Ziyuan Ma

Simon Fraser University, Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

115

H-Index

About

Ziyuan Ma is a leading researcher in multi-agent coordination, with a focus on developing scalable, decentralized solutions for large-scale robotic systems. His primary research areas include Multi-Agent Path Finding (MAPF), reinforcement learning, and heterogeneous task allocation. Ma’s most impactful contribution is his 2021 work on distributed heuristic MAPF with communication, which has garnered over 100 citations. This research addresses the fundamental challenge of achieving collision-free policies in partially observable environments by enabling agents to learn cooperative behaviors through decentralized reinforcement learning. He further advanced the field in 2023 by pioneering a novel approach that integrates graph neural networks with ant colony optimization algorithms to solve heterogeneous multi-agent task allocation problems—a critical optimization for drone swarms and multi-robot coordination. Ma’s work stands out for its practical emphasis on communication and learning in real-world, large-scale deployments, bridging the gap between theoretical algorithms and operational robotic systems. His contributions are essential reading for researchers and students working on autonomous systems, swarm robotics, and distributed AI.