Papers
12
Total Citations
106
H-Index
4
About
Negar Mehr is a robotics and autonomous systems researcher whose work spans multi-agent interactions, game-theoretic planning, safe navigation, and learning-based control. She is perhaps best known for her development of Potential iLQR and its distributed extension, algorithms that leverage differential game theory to efficiently compute equilibrium trajectories for multiple interacting agents — work that has garnered nearly 50 citations and established her as a notable voice in multi-robot coordination. Her earlier contributions to shared autonomy, particularly inferring user constraints to better assist people with motor impairments, reflect a consistent commitment to human-centered robotics. More recently, Mehr has tackled the challenge of safe robot navigation in crowded environments, integrating distributionally robust control with learned human motion models to provide formal safety guarantees — a technically demanding problem at the intersection of machine learning and control theory. Her work on constrained reinforcement learning and contraction theory-based policy learning further demonstrates her breadth, addressing provable constraint satisfaction in learned controllers. Across these diverse threads, Mehr's research is unified by a rigorous mathematical approach to making autonomous robots reliable, interactive, and safe in complex real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2Inferring and assisting with constraints in shared autonomy22 citations · 2016
- 3
- 4
- 5
- 6Learning Contraction Policies From Offline Data4 citations · 2022
- 7
- 8
- 9
- 10Distributed Autonomous Robotic Systems2 citations · 2024