Simultaneous Location and Map Construction Based on RBPF-SLAM Algorithm

Abstract

Solving the problem of simultaneous location and map construction (SLAM) is the core of robot autonomous navigation. At present, most algorithms usually only consider the odometer information of mobile robot, therefor, there exists some problems of increasing computation amount caused by a large number of sampling particles and the complexity. So, this paper developed a scheme of map construction and positioning based on RBPF-SLAM algorithm. The hardware and software platform is set up on the ROS, and it merge the odometer information of the robot into the distance information collected by the laser sensor, which effectively reduces the number of required particles and the uncertainty of the robot's pose estimates in filter prediction phase. Through the experimental, get the conclusion: simultaneous positioning based on RBPF-SLAM algorithm and map construction system can create high-precision online raster map in real time, and it is more consistent with the actual map.