Papers
7
Total Citations
68
H-Index
3
About
Ellis Ratner’s research lies at the critical intersection of human-robot interaction, multi-agent decision-making, and robust planning under uncertainty. His work addresses a fundamental challenge: how can robots operate safely and efficiently alongside humans when human behavior is unpredictable and robot models are imperfect. Ratner’s most influential contribution is a robust control framework for human motion prediction, which uses reachability analysis to guarantee safety against every possible future human state—a paradigm shift from probabilistic approaches. This work has garnered 27 citations and is foundational for robots in close physical proximity to people. He has also advanced multi-agent planning through efficient iterative linear-quadratic approximations for nonlinear differential games (20 citations), enabling robots to reason about how their decisions influence other agents. Beyond theory, Ratner developed a web-based infrastructure for recording user demonstrations of mobile manipulation tasks (12 citations), facilitating real-world learning from demonstration. His recent work tackles the practical challenge of operating with inaccurate models by integrating control-level discrepancy information into planning, and he has proposed adaptive model sets for efficient dynamics estimation. Ratner’s contributions are shaping how robots navigate the messy, uncertain reality of human environments.
Research Focus
Key Achievements
Top Papers
- 1A Robust Control Framework for Human Motion Prediction27 citations · 2020
- 2
- 3
- 4
- 5
- 6Efficient Dynamics Estimation With Adaptive Model Sets2 citations · 2021
- 7Simplifying Reward Design through Divide-and-Conquer2 citations · 2018