Xiaofeng Lin

I am a second-year Systems Engineering Ph.D. student at Boston University where I work on reinforcement learning under the supervision of Prof. Xuezhou Zhang.

I earned my Master of Science in Robotics from University of Michigan, Ann Arbor, where I was fortunate to be advised by Prof. Vasileios Tzoumas. During my master, I worked on online learnig, submodular maximization and multi-agent systems.

I earned my B.S.E in Engineering Mechanics from Tianjin University.

Resume / Email / Google Scholar / Github / X

Research

I'm broadly interested in robotics, reinforcement learning, online learning. I aspire to enable robots to make better decisions in unstructured environments.

	Efficient Reinforcement Learning in Probabilistic Reward Machines Xiaofeng Lin, Xuezhou Zhang arXiv This paper studies how to explore given the knowledge of Probabilistic Reward Machines (PRMs) and extends reward-free framework to generic non-Markovian rewards setting.
	A Real-to-Sim-to-Real Approach for Vision-Based Autonomous MAV-Catching-MAV Zian Ning, Yin Zhang, Xiaofeng Lin, Shiyu Zhao Unmanned Systems, 2024. World Scientific / PDF This paper studies the task of vision-based MAV-catching-MAV, where a catcher MAV can detect, localize, and pursue a target MAV autonomously. This paper proposes a real-to-sim-to-real approach and sucessfully implements a fully autonomous vision-based MAV-catching-MAV system.
	Leveraging Untrustworthy Commands for Multi-Robot Coordination in Unpredictable Environments: A Bandit Submodular Maximization Approach Zirui Xu, Xiaofeng Lin, Vasileios Tzoumas American Control Conference (ACC), 2024. arXiv / code The algorithm leverages a meta-algorithm to learn whether the robots should follow untrustworthy commands or a recently developed submodular coordination algorithm, Bandit Sequential Greedy (BSG), which has performance guarantees. The algorithm asymptotically can achieve the better performance out of the commands and the BSG algorithm.
	Bandit Submodular Maximization for Multi-Robot Coordination in Unpredictable and Partially Observable Environments Zirui Xu, Xiaofeng Lin, Vasileios Tzoumas Robotics: Science and Systems (RSS), 2023. arXiv / code / simulation videos / presentation The algorithm generalizes the seminal Sequential Greedy algorithm by Fisher et al. to the bandit setting, by leveraging submodularity and algorithms for the problem of tracking the best action. We validate our algorithm in simulated scenarios of multi-target tracking.

Thanks for Jon Barron for this amazing template.