Parameter Optimization Analysis of Gmapping Algorithm Based on Improved RBPF Particle Filter

Abstract

Abstract Aiming at the problem of inaccurate mapping caused by the robot hardware configuration that cannot meet the needs when the Gmapping algorithm is used for indoor mapping, a method for optimizing algorithm parameters is proposed. First, the principle of the Gmapping algorithm based on improved Rao-Blackwellised particle filtering is analyzed. By using the open source robot operating system (ROS) to run the Gmapping algorithm on a mobile robot, an indoor map is constructed. Then, by configuring different parameters, the accuracy of map construction is optimized. The experimental results show that by optimizing the algorithm parameters in an indoor environment, the algorithm can better adapt to the hardware configuration and improve the quality of the map.