Zehui Lu

Purdue University West Lafayette, American Institute of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Zehui Lu is a rising researcher at the intersection of robotics, human-robot interaction, and multi-agent systems. Her work focuses on enabling robots to learn efficiently from non-expert human input and to coordinate seamlessly in complex, dynamic environments. In her highly cited 2022 paper, "Learning From Human Directional Corrections," Lu introduced a novel framework that allows robots to learn objective functions from intuitive directional feedback rather than precise magnitude corrections—a significant advancement that reduces cognitive load on human teachers and accelerates robot learning. This work has garnered 18 citations, establishing her as a key voice in interactive robot learning. More recently, Lu has tackled the grand challenges of multi-robot coordination. Her 2024 paper on real-time mission planning addresses the critical need for collision-aware task allocation and pathfinding in cluttered settings, achieving computational efficiency and scalability. She has also pioneered a double-layered framework for multi-robot formation control that integrates human-on-the-loop oversight, allowing operators to guide heterogeneous robot teams without constant intervention. Through these contributions, Lu is shaping a future where robots learn naturally from people and work together safely in real-world, unpredictable environments.