The Modeling of Motion Control for Wheeled Mobile Robot Based on PSO-MLS Method

Abstract

The present paper describes a PSO-MLS modeling method for the motion control of wheeled mobile robot. The proposed method is based on Particle Swarm Optimization (PSO) in combination with Moving Least Squares (MLS). The PSO method is used for parameter setting in MLS method, which significantly influences the fitting accuracy. However, its use in the motion control of wheeled mobile robot has not been yet widely explored. So in this paper, the PSO-MLS method is used for the modeling of motion control of wheeled mobile robot, which is based on the local approximation method. Main advantages of this method are the PSO method used in the optimization of parameter setting and the local approximation used in MLS method. Application in the motion control of wheeled mobile robot based on PSO-MLS method is performed in this paper, and the performance of it is verified by the comparison with conventional least squares (LS) method in data fitting. Finally, the main conclusions of this study are exposed.