Xiaobai Ma

University of Michigan–Ann Arbor, Stanford University

Papers

Total Citations

H-Index

About

Xiaobai Ma is a leading researcher in multi-agent systems, with a focus on autonomous aerial robotics and reinforcement learning. Their work addresses the fundamental challenge of enabling multiple intelligent agents to coordinate and navigate safely in complex, obstacle-rich environments. Ma's highly cited 2016 paper introduced a decentralized prioritized motion planning method for multiple autonomous UAVs in 3D polygonal obstacle environments, combining a prioritized A* algorithm for global planning with barrier functions for local coordination—a breakthrough that has garnered 30 citations and remains foundational for drone swarm navigation. More recently, Ma has advanced the field of multi-agent reinforcement learning (MARL) with their 2022 work on Recursive Reasoning Graphs, which tackles the critical problem of agents anticipating each other's influence during complex interactions. This innovative framework addresses a key limitation in existing MARL algorithms, enabling more sophisticated cooperative behaviors. Ma's research bridges theoretical rigor with practical applications in autonomous systems, making significant contributions to both motion planning and learning-based coordination that continue to shape how multiple robots operate safely and efficiently in shared spaces.