Concurrent Learning Adaptive Command Filtered Backstepping Control for High-Order Strict-Feedback Systems

Abstract

This paper is devoted to trajectory tracking control for the second-and high-order strict-feedback systems (SFSs) with accurate parameter estimations. By skillfully fusing the techniques of concurrent learning (CL), adaptive command filtered backstepping (ACFB) and high-order fully-actuated (HOFA) system approach, a novel CL-based high-order ACFB (CL-HOACFB) controller is constructed. The typical feature of the proposed controller is that it directly utilizes the HOFA feature to design controller without turning the original second-and high-order strict-feedback systems into the first-order state-space approach to reduce backstepping steps, and circumvents the complexity arising due to repeatedly differentiating the virtual control. Remarkably, the proposed controller provides the ability to identify unknown parameters by only checking the linear independence of the recorded data, which largely relaxes the requirement of persistent excitation (PE) needed in the previous approaches. Theoretically, it is demonstrated that the tracking error can be adjusted to be as small as desired by tuning predetermined parameters. Finally, a benchmark application in the electromechanical system is given to illustrate the validity and potential of the proposed scheme.