Universal Low-Cost Force-Controlled Gripper for Learning Delicate Object Grasping
In Submission, 2025
TL;DR: A versatile and affordable force-controlled gripper with tactile sensors for delicate object manipulation tasks.
Robotics Researcher
Tongxuan Tian
Hi, my name is Tongxuan Tian. I am a researcher in robotics.
I received my bachelor's degree from Tongji University and my master's degree from the University of Virginia. I am grateful to all collaborators who have supported and shaped my journey.
I aim to build robots that bring value to humans. Towards my goal, I study the emergence, alignment, and democratization of physical intelligence. If you find my work interesting, feel free to reach out.
Contact: tongxuan259 [at] gmail (dot) com
Selected Publications
Full list
O³Afford: One-Shot 3D Object-to-Object Affordance Grounding for Generalizable Robotic Manipulation
Conference on Robot Learning (CoRL), 2025
TL;DR: A novel framework using vision foundation models for one-shot 3D affordance learning for robotic manipulation.
UniClothDiff: Diffusion Dynamics Model with Generative State Estimation for Cloth Manipulation
Conference on Robot Learning (CoRL), 2025
TL;DR: A novel approach for estimating and learning cloth state and dynamics with diffusion models for cloth manipulation.
Deep Unrolled Weighted Graph Laplacian Regularization for Depth Completion
International Journal of Computer Vision (IJCV), 2024
TL;DR: We propose the deep unrolled Weighted Graph Laplacian Regularization (WGLR) for depth completion task which enforces input constraints in the network design.