A Design of Neural Network Direct Adaptive Tracking Control for a Class of Strict-Feedback Uncertain Nonlinear Systems

Authors

  • Hui Hu, Yang Li, Honghua Guo, Wei Yi, Yuebiao Wang

Abstract

The paper designs a control algorithm for the strict-feedback uncertain nonlinear systems based on neural network direct adaptive tracking. In order to avoid the backstepping design, the above nonlinear system is transformed into the normal affine nonlinear system. There aren’t state observers and robustifying control terms in the system. In adaptive law only the output error is used and the parameters are optimized by a gradient descent algorithm to reduce a cost function of the error between controllers of the actual ideal and the neural network. Finally, the convergence and boundedness of tracking error and adaptive parameters are proved by Lyapunov approach. Some simulation results verifies the effectiveness of the control algorithm.

Published

2020-03-31

Issue

Section

Articles