Tianqiang Zhu
Papers
5
Total Citations
114
H-Index
4
About
Tianqiang Zhu is a leading researcher in robotic dexterous manipulation, with a focus on enabling robots to achieve human-like, functional grasping. His work centers on bridging the gap between stable, task-agnostic grasps and the semantically rich, task-oriented grasps that humans perform intuitively. Zhu’s major contributions include the development of a grasp synthesis framework that leverages semantic object-hand representations, allowing dexterous robotic hands to understand not just *how* to hold an object, but *why*—enabling functional, post-grasp manipulation. His most cited paper (2021, 50 citations) laid the groundwork for this approach, while subsequent works (2023, 27 and 23 citations) refined the concept of “FunctionalGrasp.” To support this research, Zhu created the DexFuncGrasp dataset (2024), a cost-effective real-simulation annotation system for high-degree-of-freedom hands. His latest work (2025) explores visuo-tactile fusion via multi-agent deep reinforcement learning, tackling the high-dimensional control challenge of multi-fingered hands. With a growing citation impact and a clear trajectory toward more intelligent, human-like robotic interaction, Zhu’s research is pivotal for the future of autonomous manipulation in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5