Kazuhiro Ohkura
Papers
62
Total Citations
430
H-Index
11
About
Kazuhiro Ohkura is a leading researcher in swarm robotics and multi-robot systems, whose work has significantly advanced our understanding of how autonomous robot collectives can generate intelligent, cooperative behavior without centralized control. His research spans evolutionary computation, deep reinforcement learning, and bio-inspired algorithms, applying these techniques to some of the most challenging problems in collective robotics. Ohkura's early contributions established foundational methods for autonomous role assignment and fault-tolerant homogeneous robot teams, demonstrating that decentralized systems could reliably coordinate complex tasks. He later pioneered the use of topology- and weight-evolving neural networks to coordinate adaptive swarm behavior, and explored response threshold models drawn from insect societies to achieve autonomous task allocation. His more recent work embraces deep reinforcement learning — including deep Q-learning and experience-sharing frameworks — to develop end-to-end control policies for robotic swarms, reflecting a forward-looking integration of modern AI into collective robotics. His 2022 study on Lévy flight-based exploration further demonstrates his commitment to biologically inspired search strategies for large-scale environments. With his most-cited works accumulating tens to hundreds of citations collectively, Ohkura's research offers students and practitioners an invaluable roadmap for designing scalable, adaptive, and resilient autonomous swarm systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6A homogeneous mobile robot team that is fault-tolerant18 citations · 2006
- 7Autonomous Role Assignment in a Homogeneous Multi-Robot System15 citations · 2005
- 8
- 9
- 10