Skip to Main Content
Localization is a key issue in multirobot formations, but it has not yet been sufficiently studied. In this paper, we propose a ceiling vision-based simultaneous localization and mapping (SLAM) methodology for solving the global localization problems in multirobot formations. First, an efficient data-association method is developed to achieve an optimistic feature match hypothesis quickly and accurately. Then, the relative poses among the robots are calculated utilizing a match-based approach, for local localization. To achieve the goal of global localization, three strategies are proposed. The first strategy is to globally localize one robot only (i.e., leader) and then localize the others based on relative poses among the robots. The second strategy is that each robot globally localizes itself by implementing SLAM individually. The third strategy is to utilize a common SLAM server, which may be installed on one of the robots, to globally localize all the robots simultaneously, based on a shared global map. Experiments are finally performed on a group of mobile robots to demonstrate the effectiveness of the proposed approaches.