5 papers
HapticWorld: an Interactive World Simulator with Real-time Torque Feedback
Shaoting Peng, Litian Liang, Yixuan Wang +4
Contact-rich manipulation depends on force sensing that is hard to infer from visual signals alone, both for collecting demonstrations and for training policies. Force-annotated da…
One Demo is Worth a Thousand Trajectories: Action-View Augmentation for Visuomotor Policies
Chuer Pan, Litian Liang, Dominik Bauer +4
Visuomotor policies for manipulation have demonstrated remarkable potential in modeling complex robotic behaviors, yet minor alterations in the robot's initial configuration and un…
Universal Manipulation Exoskeleton: Learning Compliant Whole-body Policies with Real-time Torque Feedback
Litian Liang, Jingxi Xu, Xinda Qi +7
For robots to work safely in household environments, they need to be compliant and react to torque and force feedback during contact. However, the majority of existing data collect…
When Should We Prefer State-to-Visual DAgger Over Visual Reinforcement Learning?
Tongzhou Mu, Zhaoyang Li, Stanisław Wiktor Strzelecki +4
Learning policies from high-dimensional visual inputs, such as pixels and point clouds, is crucial in various applications. Visual reinforcement learning is a promising approach th…
Robo360: A 3D Omnispective Multi-Material Robotic Manipulation Dataset
Litian Liang, Liuyu Bian, Caiwei Xiao +6
Building robots that can automate labor-intensive tasks has long been the core motivation behind the advancements in computer vision and the robotics community. Recent interest in…