UAV and UGV Collaboration for Active Ground Feature Search and Localization

Abstract

This paper describes the implementation of a decentralized architecture for autonomous teams of aerial (UAV) and ground (UGV) vehicles engaged in actively searching for and localizing ground features. We provide a theoretical framework based on an established approach to the underlying sensor fusion problem. This provides transparent integration of information from heterogeneous sources. The approach is extended to include an information theoretic utility measure that captures the task objective and robot interdependencies. A distributed solution mechanism is employed to determine information maximizing trajectories subject to the constraints of individual vehicle and sensor sub-systems. This architecture enables the benefit of the complementary aerial and ground based vehicle and sensor capabilities to be realized. The approach is applied to multi-vehicle missions involving searching for and tracking multiple ground targets. Results for a team of fixed wing UAVs and all terrain UGVs equipped with vision sensors are presented.