Drew Bagnell
Papers
5
Total Citations
198
H-Index
5
About
Drew Bagnell is a leading roboticist whose research spans autonomous manipulation, multi-robot coordination, and decision-making under uncertainty. His work is distinguished by its rigorous mathematical foundations and practical impact on field robotics. Bagnell made major contributions to modeling planar sliding mechanics, introducing a convex polynomial framework that elegantly captures the complex force-motion mapping for robotic pushing and manipulation—work that has garnered 59 citations and enabled more reliable object interaction. He also pioneered market-based approaches for multi-robot systems, showing how to learn opportunity costs to overcome human cognitive and communication limitations in remote exploration, a contribution cited 47 times. In Bayesian active learning, Bagnell developed near-optimal strategies for sequential decision-making that prioritize information gathering for effective decisions rather than pure uncertainty reduction (43 citations). His work on perceiving and exploiting object affordances for autonomous pile manipulation (33 citations) addresses fundamental challenges in unstructured environments, while his efficient optimization of control libraries (16 citations) provides practical solutions for high-dimensional control problems. Bagnell’s research consistently bridges theory and application, advancing both the mathematical foundations and real-world capabilities of autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Learning Opportunity Costs in Multi-Robot Market Based Planners47 citations · 2006
- 3Near Optimal Bayesian Active Learning for Decision Making43 citations · 2014
- 4
- 5Efficient Optimization of Control Libraries16 citations · 2011