Achieving cyclic manipulation tasks in an end-to-end imitation manner, without relying on auxiliary models or incurring heavy computational overhead.

A generalizable dexterous framework that achieves omni-capabilities in user prompting, dexterous embodiment, and grasping tasks.

A novel task enables robots to perform dexterous grasping based on human commands.

A wearable FBG-Based tactile sensor which can transfered form human to robot.

Open-Set Language-guided dexterous grasp based on generalizable-instructive Affordance.

Endow the robots with the ability to learn new manipulation skills based on the previous learned knowledge without re-training.

A decoupled interaction framework that categorizes bimanual manipulation tasks into non-collaborative and collaborative types, and performs both effectively.

Human-Object interaction generation by joint-based kinematic chain modeling

A novel task enables robots to perform dexterous grasping based on human commands.

A new Real-to-Sim framework for 6-DoF Grasp detection, with the key insight of bridging sim-to-real gap in a real-to-sim way.

A new economic grasping framework for 6-DoF grasp detection to economize the training resource cost and meanwhile maintain effective grasp performance.

A novel discriminative framework for dexterous grasp generation by formulating it as a set prediction task.

A novel task of generating human grasps based on single-view scene point clouds.