论文arXiv 2611.190815 天前
RoboCoEvo框架实现Agent与VLA模型协同进化提升自主任务性能
本文由 AI 根据公开内容整理,可能有误,请以原文为准:RoboCoEvo: Expanding Robot Capabilities through Agent-Policy Co-Evolution(arXiv 2611.19081)
核心内容
针对MLLM智能体与机器人基础模型能力解耦导致系统性能受限的问题,研究提出RoboCoEvo失败引导协同进化框架。该方法将智能体显式可复用知识与VLA模型参数学习耦合,通过在BEHAVIOR-1K、RoboCasa及RoboTwin基准上的实验,证实多轮协同进化能持续提升VLA模型的自主任务执行性能,解决了组件独立优化无法互益的瓶颈。
技术上采用Code-as-Policy范式,由智能体开发并优化可执行的故障检测与恢复程序,生成针对性纠正经验数据。这些修正行为通过后训练整合至VLA模型端到端策略中,而更新后的策略失败分布变化又反向指导智能体程序修订。这种闭环机制实现了从显式代码逻辑到隐式神经网络参数的知识蒸馏与双向增强,强化了长尾场景下的鲁棒性。
关键看点
- 协同进化机制:构建失败引导闭环,将Agent生成的故障恢复代码转化为VLA模型后训练数据,实现显式知识与隐式策略的双向迭代增强。
- Code-as-Policy范式:智能体编写可执行故障检测与恢复程序作为策略载体,自动生成针对性纠正经验,解决传统强化学习样本效率低问题。
- 多基准性能验证:在BEHAVIOR-1K、RoboCasa和RoboTwin三大仿真平台测试,证明随协同进化轮次增加,VLA模型自主任务成功率持续稳定提升。
工程与产业观察
该工作为具身智能提供了一种解决大模型与底层控制策略割裂的工程范式。通过代码中介实现Agent推理能力向VLA端侧策略的有效蒸馏,不仅提升了Sim-to-Real迁移潜力,也为利用合成数据自动化修复机器人长尾失效提供了可复用的算法基础设施参考。
英文摘要 (Abstract)
Integrating MLLM-based agents with robotic foundation models has attracted increasing attention as a means of enhancing robotic capabilities. However, the improvement of such integrated systems remains largely decoupled across their agent and model components, with advances in one component not necessarily benefiting the other, substantially limiting the capabilities of the combined system. As both MLLMs and robotic foundation models become increasingly capable, fully realizing their combined potential calls for co-evolution, an iterative process in which improvements in agents and robotic foundation models reinforce one another. We propose RoboCoEvo, a failure-guided co-evolution framework that couples the refinement of the agent's explicit and reusable executable knowledge with parameter learning in robotic foundation models, such as vision-language-action (VLA) models. Through a code-as-policy formulation, the agent develops and refines executable failure-detection and recovery programs, generating targeted corrective experience. Post-training consolidates the resulting corrective behaviors into the robotic model's end-to-end policy, improving autonomous task execution. Changes in the updated policy's failure distribution, in turn, guide subsequent revisions to the agent's programs. Experiments on BEHAVIOR-1K, RoboCasa, and RoboTwin demonstrate sustained improvements in autonomous VLA task performance across successive co-evolution rounds.
作者:Ran Qi, Yixin Xiao, Dawei Wang, Zhiyuan Li, Jinbang Huang, Xuan Zhao, Yintao Ma, Yangzheng Wu, Tongtong Cao, Yingxue Zhang
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)