Papers
11
Total Citations
66
H-Index
5
About
Brian Carse is a researcher whose work sits at the intersection of computational intelligence, autonomous robotics, and human-robot interaction. His career has been defined by a sustained investigation into how fuzzy logic, reinforcement learning, and evolutionary computation can be harnessed to create adaptive, intelligent systems — particularly for mobile robot control. Carse's most influential contribution is his development and refinement of Fuzzy Q-Learning (FQL) with adaptive representations, exploring how reinforcement learning can be extended to handle large or continuous state spaces across domains from data mining to robotics. Alongside this, he has conducted extensive comparative studies of Michigan- and Pittsburgh-style Fuzzy Classifier Systems, systematically evaluating their effectiveness in enabling robots to autonomously acquire reactive behavioural competencies through evolutionary methods. His later work expanded into hierarchical fuzzy rule-based systems and human-robot interaction, including pioneering research on how humanoid robots functioning as avatars shape collaboration dynamics around interactive tabletops. This breadth reflects a researcher consistently pushing the boundaries of how intelligent machines learn, navigate, and cooperate. With citations spanning robotics, machine learning, and cognitive systems, Carse's body of work offers valuable foundational and applied insights for students and researchers working in autonomous systems and adaptive AI.
Research Focus
Key Achievements
Top Papers
- 1Fuzzy Q-Learning with an adaptive representation15 citations · 2008
- 2
- 3
- 4
- 5
- 6
- 7
- 8A topological map based navigation system for mobile robotics3 citations · 2002
- 9
- 10Further experiments in Fuzzy Classifier Systems for mobile robot control2 citations · 2003