|
Zhengtong Xu 徐政通 Email: xu1703 AT purdue.edu I'm a third-year PhD candidate at Purdue University, advised by Professor Yu She. I received my Bachelor's degree in mechanical engineering at Huazhong University of Science and Technology. G. Scholar / Twitter / Github / LinkedIn |
|
News
|
ResearchMy research aims to develop robots capable of performing everyday manipulation tasks with human-level capabilities. Towards this goal, my current focus is on differentiable optimization for policies, generative models in robotics, and tactile sensing. |
|
LeTac-MPC: Learning Model Predictive Control for Tactile-reactive Grasping
A generalizable end-to-end tactile-reactive grasping controller with differentiable MPC, combining learning and model-based approaches. |
|
|
UniT: Unified Tactile Representation for Robot Learning
Learn a unified tactile representation with transferability and generalizability only by a single simple object. |
|
|
VILP: Imitation Learning with Latent Video Planning
VILP integrates the video generation model into policies, enabling the representation of multi-modal action distributions while reducing reliance on extensive high-quality robot action data. |
|
|
LeTO: Learning Constrained Visuomotor Policy with Differentiable Trajectory Optimization
LeTO is a "gray box" method which marries optimization-based safety and interpretability with representational abilities of neural networks. |
|
|
VisTac: Toward a Unified Multimodal Sensing Finger for Robotic Manipulation
VisTac seamlessly combines high-resolution tactile and visual perception in a single unified device. |
Awards
|
Reviewer Service
|
Teaching
|
|
Website template from Jon Barron's website. |