Kenneth Bogert
University of Georgia, University of North Carolina at Asheville
Papers
3
Total Citations
53
H-Index
3
About
Kenneth Bogert is a robotics researcher whose work sits at the intersection of machine learning, multi-robot systems, and inverse reinforcement learning (IRL). His research tackles a particularly challenging real-world problem: how can a robot learn the behavioral intentions and reward structures of other robots when it can only partially observe them? This problem, known as multi-robot inverse reinforcement learning under occlusion, forms the cornerstone of his academic contributions. Bogert's most influential work, "Multi-Robot Inverse Reinforcement Learning under Occlusion with Interactions" (2015, 26 citations), introduced a framework enabling a subject robot to infer the behaviors of multiple mobile agents even when portions of their trajectories are hidden from view. His subsequent research refined this approach by incorporating estimation of unknown state transitions—a critical relaxation of prior assumptions that makes IRL far more practical for real-world deployment. His 2018 paper expanded these ideas further, accumulating 16 additional citations. Collectively, Bogert's contributions advance the field of autonomous robot learning by addressing fundamental limitations in observability and agent modeling. His work is particularly valuable for researchers developing robots that must understand and safely navigate environments shared with other autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Multi-Robot Inverse Reinforcement Learning under Occlusion with Interactions26 citations · 2015
- 2
- 3