Adaptive Finite Time Control for Wearable Exoskeletons Based on Ultra-local Model and Radial Basis Function Neural Network

This paper investigates an adaptive finite time control scheme for wearable exoskeletons to realize trajectory tracking control. The main feature of the proposed scheme is model-free in which no dynamic models are required except for the input and output data. Firstly, a second order ultra-local model is employed to replace the complex dynamic model of wearable exoskeleton as the controlled object, which is a novel designed model. In addition, a nonsingular fast terminal sliding surface is proposed to design the controller to guarantee the finite time convergence of tracking errors, and a radial basis function (RBF) neural network is developed to approximate the lumped disturbance in the ultra-local model. Then the stability of closed-loop system is proved by Lyapunov theory. Finally, to validate the proposed control scheme, virtual prototype is designed in SolidWorks and transferred to MATLAB, then visual simulation program is implemented based on SimMechanics. What’s more, reference trajectories are extracted from the measured data of DELSYS Electromyography (EMG) recording system. The effectiveness of proposed scheme is demonstrated by the simulation results.