Fixed-time composite DREM neural learning control for nonlinear strict-feedback systems

A new neural control learning method is proposed for a class of uncertain nonlinear strict feedback systems. Firstly, to approximate the unknown function in fixed-time, an enhanced neural network weight adaptive law (NNAL) is proposed, which uses the dynamic regressor extension and mixing (DREM) technique and the composite learning (CL) method. Notably, the proposed NNAL ensures each weight converges individually without the need for persistent excitation (PE) conditions. Secondly, a new adaptive practical fixed-time control strategy is devised by utilizing the dynamic surface control algorithm and fixed-time stability theory. Theoretical analysis proved that the closed-loop system manifests practical fixed-time stability under the proposed control strategy. Finally, simulation results are given to demonstrate the feasibility and effectiveness of the proposed algorithm.