Finite-Time Synergetic Control of Mechanical System Based on Model-Free Friction Compensation

Abstract

In this paper, a finite-time synergetic control scheme is proposed for a class of mechanical systems with friction compensation problem. Since the nonlinear parts in the friction model is difficult to be known in prior accurately, two neural networks are employed to approximate the static function and dynamic function of the friction model, respectively. Then, a finite-time synergetic controller is developed by combining synergetic control theory and nonsingular terminal sliding mode technique. The proposed control scheme can guarantee the finite-time convergence of tracking errors and avoid the singularity problem. Simulation results are provided to verify the effectiveness of the proposed scheme.