A sonar data based particle filtering localization method for mobile robot

A global localization method is proposed for mobile robot. This algorithm is based on Bayesian filtering theory and MarKov process. A set of particles is utilized to describe the distribution of robot localization. The set is updated according to the map information and sonar data step by step. In this way the particles converge to the true localization rapidly. In this paper, the map is recorded as set of sections of line, which takes little storage space and benefit calculation. Simulation results show the effectiveness of this algorithm.