Multi-L: A Novel Multi-Robot Cooperative Localization Method in Indoor Environment

Mobile robots often lose global localization in highly dynamic indoor environments. Enabling a mobile robot to determine where a catastrophic localization failure has occurred and how to recover from it is known as the kidnapped robot problem (KRP). However, due to the limitations of the view of sensors carried by individual robots, it is difficult for individual robots to recover from KRP. To address this issue, we can use the collaboration between multiple robots to increase the coverage of robot swarms and reduce the occurrence of localization errors. Here, we propose Multi-L, a novel multi-robot collaborative localization framework, Multi-L utilizes RSSI (Received Signal Strength Indicator) to determine the relative positions between robots when multiple robots communicate and combines the communication exchange of positional information to determine whether there is a significant error in the current localization and to help it recover. We have experimented and evaluated the Multi-L proposed in this paper in both simulated and real-world environments. The experimental results show that Multi-L greatly improves the probability of localization recovery when KRP occurs and can effectively help the robot reach its scheduled position.