An Adaptive Terminal Sliding Mode Control Scheme via Neural Network Approach for Path-following Control of Uncertain Nonlinear Systems

In this paper, we design a neural network-based non-singular fast terminal sliding mode control (NFTSMC) for path tracking control of uncertain nonlinear systems (UNSs). The major features of the proposed control algorithm are to combine merits of feed-forward neural network (FFNN) and NFTSMC, such as approximation ability, rapid convergence rate, along with strong properties to external disturbances and uncertain dynamics. A controller is derived from the above combination for high control performance and fast convergence of state variables in presence of those unknown components. Besides, a fast-reaching control law (FRCL) is also applied to obtain a faster convergence time. The new contribution of the proposed method is to propose a PI-nonsingular fast terminal sliding mode surface (PI-NFTSMS) and to use a non-singular fast terminal sliding mode-based-error filter into FFNN. The contribution of the proposed method is a PI-NFTSMS and use of a non-singular fast terminal sliding mode-based-error filter into FFNN. With this approach, the lumped uncertain dynamics (LUDs) have been quickly and fully offset. Accordingly, the proposed controller can obtain a faster convergence time, higher tracking accuracy, lesser chattering than classical NFTSMC, and avoid the glitch of state variables in reaching the sliding manifold. Consequently, the stabilization of the whole closed loop control system is then guaranteed with strong control performance. The correctness of the stability is also proven based on Lyapunov analysis. Control performance obtained from the simulation example for an uncertain 3-DOF robot and experiment for an uncertain magnetic levitation system (MLS) has verified the effectiveness and practicality of the method.

Automated summary

A neural network-based non-singular fast terminal sliding mode control method is proposed for path tracking control of uncertain nonlinear systems, achieving faster convergence time, higher tracking accuracy, and less chattering with strong control performance for the entire closed-loop control system.