Hitoshi Iima

About

Hitoshi Iima is a researcher specializing in swarm intelligence, reinforcement learning, and multi-robot systems, with a particular focus on developing novel algorithms that enhance the efficiency and reliability of autonomous agent coordination. His most significant contribution is the development of **swarm reinforcement learning methods**, a framework in which multiple agent-environment pairs learn collaboratively by exchanging information, accelerating convergence and improving policy quality beyond what individual reinforcement learning achieves alone. Iima's foundational 2011 work introduced swarm reinforcement learning for continuous state-action spaces, garnering 9 citations, and laid the groundwork for his subsequent and most impactful research. His 2013 paper applying these methods to multi-robot formation problems — where robots must autonomously assign themselves to distinct goal positions and optimize their individual routes — earned 15 citations. His refined 2015 follow-up, which specifically addressed certainty of learning in this challenging coordination task, became his most cited work with 21 citations. Collectively, Iima's research addresses a core challenge in robotics and AI: enabling multiple autonomous agents to coordinate effectively without centralized control. His work offers practical and theoretical advances relevant to robotics, distributed artificial intelligence, and autonomous systems research.

Research Focus

Key Achievements

Top Papers

1. 1
   Swarm reinforcement learning methods improving certainty of learning for a multi-robot formation problem

   21 citations · 2015
2. 2
   Swarm Reinforcement Learning Method for a Multi-robot Formation Problem

   15 citations · 2013
3. 3
   Swarm reinforcement learning methods for problems with continuous state-action space

   9 citations · 2011

Key Collaborators

Contact & Links