Learning from adaptive neural control of SISO strict-feedback nonlinear systems

This paper emphasizes learning from adaptive neural control of SISO strict-feedback nonlinear systems with completely unknown system dynamics. The SISO strict-feedback nonlinear systems are transformed into the affine nonlinear systems. Then, an adaptive neural controller is designed, which achieves tracking convergence of the plant states to the recurrent reference states, so that the partial persistent excitation (PE) condition is satisfied. Consequently, exponential stability of the closed-loop error system which is in the form of a class of linear time-varying (LTV) systems is confirmed in theory, convergence of partial neural weights to their optimal values is guaranteed, and locally-accurate NN approximation of the unknown closed-loop system dynamics is achieved within a local region along the recurrent tracking trajectory. The learned knowledge stored as constant neural weights can be used to improve the control performance, and can also be reused in the same or similar control task. Finally, Simulation results show the effectiveness of the proposed approach.