Abstract :
Researchers have developed a map-merging technique for multi-robot systems (MRS). Their method can produce an accurate collective map from individual maps built by multiple robots, which allows for more accurate and efficient path planning. When multi-robot systems are implemented in unknown environments, they need a collective map which describes the robots´ surroundings. Merging produces this collective map by fusing the individual maps of each robot; this is trivial if the robots know their initial relative positions, because the map transformation matrix (MTM) is easily derived. However, if the robots do not know their initial relative positions, map merging becomes challenging because the MRS needs additional information to acquire the MTM. Overcoming this limitation was the goal of the researchers, which would then afford greater utility for MRS.
