The Robot Is Not Its Description: GaugeBench for Representation Robustness in Morphology-Aware Policies

📄 arXiv: 2610.07597v1 📥 PDF

作者: Rahath Malladi, Arshia Sangwan, Rajesh K. Gupta, Tauhidur Rahman

分类: cs.RO, cs.LG

发布日期: 2026-10-06


💡 一句话要点

提出GaugeBench以解决机器人描述与行为不一致问题

🎯 匹配领域: 支柱七:动作重定向 (Motion Retargeting)

关键词: 机器人描述 行为鲁棒性 形态感知策略 跨实体评估 GaugeBench

📋 核心要点

  1. 现有的形态感知策略在跨实体评估时,通常保持机器人描述的约定不变,未能探讨描述变化对行为的影响。
  2. 本文提出GaugeBench,通过在物理上等效的约定下重写固定机制,评估描述变化对策略性能的影响。
  3. 实验结果表明,新的描述可能导致性能显著下降,轴反转的影响尤为明显,但通过特定训练可以恢复性能。

📝 摘要(中文)

机器人描述不仅仅是物理机制的规范,它还编码了关节轴方向、关节角零点及链接和关节的顺序与名称等约定。现有的形态感知策略在跨实体评估时通常保持这些约定不变,而本文探讨了在机器人保持不变的情况下,描述变化对行为的影响。GaugeBench通过在物理上等效的约定下重写固定机制,验证其物理特性和策略接口的保留,并评估相同的策略权重。结果显示,新的描述可能比新的机器人更具破坏性,且轴反转会导致性能崩溃,而关节角零点的变化几乎无害。通过跨等效轴约定训练,可以显著提高在轴反转下的保留回报。

🔬 方法详解

问题定义:本文旨在解决机器人描述变化对行为影响的问题,现有方法未能考虑描述变化带来的潜在风险,导致评估结果不准确。

核心思路:通过GaugeBench重写固定机制,保持物理等效性,评估描述变化对策略性能的影响,从而揭示描述与行为之间的关系。

技术框架:整体框架包括固定机制的重写、物理特性验证、策略接口评估和性能测试四个主要模块,确保在描述变化下的行为一致性。

关键创新:最重要的创新在于将描述变化与行为表现的关系明确分离,揭示了描述的鲁棒性与机制的鲁棒性之间的差异。

关键设计:在实验中,采用了特定的策略权重评估方法,关注轴反转、关节角零点变化和重排序等因素对性能的影响。

🖼️ 关键图片

fig_0
img_1
img_2

📊 实验亮点

实验结果显示,三种MetaMorph策略在80个熟悉的机器人上得分为4030.6,但在相同机器人重新描述后仅得51.6,表明描述变化对性能的显著影响。通过训练,轴反转下的保留回报从3.6%提升至80.6%。

🎯 应用场景

该研究的潜在应用领域包括机器人控制、自动化系统和人机交互等。通过提高机器人在不同描述下的鲁棒性,可以增强其在复杂环境中的适应能力,推动智能机器人技术的实际应用与发展。

📄 摘要(原文)

A robot description does more than specify a physical mechanism: it also encodes arbitrary conventions, such as joint-axis direction, joint-angle zero, and the order and names of links and joints. Morphology-aware policies consume interfaces built from these descriptions, yet cross-embodiment evaluation typically changes the robot while keeping those conventions fixed. This leaves a simple question unanswered: does behavior survive when the robot stays fixed but its description changes? GaugeBench isolates this case by rewriting a fixed mechanism under physically equivalent conventions, verifying that its physics and policy interface are preserved, and then evaluating the same policy weights. The result is stark: three MetaMorph policies score 4030.6 on 80 familiar robots, but only 51.6 when those same robots are equivalently re-described, while 98 genuinely held-out robots score 1489.6. A new description can therefore be more damaging than a new robot. Tracing the failure reveals that axis reversal alone reproduces the collapse, joint-angle zero changes are nearly harmless, and reordering lies between them; moreover, changing joint-state and torque coordinates alone is sufficient to cause the failure, while changing description-derived features alone is not. The same phenomenon appears in ModuMorph and an unrelated PyBullet framework. Yet it is not irreversible: exact two-description transport restores the original controller, and training across equivalent axis conventions raises retained return under axis reversal from 3.6% to 80.6%. Together, these results separate mechanism robustness from representation robustness and show that cross-embodiment evaluation should test both.