Radial Basis Function Neural Network and Genetic Algorithm in trajectory tracking control of the Omni-Directional mobile robot

Abstract

The paper’s aim is to combine a radial basis function (RBF) neural network and genetic algorithm in trajectory tracking control of the Omni-directional mobile robot.The radial basis function neural network is considered as an adaptive controller in the adaptive sliding mode control law. This is self-learning, selforganizing, and adaptive, possess fast training speed, and global convergenceneural network.The genetic algorithm is used to optimize the number of neurons in the hidden layer, centers, widths and initial weights of the radial basis function neural network. After optimizing, the radial basis function neural network is online trained by Quasi - Newton algorithm. The simulation results in MATLAB/SIMULINK show that the proposed controller is efficient, the response of the Omni-directional mobile robot in simulation model converge to reach the trajectory with steady-state error is about0.001±0.0005(m) , and the overshoot is about 0.15 ±0.05(%)