Chris Gaskett
Papers
8
Total Citations
201
H-Index
7
About
Chris Gaskett is a pioneering researcher in reinforcement learning for robotics, with a career spanning mobile robots, humanoid systems, and autonomous underwater vehicles (AUVs). His work focuses on enabling robots to learn complex behaviors—such as visual servoing, reaching, and object recognition—without requiring explicit models or calibration. Gaskett’s most influential contribution is the application of Q-learning to continuous state-action spaces, demonstrated in his highly cited 2002 paper “Reinforcement learning for a vision based mobile robot” (45 citations), where a robot learned vision-based behaviors like wandering and servoing from scalar rewards alone. His related work “Q-Learning for Robot Control” (39 citations) further established this approach as a means to reduce programming effort. Gaskett also advanced underwater robotics, developing visually-guided AUVs for autonomous exploration and inspection (37 citations), and contributed to humanoid control with online learning of motor maps for reaching (24 citations). His research on support vector machines and Gabor kernels for active foveated vision (14 citations) highlights his versatility. With over 200 total citations, Gaskett’s work remains foundational for students and researchers in robot learning and autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Reinforcement learning for a vision based mobile robot45 citations · 2002
- 2Q-Learning for Robot Control39 citations · 2002
- 3Development of a visually-guided autonomous underwater vehicle37 citations · 2002
- 4Online learning of a motor map for humanoid robot reaching24 citations · 2003
- 5Reinforcement learning for visual servoing of a mobile robot24 citations · 2000
- 6
- 7Autonomous Guidance and Control for an Underwater Robotic Vehicle13 citations · 1999
- 8