Indoor Environment Mapping of Mobile Robot Based on Efficient DSM-PF Method

A mobile robot uses an odometer system for localization, and uses a lidar sensor to obtain environmental information, to complete the process of Simultaneous Localization and Mapping (SLAM). This paper proposes an efficient Double Simultaneous Majorization Particle Filter (DSM-PF) algorithm, which is based on the majorization of the Rao-Blackwellized Particle Filtering (RBPF) method. The main purpose of the algorithm is to improve the accuracy of particle pose and maintain the weight of particles, to improve the quality of the results. For the motion distortion generated by the mobile robot, a particle pose majorization algorithm is proposed to improve the localization accuracy of the robot while in motion. At the same time, a weight majorization algorithm is proposed to maintain the weight of each particle, slow down the degradation of particles and improve the accuracy of the map carried by particles. In addition, for different particles that require resampling, an adaptive hierarchical resampling method is performed to maintain particles with higher weights. In this paper, the feasibility of the proposed algorithm is verified in different indoor environments. The experimental results show that the algorithm can slow down the degradation of particles and prevent the number of particles from declining. What is more, the experimental results of particle resampling show that the proposed algorithm has better mapping ability.

Automated summary

This paper proposes an efficient Double Simultaneous Majorization Particle Filter algorithm for localization and mapping of a mobile robot. By using pose majorization and weight majorization algorithms, the accuracy of robot localization and particle-carried map is improved. The proposed adaptive hierarchical resampling method maintains particles with higher weights.