ROSA：利用机器人状态实现视觉-语言与动作对齐
ROSA: Harnessing Robot States for Vision-Language and Action Alignment
AI总结：
针对现有VLA模型微调存在时空鸿沟、数据效率低的问题，提出ROSA训练范式，通过整合自动化获取的机器人状态估计数据增强模型空间理解与自我感知，在低数据场景下提升了机器人控制性能与泛化性。
摘要：
视觉-语言-动作（Vision-Language-Action, VLA）模型凭借视觉-语言模型（Vision-Language Models, VLMs）的强泛化能力，近期在多任务端到端机器人控制领域取得了显著进展。开发这类模型的核心挑战在于如何有效对齐视觉-语言空间与机器人动作空间。现有方法通常依赖专家演示直接微调VLMs，但该策略存在时空鸿沟，导致数据效率低下且高度依赖人力：空间上，VLMs在高层语义空间运行，而机器人动作基于底层3D物理空间；时间上，VLMs主要解读当前状态，而VLA模型需要预测未来动作。为解决这些问题，我们提出了一种新型训练范式ROSA，该范式利用机器人状态估计来提升视觉-语言与动作空间的对齐效果。通过整合自动化流程获取的机器人状态估计数据，ROSA使VLA模型获得更强的空间理解能力与自我感知能力，从而提升性能与泛化性。在模拟环境与真实环境中开展的大量实验证明了ROSA的有效性，尤其在低数据量场景下表现突出。
英文摘要：
Vision-Language-Action (VLA) models have recently made significant advance in multi-task, end-to-end robotic control, due to the strong generalization capabilities of Vision-Language Models (VLMs). A fundamental challenge in developing such models is effectively aligning the vision-language space with the robotic action space. Existing approaches typically rely on directly fine-tuning VLMs using expert demonstrations. However, this strategy suffers from a spatio-temporal gap, resulting in considerable data inefficiency and heavy reliance on human labor. Spatially, VLMs operate within a high-level semantic space, whereas robotic actions are grounded in low-level 3D physical space; temporally, VLMs primarily interpret the present, while VLA models anticipate future actions. To overcome these challenges, we propose a novel training paradigm, ROSA, which leverages robot state estimation to improve alignment between vision-language and action spaces. By integrating robot state estimation data obtained via an automated process, ROSA enables the VLA model to gain enhanced spatial understanding and self-awareness, thereby boosting performance and generalization. Extensive experiments in both simulated and real-world environments demonstrate the effectiveness of ROSA, particularly in low-data regimes.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
