FOCUS: Foot Observation Confidence for Robust Humanoid Proprioceptive Odometry

📄 arXiv: 2609.02222v1 📥 PDF

作者: Kaixin Feng, Angsong Li, Shaopeng Zhang, Enyu Li, Peiwen Lin, Chuang Wang, You Li, Haiyu Lan

分类: cs.RO

发布日期: 2026-09-02

备注: 8pages,6figures


💡 一句话要点

提出FOCUS以解决腿部自我定位中的接触决策问题

🎯 匹配领域: 支柱一:机器人控制 (Robot Control)

关键词: 自我定位 动态运动 机器人技术 脚部运动学 扩展卡尔曼滤波 可靠性权重 IMU融合 仿真训练

📋 核心要点

  1. 现有方法依赖于二元接触决策,无法有效处理动态运动中的部分支撑和脚滑现象,导致轨迹漂移。
  2. FOCUS通过预测连续的FK可靠性权重,融合IMU和FK观测,避免了硬接触切换,提升了定位精度。
  3. 实验显示FOCUS在模拟步态中将绝对轨迹误差降低了83.7%,在真实步态中降低了70.8%,有效提升了动态运动的保真度。

📝 摘要(中文)

脚部前向运动学(FK)广泛用于提高腿部自我定位的可靠性,然而现有的接触辅助估计方法依赖于二元接触决策,导致在动态运动中出现显著漂移。为此,本文提出FOCUS(来自未标注仿真的脚部观察置信度),通过预测每只脚的连续FK可靠性权重,替代传统的二元接触估计。FOCUS将预测的可靠性权重与IMU传播的身体速度融合,适应扩展卡尔曼滤波器的观测协方差,从而实现平滑的可靠性感知融合。实验结果表明,FOCUS在模拟和真实步态中显著降低了绝对轨迹误差(ATE),提升了动态运动的保真度。

🔬 方法详解

问题定义:现有的接触辅助估计方法通常依赖于二元接触决策来判断脚部运动学测量的可靠性,但在动态运动中,接触并不总是意味着可靠性,因此导致了显著的轨迹漂移。

核心思路:FOCUS提出了一种新的方法,通过预测每只脚的连续FK可靠性权重,而不是简单的二元接触判断。这种设计允许在动态运动中更灵活地处理脚部状态,提高了自我定位的准确性。

技术框架:FOCUS的整体架构包括一个基于IMU和关节运动学测量的网络模型,该模型通过FK加权的速度一致性损失进行训练,使用轻量级的仿真信号生成。模型输出的可靠性权重用于融合FK速度观测和IMU传播的身体速度,并适应EKF的观测协方差。

关键创新:FOCUS的主要创新在于引入了连续的FK可靠性权重预测,而不是依赖于传统的二元接触决策。这一方法显著减少了在动态环境中由于接触判断错误而导致的轨迹漂移。

关键设计:在训练过程中,FOCUS使用了FK加权的速度一致性损失和轻量级的仿真接触正则化,避免了对手动标注的FK可靠性标签的依赖,确保了模型的实用性和适应性。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

FOCUS在模拟步态中将绝对轨迹误差(ATE)降低了83.7%,在19个真实步态段中降低了70.8%,并在四个真实动态运动例程中平均降低了42.7%。这些结果表明FOCUS在动态运动中的有效性和可靠性,显著提升了运动保真度。

🎯 应用场景

FOCUS的研究成果在机器人导航、动态步态控制和人形机器人等领域具有广泛的应用潜力。通过提高自我定位的准确性,该方法能够增强机器人在复杂环境中的自主性和适应能力,推动智能机器人技术的发展。

📄 摘要(原文)

Foot forward kinematics (FK) is widely used to improve proprioceptive legged odometry by providing reliable velocity constraints during foot support. Existing contact-aided estimators generally rely on binary contact decisions to determine whether the FK measurements of an entire foot should be trusted. However, contact does not necessarily imply FK reliability. Dynamic locomotion often involves partial support, toe dragging, and foot slip, causing binary contact decisions to accumulate significant drift over long trajectories. To address this limitation, we propose FOCUS (Foot Observation Confidence from Unannotated Simulation), which predicts a continuous FK reliability weight for each foot instead of estimating binary foot contact. Rather than replacing the model-based estimator, the predicted reliability weights are used to blend FK velocity observations with IMU-propagated body velocity and to adapt the observation covariance of an extended Kalman filter (EKF), enabling smooth reliability-aware fusion without hard contact switching. The network is trained from automatically generated simulation signals using an FK-weighted velocity consistency loss with lightweight simulator-contact regularization, without manually annotated continuous FK-reliability labels. The deployed model relies only on IMU and joint kinematic measurements, making it suitable for hardware platforms with unreliable torque sensing. Experiments demonstrate that FOCUS reduces absolute trajectory error (ATE) by 83.7% on simulated walking episodes, preserves simulated dynamic-motion fidelity in motion scale and spectral energy, reduces ATE by 70.8% across 19 real walking segments, and reduces mean ATE by 42.7% across four real dynamic-motion routines.