# 四足机器人负载交互框架：仅凭本体传感实现柔顺搬运与轨迹引导

> 原文链接：https://robotics.hellobike.com/papers/2026-10-06-2610-08637
> 频道：papers | 更新日期：2026-10-06 | 标签：auto, arxiv
> 核心摘要：研究提出力感知运动框架，解决四足机器人搬运无约束负载时的人机交互难题。通过解耦力估计与柔顺策略训练，仅利用本体测量即可区分负载运动与外部意图，实现稳定搬运、柔顺响应及基于推断力的人机协作轨迹引导。

# 四足机器人负载交互框架：仅凭本体传感实现柔顺搬运与轨迹引导

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[Feeling Through the Load: Compliant Quadruped Locomotion under Payload Interactions](https://arxiv.org/abs/2610.08637)（arXiv 2610.08637）

## 核心内容

针对四足机器人在人类环境中搬运无约束物体时难以区分负载自然晃动与人为交互的痛点，该研究提出一种力感知运动框架。核心创新在于将负载交互视为组合系统运动指令，采用Estimator-in-the-loop数据聚合与微调策略，分离训练柔顺负载搬运策略与因果力估计器。仿真与真机实验证实，该系统仅依赖机载传感器即可在维持负载平衡与步态稳定的同时，准确推断外部作用力并支持人机协作变轨。

技术架构上，系统摒弃额外力传感器，完全基于本体 proprioceptive 测量进行状态推断。因果力估计器负责从复杂动力学中解耦出真实交互力，作为高层指令输入至柔顺控制策略。这种解耦设计解决了无约束负载动力学耦合导致的误触发问题，使机器人能在不增加硬件成本的前提下，实现对非结构化物理交互的语义理解与柔顺响应，为具身智能在动态人机环境中的操作提供了纯算法侧解决方案。

## 关键看点

- **解耦训练范式**：分离柔顺搬运策略与因果力估计器训练，采用Estimator-in-the-loop数据聚合微调，解决无约束负载动力学耦合干扰
- **纯本体感知交互**：无需外部力/触觉传感器，仅凭机载本体测量推断人机交互意图，显著降低硬件复杂度与系统集成成本
- **交互即指令机制**：将推断的外部作用力直接映射为组合系统运动命令，实现负载搬运过程中的柔顺让行与人机协作轨迹引导
- **无约束负载适配**：专门针对未固定负载场景优化，有效区分负载惯性运动与主动外力，保障复杂动态环境下的搬运稳定性

## 工程与产业观察

该工作验证了仅靠本体感知与先进控制学习即可替代昂贵力传感实现精细人机协作，为四足机器人进入家庭、仓储等动态环境提供低成本技术路径。其Estimator-in-the-loop训练范式对解决具身智能中“感知-控制”耦合难题具有通用参考价值，提示行业在追求多模态传感融合之余，应重视挖掘本体信号在物理交互理解中的潜力。

## 英文摘要 (Abstract)

Quadruped robots are increasingly expected to carry objects while moving through human environments. But what happens when a person interacts directly with the payload rather than with the robot? If the payload is unrestrained, the robot must distinguish intentional external interactions from ordinary payload motion, while still keeping the load balanced and maintaining stable locomotion. How can a quadruped infer and compliantly respond to such interactions using only onboard measurements? In this work, we develop a force-aware locomotion framework that treats payload interactions as commands that shape the motion of the combined robot-payload system. Our approach separates the learning of force-aware locomotion and force estimation on an unrestrained payload. We combine a compliant load-carrying policy with a causal force estimator, trained through estimator-in-the-loop data aggregation and finetuning, to predict interactions from onboard robot measurements. Our simulations and real-world experiments show that the resulting controller can maintain stable payload-carrying locomotion, yield compliantly to external interactions, and use the inferred force to support human-guided changes in the robot's trajectory.

作者：Shaunak A. Mehta, Mayank Mishra, Prajit KrisshnaKumar, Sebastian Scherer, Koichiro Niinuma
:::

## 原始参考来源 (Sources)

- [arXiv 2610.08637：Feeling Through the Load: Compliant Quadruped Locomotion under Payload Interactions](https://arxiv.org/abs/2610.08637)