Robot Mapping and Navigation System Based on Multi-sensor Fusion

Abstract

To deal with the problem of low accuracy of two-dimensional grid map and obstacle avoidance in dynamic environment of mobile robot with single 2D LiDAR, based on Turtlebot3 robot and Jetson TX2 development board, combining Bayesian estimation data fusion method, improved A* algorithm and dynamic window algorithm, this paper develops a robot mapping and navigation system. This system can fuse the data of LiDAR and RGB-D camera to enhance the mapping exactness, and realize dynamic environment navigation. Experimental results in real environment show that the mobile robot's exactness of two-dimensional grid map is enhanced, the path planning efficiency is improved, and the terminal positioning error is competent for the mobile robot's mapping and navigation tasks.