{"ID":23475862,"CreatedAt":"2026-09-18T01:09:05.407443952Z","UpdatedAt":"2026-09-20T18:11:56.143995915Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2609.19340","arxiv_id":"2609.19340","title":"ViLoMan: Learning Visual-Proprioceptive Whole-Body Loco-Manipulation Skills for Humanoid Robots","abstract":"Humanoid loco-manipulation requires adaptive whole-body coordination to seamlessly integrate locomotion and physical interaction. Despite recent advances, learning autonomous loco-manipulation remains challenging due to the scarcity of diverse, physically executable robot-object interaction data and the difficulty of learning unified whole-body control directly from onboard observations. We present ViLoMan, a scalable framework for autonomous humanoid loco-manipulation. ViLoMan first transforms partial kinematic demonstrations of human-object interactions into complete, physically executable robot trajectories. It then leverages these trajectories within a teacher-student distillation framework to learn a unified policy that maps egocentric depth observations and proprioceptive measurements directly to joint-level whole-body actions. During deployment, the policy requires neither reference motions nor intermediate commands. We evaluate ViLoMan on door-closing tasks across diverse door configurations and robot initial conditions in both simulation and the real world. Experimental results demonstrate that a single policy enables a Unitree G1 humanoid to complete the full task using only onboard depth sensing and proprioception, while generalizing robustly across task variations and transferring effectively from simulation to reality. Project page: viloman-anonymous.pages.dev.","short_abstract":"Humanoid loco-manipulation requires adaptive whole-body coordination to seamlessly integrate locomotion and physical interaction. Despite recent advances, learning autonomous loco-manipulation remains challenging due to the scarcity of diverse, physically executable robot-object interaction data and the difficulty of l...","url_abs":"https://arxiv.org/abs/2609.19340","url_pdf":"https://arxiv.org/pdf/2609.19340v1","authors":"[\"Zejie Tian\",\"Ruibing Hou\",\"Bingpeng Ma\",\"Börje F. Karlsson\",\"Shiguang Shan\"]","published":"2026-09-16T19:10:13Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}