Composite learning control for strict feedback systems with neural network based on selective memory

This paper addresses the high‐precision control problem for nonlinear strict feedback systems with external time‐varying disturbances and proposes a novel composite learning control algorithm. Unlike previous research that only uses tracking errors for neural network updates, this paper prioritizes the accuracy of neural network learning. The article uses a selective memory recursive least squares algorithm to construct system information prediction errors, which are combined with tracking errors to update the neural network weights. A new composite learning control algorithm is developed to design dynamic surface control and neural network disturbance observers, which achieves high‐precision control of nonlinear strict feedback systems under external time‐varying disturbance conditions. Lyapunov's method demonstrates the stability of the closed‐loop system and the boundedness of errors. The simulation results show that the proposed control algorithm can effectively estimate system nonlinearity and suppress the impact of disturbances.