Co-VLA：面向双臂视觉-语言-动作系统的协调感知结构化动作建模
Co-VLA: Coordination-Aware Structured Action Modeling for Dual-Arm Vision-Language-Action Systems
AI总结：
针对双臂紧耦合任务中隐式协调不足的问题，提出Co-VLA框架，通过结构化动作专家和潜在感知控制器显式引入协调先验，在仿真和真实场景中显著提升成功率和效率。
摘要：
视觉-语言-动作（VLA）模型在单臂和双臂机器人操作中展现出强大能力。先前研究表明，通过端到端学习，利用大型视觉-语言骨干网络和连续动作预测，可以涌现出协调的双臂行为。然而，随着双臂任务变得紧密耦合且执行约束变得关键，仅靠隐式协调不足以确保可靠、可解释且稳定的行为。在这项工作中，我们提出了Co-VLA，一个协调感知的双臂操作框架，将显式结构先验引入VLA模型。我们在一个最先进的视觉-语言骨干网络上实例化我们的方法，用专为双臂协调设计的结构化动作专家（SAE）替换其单一动作头。具体来说，我们在动作生成层面引入显式结构，采用模块化的协调感知损失，根据任务特定结构塑造共享和残差潜在变量。共享潜在变量编码任务级协调意图，而残差潜在变量捕获每个手臂的执行调整。在部署时，潜在感知控制器（LAC）解释学习到的表示，以实时调节同步强度、执行不对称性、平滑性和安全约束。LAC在关节命令级别运行，并与标准控制流水线兼容，无需力或阻抗控制。在仿真和真实世界基准上的实验表明，Co-VLA显著优于单一基线，在紧协调任务中成功率达到27%的提升，在OOD真实世界场景中性能翻倍（从13%提升至27%），并将任务完成时间减少高达25%。
英文摘要：
Vision-language-action (VLA) models show strong capabilities in single and dual-arm robotic manipulation. Prior works show coordinated bimanual behaviors can emerge from end-to-end learning, leveraging large vision-language backbones with continuous action prediction. However, as bimanual tasks become tightly coupled and execution constraints become critical, implicit coordination alone is insufficient to ensure reliable, interpretable, and stable behavior. In this work, we propose Co-VLA, a coordination-aware bimanual manipulation framework introducing explicit structural priors into VLA models. We instantiate our method on a state-of-the-art vision-language backbone by replacing its monolithic action head with a Structured Action Expert (SAE) designed for bimanual coordination. Specifically, we introduce explicit structure at the action generation level with a modular coordination-aware loss that shapes shared and residual latents according to task-specific structures. The shared latent encodes task-level coordination intent, while residual latents capture execution adjustments for each arm. At deployment, a Latent-Aware Controller (LAC) interprets the learned representations to modulate synchronization strength, execution asymmetry, smoothness, and safety constraints in real time. LAC operates at the joint-command level and remains compatible with standard control pipelines without requiring force or impedance control. Experiments across simulation and real-world benchmarks show Co-VLA significantly outperforms monolithic baselines, achieving a 27% success rate gain in tight-coordination tasks, more than doubling performance in OOD real-world scenarios (from 13% to 27%), and reducing task completion time by up to 25%.
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。
