A Survey of Implementation of Multi-Robot Simultaneous Localization and Mapping

Abstract

For any robot to effectively traverse its environment, it requires a map and its location along with the orientation within that map. Often a map may not be available if the robot is presented with an unfamiliar or unknown environment. This requires the robot to construct the map as it localizes itself and navigates. This problem of concurrently building a map and localizing a robot in that environment is defined as Simultaneous Localization and Mapping (SLAM). Many SLAM applications already exist for single robot such as navigating the unmanned mines and exploring the sites with natural calamities. With the advent of swarming robots that must interact with each other, extensive research is being conducted for the extension of SLAM problem to multiple robots known as Multi-Robot SLAM (MRSLAM). In the MRSLAM environment, the efficiency is improved, and time constraints are reduced, but it's implementation is restricted due to constraints such as communication bandwidth, memory requirements and problems faced during map merging and co-ordinate transformation. This paper represents many different approaches that are being used for an implementation of MRSLAM.