Papers
14
Total Citations
268
H-Index
7
About
Keisuke Okumura is a leading researcher in multi-agent pathfinding (MAPF), a core challenge in multi-robot coordination for applications like automated warehouses and swarm robotics. His most influential work, the 2022 paper on "Priority inheritance with backtracking" (110 citations), introduced a powerful iterative refinement technique that dramatically improves solution quality in large-scale MAPF problems. He followed this with LaCAM (2023, 55 citations), a complete, search-based algorithm that achieves remarkably quick solutions for complex multi-agent scenarios. Okumura has also pioneered the study of "time-independent" path planning, enabling robots to execute plans without strict temporal synchronization—a critical advance for real-world deployment. His work on combining target assignment with path planning (2023, 22 citations) addresses the challenging unlabeled MAPF problem, where agents must both decide which goal to reach and how to get there. Beyond algorithmic theory, Okumura has applied robotics to nuclear decommissioning, including work on remotely operated vehicles for fuel debris detection at the Fukushima Daiichi site. His research consistently bridges rigorous algorithmic foundations with practical, real-time constraints, making him a key figure in advancing the state of the art in multi-robot coordination.
Research Focus
Key Achievements
Top Papers
- 1Priority inheritance with backtracking for iterative multi-agent path finding110 citations · 2022
- 2LaCAM: Search-Based Algorithm for Quick Multi-Agent Pathfinding55 citations · 2023
- 3
- 4Iterative Refinement for Real-Time Multi-Robot Path Planning19 citations · 2021
- 5
- 6Offline Time-Independent Multiagent Path Planning12 citations · 2023
- 7Quick Multi-Robot Motion Planning by Combining Sampling and Search10 citations · 2023
- 8Offline Time-Independent Multi-Agent Path Planning6 citations · 2022
- 9Amoeba Exploration: Coordinated Exploration with Distributed Robots5 citations · 2018
- 10