Papers
24
Total Citations
894
H-Index
15
About
Michael Bowling is a prominent researcher whose work spans robotics, multiagent systems, and machine learning, with particular strengths in autonomous robot control and multi-robot coordination. His most influential contribution, "Automatic Gait Optimization with Gaussian Process Regression" (2007, 247 citations), demonstrated how robots could autonomously optimize locomotion patterns without relying on brittle local search methods — a significant advance for both quadrupedal and bipedal robotics. Alongside this, Bowling made foundational contributions to multi-robot teamwork in adversarial environments, most notably through the STP (Skills, Tactics, and Plays) framework (2005, 110 citations), which provided a principled architecture for coordinating robot soccer teams under time pressure. His work on multiagent learning explored how agents can adapt strategically when teammates and opponents have varying limitations, addressing one of the field's core challenges. Bowling also advanced dimensionality reduction for sequential decision-making through Action Respecting Embedding (2005, 53 citations), connecting representation learning to robot perception and localization. His repeated contributions to RoboCup robot soccer — including the CMUnited-98 championship — demonstrate both theoretical depth and real-world impact, making him a defining figure in intelligent autonomous systems research.
Research Focus
Key Achievements
Top Papers
- 1Automatic gait optimization with Gaussian process regression247 citations · 2007
- 2
- 3
- 4Action respecting embedding53 citations · 2005
- 5
- 6Multi-robot team response to a multi-robot opponent team49 citations · 2004
- 7Multiagent learning in the presence of agents with limitations46 citations · 2003
- 8The CMUnited-98 Champion Small-Robot Team43 citations · 1999
- 9Simultaneous adversarial multi-robot learning39 citations · 2003
- 10Subjective Localization with Action Respecting Embedding35 citations · 2007