Global Loop Closure Detection of Multi-Robot Based on Particle Filter

Abstract

A loop closure detection method based on the combination of CSM (correlative scan matching) and particle filter is proposed for the uncertainty of initial relative pose of the robot during the multi-robot SLAM(simultaneous localization and mapping). This method applies the concept of particle filter location to multi-robot loop closure detection, and use the matching of sub-maps to calculate the similarity score, which can quickly and effectively detect the loop closure state between multiple robots, thus it can providing a basis for fast and consistentjoint mapping of multi-robot.