Papers
21
Total Citations
210
H-Index
8
About
Minghui Zhu is a researcher whose work sits at the intersection of robotics, cyber-physical systems security, and multi-agent coordination. His research spans three interconnected domains: anomaly and attack detection in mobile robotic systems, multi-robot motion planning, and distributed machine learning for robotic networks. Zhu's most influential contribution, "RoboADS" (2018, 49 citations), introduced a pioneering anomaly detection framework that safeguards mobile robots against sensor and actuator misbehaviors by exploiting the tight coupling between cyberspace and physical dynamics — a theme he further developed in his 2017 work on detecting actuator and sensor attacks. In the realm of motion planning, Zhu has made substantial strides in formalizing multi-robot coordination as game-theoretic problems, developing both trajectory-based and policy-based algorithms that compute Nash equilibrium and Pareto optimal solutions across distributed robot teams. His more recent work ventures into federated reinforcement learning for zero-shot generalization in robot motion planning and communication-aware Gaussian process regression for real-time collaborative learning. Collectively, his publications reflect a sustained commitment to making autonomous robotic systems both safer and more intelligent, earning him over 160 cumulative citations across his most recognized contributions.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Pareto Optimal Multirobot Motion Planning21 citations · 2020
- 4
- 5
- 6Pareto optimal multi-robot motion planning10 citations · 2018
- 7
- 8
- 9
- 10