Wenbin Hu
Papers
7
Total Citations
73
H-Index
4
About
Wenbin Hu is a robotics researcher specializing in dexterous robotic manipulation, machine learning-based control, and autonomous grasping systems. His work sits at the intersection of deep reinforcement learning, tactile sensing, and human-inspired robot skill acquisition, addressing some of the most challenging problems in robot manipulation. Hu's most cited contribution, "Learning Pregrasp Manipulation of Objects from Ungraspable Poses" (28 citations), drew inspiration from human bimanual manipulation to enable robots to handle objects in configurations previously inaccessible to standard grasping algorithms — a meaningful advance for real-world deployment. His research into learning grasping policies from teleoperated human demonstrations (18 citations) demonstrated how human expertise can be effectively transferred to robots through multi-sensing neural network architectures. More recently, his 2024 work on dexterous in-hand manipulation of slender cylindrical objects using tactile sensing (14 citations) showcases his growing focus on fine motor control. Beyond grasping, Hu has tackled dynamic challenges such as in-flight object catching using modular neural network frameworks, and developed unified policies for reaching, grasping, and re-grasping under uncertainty. Together, his portfolio reflects a sustained commitment to building robots capable of adaptive, human-like manipulation in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1Learning Pregrasp Manipulation of Objects from Ungraspable Poses28 citations · 2020
- 2
- 3
- 4
- 5Reaching, Grasping and Re-grasping: Learning Multimode Grasping Skills4 citations · 2020
- 6Learning Motor Skills of Reactive Reaching and Grasping of Objects3 citations · 2021
- 7Learning Pregrasp Manipulation of Objects from Ungraspable Poses2 citations · 2020