Papers
17
Total Citations
227
H-Index
8
About
Haiyue Zhu is a versatile robotics and control systems researcher whose work spans precision motion control, variable stiffness robotics, and data-driven learning for robotic manipulation. With contributions spanning over three decades — from early foundational work on model-based control of robotic manipulators in the 1990s to cutting-edge machine learning applications in the 2020s — Zhu has consistently pushed the boundaries of intelligent automation. Among Zhu's most impactful contributions is a data-driven multiobjective controller optimization framework for magnetically levitated nanopositioning systems (53 citations), which elegantly addresses the limitations of traditional model-dependent control by leveraging real system data. This work reflects a broader research theme: replacing fragile model assumptions with adaptive, learning-based alternatives. Equally significant is Zhu's development of structure-controlled variable stiffness robotic joints (35 citations), enabling safer and more dexterous human-robot interaction. Zhu has also made notable strides in robotic perception, advancing grasping detection through semi-supervised domain adaptation (30 citations) and few-shot incremental object detection — capabilities critical for flexible real-world deployment. Together, these contributions, accumulating over 200 citations, establish Zhu as a compelling bridge between classical control theory and modern data-driven robotics, making their work essential reading for researchers in intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6Incremental Few-Shot Object Detection for Robotics15 citations · 2022
- 7An enhanced scheme for the model-based control of robot manipulators15 citations · 1992
- 8
- 9Towards Generalized and Incremental Few-Shot Object Detection6 citations · 2021
- 10Motion control of robotic manipulators with disturbance decoupling4 citations · 2002