DiffusionHandover: Reliable Human-to-Robot Handover Generation With Anthropomorphic Hand
- 13 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (5) , 6272-6279
Abstract
Human-to-robot handover is a fundamental capability in human-robot interaction, critical for effective collaboration in service and assistive domains. Despite recent progress, ensuring both reliability and safety—particularly collision-free interaction with the human hand—remains a major challenge, especially when using anthropomorphic robotic hands. In this work, we propose DiffusionHandover, a novel framework built on a Decomposed Vector-Quantized VAE (DVQ-VAE) latent diffusion model, further enhanced with reinforcement learning from human feedback (RLHF) to improve grasp reliability and alignment with human preferences. We validate our approach extensively in both simulation and the real world using a Schunk SVH anthropomorphic hand. Our method achieves an average success rate above 80 % with diverse grasp configurations on unseen objects. In addition, we conduct ablation studies to assess individual submodules, as well as comparative evaluations against state-of-the-art baselines.Keywords
Funding Information
- National Natural Science Foundation of China (82427802)
- China Scholarship Council (202304910272)
- Natural Science Foundation of Beijing Municipality (L232035)
- Natural Science Foundation of Beijing Municipality (L222154)
This publication has 36 references indexed in Scilit:
- Robot learning in the era of foundation models: a surveyNeurocomputing, 2025
- UGG: Unified Generative Grasping2024
- Decomposed Vector-Quantized Variational Autoencoder for Human Grasp Generation2024
- GraspXL: Generating Grasping Motions for Diverse Objects at Scale2024
- Uniform Transformation: Refining Latent Representation in Variational Autoencoders2024
- Surface Representation for Point Clouds2022
- High-Resolution Image Synthesis with Latent Diffusion Models2022
- Fast Point Transformer2022
- OakInk: A Large-scale Knowledge Repository for Understanding Hand-Object Interaction2022
- Interactively shaping robot behaviour with unlabeled human instructionsAutonomous Agents and Multi-Agent Systems, 2020