Papers
7
Total Citations
208
H-Index
7
About
Thomas Weng is a roboticist whose research bridges human-robot interaction and dexterous manipulation, with a focus on making robots more effective collaborators. His work is defined by two major threads: enabling robots to communicate naturally with humans, and equipping them with the tactile intelligence needed to handle complex, deformable objects. In his highly cited 2016 paper (64 citations), Weng demonstrated that robot nonverbal behaviors—such as gestures and gaze—significantly improve task performance, especially in difficult collaborations. This foundational work led to models for legible object referencing using verbal cues and even situated projections, making robot intent transparent to human partners. More recently, Weng has tackled the challenging domain of cloth manipulation. His 2022 paper introduced a method for singulating layers of fabric using tactile feedback, addressing a notoriously difficult problem due to cloth's high degrees of freedom and self-occlusion. In 2023, he advanced robot grasping with Neural Grasp Distance Fields (NGDF), a novel neural field approach that predicts continuous grasp manifolds for arbitrary objects. With over 200 total citations, Weng's contributions are shaping how robots both understand their physical environment and communicate their intentions—a dual focus essential for the next generation of collaborative robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Neural Grasp Distance Fields for Robot Manipulation33 citations · 2023
- 4Learning to Singulate Layers of Cloth using Tactile Feedback23 citations · 2022
- 5
- 6RobotIST17 citations · 2018
- 7Robot Object Referencing through Legible Situated Projections14 citations · 2019