MEAM Seminar: “Multi-robot Coordination Via Graph-based Computation”

Multi-robot coordination and cooperation are critical behaviors that improve team performance and enable new tasks in application areas like autonomous construction, agriculture, and extended
operation in large unknown regions. This talk explores the utility of graph-based modeling and computational frameworks to improve the coordination and cooperation of multi-robot teams in a variety of problems. I will introduce a graph-based modeling and solution approach to multi-robot task allocation in complex multi-task missions where coordination and cooperation are explicitly required. I will discuss preliminary results, which show a vast improvement over the state of the art, and limitations of these results and my plans to address these limitations. I will also briefly introduce graph-based computation via Graph Neural Networks applied to the distributed multi-agent coverage control problem. These modeling approaches and algorithms bring performance improvements that further the state of the art by leveraging the fundamental graph structure present in some multi-robot problems.