Wanli Peng
Papers
2
Total Citations
46
H-Index
2
About
Wanli Peng is a robotics researcher whose work centers on enabling robots to perform human-like, task-oriented grasping—a critical step toward more capable and intuitive robotic manipulation. His research bridges computer vision, semantic representation, and grasp pose estimation to move beyond simple stable grasps toward functional grasps that support subsequent manipulation tasks. Peng’s major contributions include the development of **FunctionalGrasp**, a framework that leverages semantic hand-object representations to learn grasps optimized for specific post-grasp actions, rather than merely for stability. This work, published in 2023, has already garnered 23 citations, reflecting its timely impact on the field. Additionally, his **TransGrasp** system (2022, also 23 citations) addresses the practical challenge of generalizing grasp poses across object categories from only a single labeled instance, significantly reducing the data burden for training robust grasp estimators. Through these innovations, Peng is advancing the frontier of robotic dexterity, making it possible for robots to not only pick up objects but to do so in a way that prepares them for meaningful interaction—a foundational capability for service robots, manufacturing, and assistive technologies. His work is essential reading for researchers in robotic manipulation, computer vision, and human-robot interaction.
Research Focus
Key Achievements
Top Papers
- 1
- 2