Anytime Planning for Decentralized Multirobot Active Information Gathering

Abstract

This letter considers the problem of reducing uncertainty about a physical process of interest by designing sensing trajectories for a team of robots. This active information gathering problem has applications in environmental monitoring, search and rescue, and security and surveillance. Our previous work developed a search-based planning method for information gathering which prunes uninformative trajectories from the search space while providing suboptimality guarantees and decentralizes the planning across multiple robots via coordinate descent. The novelty of this letter is to demonstrate the practical feasibility of these algorithms in a target tracking scenario featuring three collaborating UAVs and five mobile targets. To achieve this, we relax the previous requirement of having centralized estimation by performing distributed information filtering. We then develop an anytime planning algorithm that progressively reduces the suboptimality of the information gathering plans while respecting real-time constraints. These contributions enable robust and scalable information gathering using a team of agile robots that adapt their cooperation to timing constraints and ad hoc communication without the need for external or centralized computation.