{"ID":2886113,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.03068","arxiv_id":"2508.03068","title":"Hand-Eye Autonomous Delivery: Learning Humanoid Navigation, Locomotion and Reaching","abstract":"We propose Hand-Eye Autonomous Delivery (HEAD), a framework that learns navigation, locomotion, and reaching skills for humanoids, directly from human motion and vision perception data. We take a modular approach where the high-level planner commands the target position and orientation of the hands and eyes of the humanoid, delivered by the low-level policy that controls the whole-body movements. Specifically, the low-level whole-body controller learns to track the three points (eyes, left hand, and right hand) from existing large-scale human motion capture data while high-level policy learns from human data collected by Aria glasses. Our modular approach decouples the ego-centric vision perception from physical actions, promoting efficient learning and scalability to novel scenes. We evaluate our method both in simulation and in the real-world, demonstrating humanoid's capabilities to navigate and reach in complex environments designed for humans.","short_abstract":"We propose Hand-Eye Autonomous Delivery (HEAD), a framework that learns navigation, locomotion, and reaching skills for humanoids, directly from human motion and vision perception data. We take a modular approach where the high-level planner commands the target position and orientation of the hands and eyes of the huma...","url_abs":"https://arxiv.org/abs/2508.03068","url_pdf":"https://arxiv.org/pdf/2508.03068v2","authors":"[\"Sirui Chen\",\"Yufei Ye\",\"Zi-Ang Cao\",\"Jennifer Lew\",\"Pei Xu\",\"C. Karen Liu\"]","published":"2025-08-05T04:30:02Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}