PSO-Based Adaptive Neural Control for Trajectory Tracking of a Mobile Robot

Abstract

This paper proposes a PSO (Particle Swarm Optimization)-based adaptive neural control scheme for trajectory tracking of a nonholonomic wheeled mobile robot in the presence of uncertainties and external disturbances. The proposed controller does not require the exact kinematic and dynamic model parameters for a mobile robot. The proposed controller is adjusted according to the adaptation laws. The number of Gaussian basis functions used in a RBFNN (Radial Basis Function Neural Network) and the control gains and the adaptive gains are automatically tuned online by a PSO algorithm. The stability of the overall control system is guaranteed by the Lyapunov function analysis. A comparative simulation for the proposed adaptive neural controller with/without PSO is performed. The simulation results verify the validity and robustness of the proposed control scheme.