A hybrid positioning method for multi-robot simultaneous location and mapping

Navigating in a GPS-disabled environment is one of the basic tasks of autonomous mobile robots. Our research focuses on the topic that using a multi-robot simultaneous location and mapping (SLAM) to detect the surrounding environment. A new multi-robot SLAM framework is proposed. Firstly, each mobile robot is positioned with six degrees of freedom using cooperative identification, and then the data correlation between sub-maps is used to optimize the result of mutual positioning. Finally, sub-maps are merged for a consistent global map. Experimental results show that this method can achieve global map reconstruction, and can get more accurate results than single-robot SLAM.