Multi-UAV Collaborative Monocular SLAM Focusing on Data Sharing

Sharing data among Unmanned Aerial Vehicles (UAVs) is one of key issues in the field of multiple-robot SLAM. In this paper, aiming at problems of sharing data between UAVs during tracking lost and map fusion, we propose a robust, focusing on Date Sharing Multi-UAV visual SLAM (DSM-SLAM) based on centralized architecture. In addition, we present a two-step relocalization method based on sharing local maps, in order to support the UAV in using the data from other UAVs which have gone there before when the tracking is lost. Furthermore, we put forward a map fusion method based on hierarchical clustering to dynamically and adaptively select the order of map fusion that is more beneficial to data sharing between drones. Experimental results on popular public datasets demonstrate the feasibility and effectiveness of the system.