外观
论文arXiv 2610.2312712 天前
BeJEPA：通过无监督RL实现的目标条件人形行为世界模型
BeJEPA用无监督RL和JEPA学习目标条件行为世界模型，并在运动跟踪与Unitree G1真机上验证。
宇树 G1原文链接：BeJEPA: A Goal-Conditioned Humanoid Behavior World Model via Unsupervised RL（arXiv 2610.23127 · 发布日期：2026-10-16）
BeJEPA: A Goal-Conditioned Humanoid Behavior World Model via Unsupervised RL
BeJEPA用无监督RL和JEPA学习目标条件行为世界模型，并在运动跟踪与Unitree G1真机上验证。
核心要点
- 人形控制需协调全身姿态与世界坐标中的位置、朝向，主流根坐标轨迹模仿难以在位移后恢复目标世界姿态。
- BeJEPA结合人类运动先验与机器人交互，通过JEPA学习身体构型和世界位置的共享表征。
- Dynamic预测动作下的下一状态嵌入，Successor用单向量概括目标条件策略的折扣未来状态嵌入。
- 20秒运动跟踪中，结合在线动作细化的BeJEPA在全局身体位置误差上优于所有评估基线；项目页已给出。
验证范围与局限：验证范围包括20秒运动跟踪实验，以及Unitree G1真机上的姿态、世界位置和朝向目标、起身与扰动/目标变化后的闭环返回。
英文摘要 (Abstract)
Humanoid control requires coordinating whole-body posture with position and heading in the world frame. Mainstream controllers imitate trajectories in the root frame, but struggle to regain the target world pose after displacement. We introduce BeJEPA, a goal-conditioned Behavior World Model learned through unsupervised reinforcement learning from human motion priors and robot interaction. BeJEPA learns a shared representation of body configuration and world placement through a joint-embedding predictive architecture (JEPA). A Dynamic Predictor predicts the next state embedding under an action, while a Successor Predictor summarizes a goal-conditioned policy's expected discounted future state embeddings in a single vector. During joint training, Dynamic grounds the initial representation in observed physical transitions, while Successor increasingly organizes it around long-term behavior. Their composition guides policy improvement through one physical prediction and one behavioral summary, without long imagined rollouts. In 20-second motion-tracking experiments, BeJEPA with online action refinement using Dynamic outperforms all evaluated baselines in global body-position error, reducing it on AMASS by 61.7% relative to BFM-Zero with world-pose inputs and 58.7% relative to SONIC. Physical deployment on a Unitree G1 demonstrates joint attainment of posture, world position, and heading goals, recovery from the ground, and closed-loop return after disturbances and goal changes. Project page: https://ylyem9x.github.io/BeJEPA-project.
作者：Liyun Yan, Haoyang Li, Yang Zhang, Jialing Zhu, Zhanxiang Cao, Zihao Zhou, Keqi Zhu, Yizhi Chen, Chuxuan Huang, Shengcheng Fu
来源
本页事实对应的公开页面，以原文为准。