{"ID":2877913,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.18691","arxiv_id":"2508.18691","title":"Deep Sensorimotor Control by Imitating Predictive Models of Human Motion","abstract":"As the embodiment gap between a robot and a human narrows, new opportunities arise to leverage datasets of humans interacting with their surroundings for robot learning. We propose a novel technique for training sensorimotor policies with reinforcement learning by imitating predictive models of human motions. Our key insight is that the motion of keypoints on human-inspired robot end-effectors closely mirrors the motion of corresponding human body keypoints. This enables us to use a model trained to predict future motion on human data \\emph{zero-shot} on robot data. We train sensorimotor policies to track the predictions of such a model, conditioned on a history of past robot states, while optimizing a relatively sparse task reward. This approach entirely bypasses gradient-based kinematic retargeting and adversarial losses, which limit existing methods from fully leveraging the scale and diversity of modern human-scene interaction datasets. Empirically, we find that our approach can work across robots and tasks, outperforming existing baselines by a large margin. In addition, we find that tracking a human motion model can substitute for carefully designed dense rewards and curricula in manipulation tasks. Code, data and qualitative results available at https://jirl-upenn.github.io/track_reward/.","short_abstract":"As the embodiment gap between a robot and a human narrows, new opportunities arise to leverage datasets of humans interacting with their surroundings for robot learning. We propose a novel technique for training sensorimotor policies with reinforcement learning by imitating predictive models of human motions. Our key i...","url_abs":"https://arxiv.org/abs/2508.18691","url_pdf":"https://arxiv.org/pdf/2508.18691v1","authors":"[\"Himanshu Gaurav Singh\",\"Pieter Abbeel\",\"Jitendra Malik\",\"Antonio Loquercio\"]","published":"2025-08-26T05:25:57Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}