Shoichi Noda

About

Shoichi Noda is a pioneering researcher in the field of robot learning, with a particular focus on vision-based reinforcement learning and autonomous behavior acquisition for mobile robots. Working primarily through the 1990s and 2000s, Noda made foundational contributions to the challenge of enabling real robots to learn purposive behaviors directly from visual input, without requiring explicit knowledge of their environment's geometry or dynamics. His most celebrated work, "Purposive Behavior Acquisition for a Real Robot by Vision-Based Reinforcement Learning" (1996), has accumulated 277 citations and demonstrated that Q-learning could be applied to physical robots performing goal-directed tasks such as shooting a ball into a target. Noda also tackled the critical problem of state space construction, proposing action-based approaches to resolve the interdependence between state and action representations — a persistent challenge in robot learning. His subsequent research extended these ideas to multi-behavior coordination and adversarial scenarios, contributing directly to the RoboCup robotics competition initiative. With a body of work spanning perception, learning, and coordination in autonomous systems, Noda's research laid important groundwork for modern embodied AI and autonomous robot development.

Research Focus

Key Achievements

Top Papers

1. 1
   Purposive behavior acquisition for a real robot by vision-based reinforcement learning

   277 citations · 1996
2. 2
   Action-based sensor space categorization for robot learning

   72 citations · 2002
3. 3
   Coordination of multiple behaviors acquired by a vision-based reinforcement learning

   71 citations · 2002
4. 4
   Vision-based reinforcement learning for purposive behavior acquisition

   71 citations · 2002
5. 5
   Vision-based behavior acquisition for a shooting robot by using a reinforcement learning

   33 citations · 2002
6. 6
   Purposive Behavior Acquisition for a Real Robot by Vision-Based Reinforcement Learning

   27 citations · 1996
7. 7
   Action-Based State Space Construction for Robot Learning.

   23 citations · 1997
8. 8
   Learning in robotics. Purposive Behavior Acquisition for a Robot by Vision-Based Reinforcement Learning.

   21 citations · 1995
9. 9
   A Vision-Based Reinforcement Learning For Coordination Of Soccer Playing Behaviors

   9 citations · 2004
10. 10
    Vision-Based Reinforcement Learning for RoboCup : Towards Real Robot Competition

    7 citations · 2004

Key Collaborators

Contact & Links