into the problem of parameter variation and nonlinear dynamic friction compensation has been established, a recurrent neural network (RNN) based adaptive back stepping control (RNABC) for BLDCM in electro-mechanical actuator (EMA) system was proposed. RNN which has self feed back of the hidden neurons ensures that the outputs of RNN contain the whole past information of the system even if the inputs of RNN are only the present states and inputs of the system. The RNABC system is comprised of a back stepping controller and a robust controller. The back stepping controller containing an RNN uncertainty observer is the principal controller, and the robust controller is designed to dispel the effect of approximation error introduced by the uncertainty observer. The adaptation laws of the adaptive back stepping approach are derived in the sense of the Lyapunov function, thus, the stability of the system can be guaranteed. Simulation results verify that the proposed RNABC can achieve favorable tracking performance for servo system, even regard to parameter variations and friction disturbance.