Mitsuo Kawato
The University of Osaka, Japan Science and Technology Agency, Advanced Telecommunications Research Institute International, Research Organization of Information and Systems, Brain (Germany), University of Southern California, RF Laboratories (United States), Oxford Centre for Computational Neuroscience, Osaka City University, University of Calgary, Nara Institute of Science and Technology
Papers
60
Total Citations
6,352
H-Index
28
About
Mitsuo Kawato is a pioneering computational neuroscientist whose work has fundamentally shaped our understanding of how the brain controls movement. Based at the Advanced Telecommunications Research Institute International (ATR) in Japan, Kawato has dedicated his career to unraveling the computational principles underlying voluntary motor control, bridging neuroscience, robotics, and machine learning. Kawato's most celebrated contribution is his hierarchical neural network model for voluntary movement control, proposed in 1987 and now boasting over 1,575 citations, which elegantly framed how the central nervous system solves the complex computational challenges of transforming visual goals into precise muscular commands. Building on this, his feedback-error-learning framework provided a biologically plausible mechanism for motor adaptation that has profoundly influenced robot control systems. His influential 2001 work on impedance control and unstable dynamics (1,116 citations) demonstrated that the brain actively learns to stabilize unpredictable environments by optimizing arm stiffness — a discovery with deep implications for rehabilitation and prosthetics. Kawato has also pioneered the use of humanoid robots as scientific tools for studying human behavior, and his more recent investigations into physical human-human interaction reveal how individuals infer a partner's intentions to coordinate movement. Across more than three decades, his interdisciplinary vision has made him one of the most impactful figures in computational motor neuroscience.
Research Focus
Key Achievements
Top Papers
- 1A hierarchical neural-network model for control and learning of voluntary movement1,575 citations · 1987
- 2The central nervous system stabilizes unstable dynamics by learning optimal impedance1,116 citations · 2001
- 3
- 4Learning from demonstration and adaptation of biped locomotion400 citations · 2004
- 5
- 6
- 7Using humanoid robots to study human behavior266 citations · 2000
- 8
- 9
- 10Impedance Control Balances Stability With Metabolically Costly Muscle Activation124 citations · 2004