Papers
8
Total Citations
106
H-Index
4
About
Zerui Li is a robotics researcher whose work centers on autonomous terrain perception and classification for field robots, with a particular focus on vibration-based methods. His major contributions lie in developing machine learning approaches that enable wheeled and legged robots to identify traversing terrains using vibration signals generated by robot-terrain interaction. Li’s most cited work, “Comparative Study of Different Methods in Vibration-Based Terrain Classification for Wheeled Robots with Shock Absorbers” (2019, 43 citations), systematically evaluates classification techniques for enhancing robot safety and efficiency. He has pioneered semi-supervised and unsupervised learning frameworks for this task, including the Feature-Temporal Semi-Supervised Extreme Learning Machine (2020, 27 citations) and Laplacian Support Vector Machine (2020, 22 citations), which reduce the need for human supervision. A key challenge Li addresses is domain adaptation—ensuring terrain classifiers trained in controlled environments remain accurate in dynamic, real-world settings—as seen in his work on unsupervised domain adaptation (2022) and broad feature alignment (2021). Beyond perception, Li contributed to robot design with the SmallRhex hexapod robot (2022) and to skid-steering kinematics estimation (2018). His cumulative work, with over 100 citations, advances the reliability of autonomous navigation in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6SmallRhex: A Fast and Highly-Mobile Hexapod Robot2 citations · 2022
- 7
- 8