معرفی
Dr. Keisuke Okumura is a Visiting Professor at the University of Cambridge (as a JSPS overseas fellow) and a researcher at the National Institute of Advanced Industrial Science and Technology (AIST) in Japan. He holds a Ph.D. in Artificial Intelligence from Tokyo Institute of Technology (2023), supervised by Prof. Xavier Défago. His research focuses on multi-agent systems, particularly in path planning and coordination for autonomous agents, leveraging techniques from planning, search algorithms, machine learning, and distributed computing.
Research Interests: His work addresses challenges in multi-agent pathfinding (MAPF), real-time trajectory optimization, and scalable solutions for large-scale robotic systems. He explores hybrid approaches combining sampling-based methods with search algorithms, as well as fault-tolerant offline planning strategies. Key areas include dynamics-aware trajectory deformation, time-independent execution frameworks, and machine learning integration for enhanced pathfinding efficiency.
Publications: His recent work emphasizes advancements in algorithms like LaCAM and D4orm, targeting real-time performance and scalability. Themes include optimizing multi-robot motion, integrating diffusion models for trajectory refinement, and addressing challenges in large-scale automation. His articles reflect a blend of theoretical contributions and practical implementations in robotics and autonomous systems.
Affiliations: Alongside his visiting role at Cambridge’s Department of Computer Science and Technology, Dr. Okumura contributes to AIST’s research initiatives. His collaborations span academic and industrial partners, focusing on applied robotics and intelligent systems.
