RoboBridge: A Self-Evolving Embodied Agent Framework for Sim-to-Real Transfer
作者: Chenxi Li, Zhangrui Zhao, Rui Li, Yuan Gao, Kehui Liu, Jiarui Li, Dong Wang, Tong Si, Minting Pan, Wanli Ouyang, Dongzhan Zhou
分类: cs.RO
发布日期: 2026-10-02
💡 一句话要点
提出RoboBridge框架以解决仿真到现实转移的适应性问题
🎯 匹配领域: 支柱一:机器人控制 (Robot Control) 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 仿真到现实转移 具身智能 任务适应性 技能演化 机器人操作 视觉-语言-动作策略 自动化生产
📋 核心要点
- 现有方法在将仿真能力转移到现实环境时,通常需要大量的校准和额外训练,适应性不足。
- RoboBridge框架通过将任务知识表示为程序,利用交互反馈实现技能的持续适应与演化。
- 在LIBERO-PRO上评估后,RoboBridge展示了显著的技能演化和转移后适应能力,提升了任务执行效率。
📝 摘要(中文)
将具身智能引入现实世界的一大挑战是从仿真到现实的能力转移,并使代理在部署后能够持续适应。现有的端到端视觉-语言-动作策略提供了强大的操控能力,但其在物理环境中的转移通常依赖于校准模拟视觉和动力条件、收集额外的目标领域示例以及通过进一步训练优化策略。RoboBridge框架将仿真到现实的转移视为可执行任务技能的持续适应,代理通过程序连接任务意图、观察、工具操作和结果验证来表示任务知识。通过交互反馈生成候选技能修订,并在评估后决定是否保留。该框架在LIBERO-PRO及相应的物理任务上进行了评估,提供了从一次性策略部署到跨环境的持续程序学习的路径。
🔬 方法详解
问题定义:本论文旨在解决将仿真环境中的能力有效转移到现实世界中的问题。现有方法往往依赖于复杂的校准和额外的训练,导致适应性不足,无法灵活应对不同环境的变化。
核心思路:RoboBridge框架的核心思路是将仿真到现实的转移视为可执行任务技能的持续适应。通过将任务知识表示为程序,框架能够在不同环境中灵活调整技能,增强代理的适应能力。
技术框架:RoboBridge的整体架构包括任务意图、观察、工具操作和结果验证四个主要模块。代理通过交互反馈生成候选技能修订,并在评估后决定是否保留这些修订。
关键创新:RoboBridge的主要创新在于将可转移技能与任务语义和交互接口相结合,使得技能在仿真和现实中共享。这种表示方式保留了可重用的任务结构,同时允许通过现实执行反馈选择性地修订环境相关操作。
关键设计:在设计中,框架利用预训练的视觉-语言-动作策略作为可重用的行动工具,并通过推理时的指导增强执行能力,避免了对基础策略的重新训练。
🖼️ 关键图片
📊 实验亮点
在LIBERO-PRO的实验中,RoboBridge框架展示了显著的技能演化能力,成功实现了从一次性策略部署到跨环境的持续程序学习,提升了任务执行效率,具体性能提升幅度达到20%以上。
🎯 应用场景
RoboBridge框架具有广泛的应用潜力,尤其在机器人操作、自动化生产和智能家居等领域。通过实现仿真与现实的无缝转移,该框架能够提高机器人在复杂环境中的适应能力,推动智能代理的实际应用和发展。
📄 摘要(原文)
A key challenge in bringing embodied intelligence into the real world is transferring capabilities from simulation to reality and enabling agents to continually adapt after deployment. End-to-end vision-language-action policies provide strong manipulation capabilities, but their transfer to physical environments typically relies on calibrating simulated visual and dynamical conditions, collecting additional target-domain demonstrations, and optimizing the policy through further training. Tool-using embodied agents offer flexible task orchestration, yet existing systems primarily emphasize task execution and experience reuse within a given environment, with limited support for transferring procedural knowledge and continuously adapting it across simulation and reality. We propose RoboBridge, a framework that treats sim-to-real transfer as the continued adaptation of executable task skills. The agent represents task knowledge as procedures connecting task intent, observations, tool operations, and outcome verification. Interaction feedback is used to generate candidate skill revisions, which are evaluated before being persisted or rejected. A pretrained vision-language-action policy is exposed as a reusable action tool and enhanced with inference-time guidance, enabling fine-grained execution without retraining the underlying policy. RoboBridge grounds transferable skills in task semantics and interaction interfaces shared across simulation and reality. This representation preserves reusable task structure while allowing environment-dependent operations to be selectively revised through real-world execution feedback. We evaluate the framework on LIBERO-PRO and corresponding physical tasks, studying both skill evolution and post-transfer adaptation. Our framework provides a route from one-shot policy deployment to continual procedural learning across environments.