Man Xie

Papers

Total Citations

H-Index

About

Man Xie is a robotics researcher whose work centers on structured policy learning for robot motion control, with a particular focus on acceleration-based systems. Their major contribution is the development of RMP2, a novel policy class that leverages the RMPflow multi-task control framework to enable composable, efficient robot learning. By structuring policies within this framework, Xie’s research bridges the gap between classical control theory and modern reinforcement learning, allowing robots to learn complex motion tasks more reliably and with greater sample efficiency. This work has garnered attention in the robotics community, with their most-cited paper accumulating 11 citations, reflecting its growing influence. Xie’s approach offers a principled way to incorporate domain knowledge into learned policies, making it especially valuable for real-world applications where safety and precision are paramount. Their research stands at the intersection of robot learning, control, and multi-task optimization, providing a foundation for more adaptable and capable robotic systems.