Non-homogeneous Multi-robot Collaboration for Environment Mapping and Inference

Multi-robot collaboration is the subject of a great deal of research currently. This interest within the field of robotics is prominent because multiple robots can often complete complex tasks faster and more efficiently than a single robot. The majority of the research being done today is related to homogeneous robotic communities, wherein each robot has the same equipment and limitations. The scenario becomes more challenging by introducing non-homogenous teams of robots. In this paper, we explore a non-homogeneous robotic community where one robot (acting as master) is designated to map an area while a different robot (acting as slave) infers from that map. Such a system is demonstrated to be reliable within a small and primarily static environment using two mobile robots equipped with RGB-D cameras and planar LiDAR sensors. This type of role assignment allows for more specialized equipment for both robots as well as greater optimization of the robots' tasks.