外观
论文arXiv 2609.214471 个月前
FootQuery：面向感知式人形机器人运动的未来触地引导深度历史检索
FootQuery按预测足端触地点检索历史深度并融合视觉记忆，在仿真复杂地形和Unitree G1真实路线验证人形感知运动。
宇树 G1原文链接：FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion（arXiv 2609.21447 · 发布日期：2026-09-18）
FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion
FootQuery按预测足端触地点检索历史深度并融合视觉记忆，在仿真复杂地形和Unitree G1真实路线验证人形感知运动。
核心要点
- 复杂地形中触地点可能在落脚时不可见，需利用早先地形观测。
- 策略由本体感知预测触地位置与不确定性，查询稀疏历史深度帧。
- 训练将实际接触投影到历史图像监督检索，并融合全局视觉记忆生成控制。
- 仿真优于组件消融，真机以单策略连续通过室外楼梯和室内复合路线。
验证范围与局限：验证包括仿真中的楼梯、间隙、平台，以及Unitree G1上的室外楼梯和室内含上下楼梯、平台、间隙路线。
英文摘要 (Abstract)
Humanoid locomotion over complex terrain requires anticipating footholds that may no longer be visible at touchdown. Limited camera coverage and self-occlusion make it necessary to retrieve relevant terrain information from earlier observations. We present FootQuery, a perceptive locomotion framework that queries depth history using each foot's predicted next touchdown. The policy predicts touchdown locations and uncertainty from proprioception and uses these distributions, together with per-foot features, to query sparsely sampled historical depth frames. During training, realized contacts are projected into historical images to supervise retrieval at the regions where those contacts were visible. The retrieved per-foot features are fused with global visual memory to generate control actions. A progressive force-assistance curriculum supports early exploration, while event-consistent tread-midline shaping encourages coordinated stair contacts. Deployment requires only proprioception and onboard depth images. In simulation, the complete framework outperforms its component ablations on the most challenging tested stairs, gaps, and platforms. Real-world experiments on a Unitree G1 demonstrate continuous traversal with a single policy across outdoor stairs and indoor routes combining stair ascent and descent, platforms, and gaps. These results support organizing visual history around anticipated contacts for perceptive humanoid locomotion.
作者：Tao Dong, Jia Yu, Yuxuan Fan, Linna Zhao, Jiaqi Gong, Andong Yang, Chao Gao, Guyue Zhou
来源
本页事实对应的公开页面，以原文为准。