Model-free control using time delay estimation and fractional-order nonsingular fast terminal sliding mode for uncertain lower-limb exoskeleton

Abstract

A robotic exoskeleton is a nonlinear system, which is subjected to parametric uncertainties and external disturbances. Due to this reason, it is difficult to obtain the exact model of the system, and without knowledge of the system, it cannot be compensated accurately. In this study, time delay estimation (TDE)-based model-free fractional-order nonsingular fast terminal sliding mode control (MFF-TSM) is proposed for the lower-limb robotic exoskeleton in the existence of uncertainties and external disturbance. The main characteristic of the proposed scheme is that it controls the system without relying on the knowledge of exoskeleton dynamics. At first, the fractional-order (FO) with nonsingular fast terminal sliding mode control (NFTSM) is adopted to provide a precise trajectory tracking performance, fast finite-time speed of convergence, singularity-free and chatter-free control inputs. And then, the proposed controller employs TDE, to make the controller model independent, which directly estimates the uncertain exoskeleton dynamics with external disturbances. Later, asymptotical stability analysis of the overall system and finite-time convergence are investigated and ensured using Lyapunov theorem. Finally, the simulation results are conducted to validate the efficacy of the proposed control method.