Hisashi Tamaki

About

Hisashi Tamaki is a robotics and control systems researcher whose work spans two complementary domains: spherical rolling robot design and reinforcement learning for autonomous systems. He is best known for his pioneering contributions to gyro-driven spherical robots, having developed both the theoretical foundations and physical prototypes of a novel mechanism in which an internal gyroscope generates the driving torque needed for controlled locomotion. His 2006 paper on position and attitude control of this system has garnered 58 citations, establishing it as a landmark reference in the field, while subsequent work on dynamic modeling (2015) and experimental prototype development (2012) further solidified his expertise in this challenging area of nonholonomic robotics. Beyond robotics hardware, Tamaki has pursued intelligent control through reinforcement learning, investigating how agents can efficiently navigate partially observable Markov decision processes with high-dimensional, continuous state and action spaces. His concept of "state space filtering" addresses one of the most practical barriers to deploying reinforcement learning in real-world systems — the curse of dimensionality. Together, these research threads reflect a cohesive vision of building robots that are not only mechanically innovative but also capable of adaptive, autonomous decision-making.

Research Focus

Key Achievements

Top Papers

1. 1
   Position and attitude control of a spherical rolling robot equipped with a gyro

   58 citations · 2006
2. 2
   Dynamic Modeling and Controller Design for a Spherical Rolling Robot Equipped With a Gyro

   24 citations · 2015
3. 3
   Development of a spherical rolling robot equipped with a gyro

   12 citations · 2012
4. 4
   Developing reinforcement learning for adaptive co-construction of continuous high-dimensional state and action spaces

   6 citations · 2012
5. 5
   A State Space Filter for Reinforcement Learning in POMIDPs - Application to a Continuous State Space

   4 citations · 2006
6. 6
   A State Space Filter for Reinforcement Learning in Partially Observable Markov Decision Processes

   2 citations · 2009
7. 7
   A State Space Filter for Reinforcement Learning<br>- Concept and a Design -

   2 citations · 2006

Key Collaborators

Contact & Links