用物理接触点替代语言指令，让机器人更可靠地完成抓取等基础操作。
Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models
- 以空间中的物理接触点作为控制条件，构建可模块化复用的实用型策略
- 仅用23小时演示数据，在新环境和新机器人上即能零样本泛化，性能比先进多模态模型高56%
- 通过轻量级仿真平台EgoGym实现真实与仿真间快速迭代，加速部署
当前机器人学习普遍依赖运行时的语言提示来跨环境、跨形态、跨任务泛化，但语言抽象性与物理操作所需具体理解之间存在根本矛盾。本文提出接触锚定策略（Contact-Anchored Policies, CAP），用空间中的物理接触点替代语言条件。同时，将CAP设计为模块化实用模型库而非单一通用策略。该结构支持真实世界与仿真间的迭代优化：我们构建了EgoGym——一个轻量级仿真基准，用于快速识别失效模式并优化模型与数据集，再投入真实部署。实验表明，通过接触条件与仿真迭代，CAP在三种基础操作技能上无需微调即可泛化至新环境与新机器人形态，仅需23小时演示数据，且在零样本评估中性能优于主流大模型多模态动作系统（VLAs）56%。所有模型检查点、代码库、硬件配置、仿真环境及数据集均将开源。项目主页：https://cap-policy.github.io/
原文摘要 · Abstract (English)
The prevalent paradigm in robot learning attempts to generalize across environments, embodiments, and tasks with language prompts at runtime. A fundamental tension limits this approach: language is often too abstract to guide the concrete physical understanding required for robust manipulation. In this work, we introduce Contact-Anchored Policies (CAP), which replace language conditioning with points of physical contact in space. Simultaneously, we structure CAP as a library of modular utility models rather than a monolithic generalist policy. This factorization allows us to implement a real-to-sim iteration cycle: we build EgoGym, a lightweight simulation benchmark, to rapidly identify failure modes and refine our models and datasets prior to real-world deployment. We show that by conditioning on contact and iterating via simulation, CAP generalizes to novel environments and embodiments out of the box on three fundamental manipulation skills while using only 23 hours of demonstration data, and outperforms large, state-of-the-art VLAs in zero-shot evaluations by 56%. All model checkpoints, codebase, hardware, simulation, and datasets will be open-sourced. Project page: https://cap-policy.github.io/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。