论文/arXiv 2610.23123·12 天前

GRASP：面向双臂机器人操作的全局-局部表示对齐与空间-物理锚定

针对关节空间VLA训练目标与任务成功不一致，GRASP注入物理与视觉先验，并在仿真和真机长程双臂任务验证。

AI 整理

• 本体 SDK 与控制
• 仿真与训练
• 数据与采集

原文链接：GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation（arXiv 2610.23123 · 发布日期：2026-10-16）

GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation

针对关节空间VLA训练目标与任务成功不一致，GRASP注入物理与视觉先验，并在仿真和真机长程双臂任务验证。

核心要点

• 关节空间动作回归免去部署时逆解，但关节损失忽视关节误差的不同物理后果。
• 物理约束的 flow-matching 损失经可微 FK 监督 TCP 位姿和双臂相对位姿。
• 全局查询由冻结 DINOv3 蒸馏预测未来帧特征，局部查询用 Grounding DINO 监督注意力。
• 仿真和真机实验显示两类先验互补；分布内全模型更优，强视觉随机化下蒸馏查询更适合。

验证范围与局限：验证范围为仿真和真实机器人上的长程双臂操作任务。

英文摘要 (Abstract)

Vision-Language-Action (VLA) models that directly regress joint-space actions avoid the deployment-time cost of inverse kinematics. However, their training objectives remain misaligned with task success, as joint-space losses ignore the unequal physical consequences of joint errors and visual conditioning lacks explicit focus on hand-object interaction regions. To address these issues, we propose a training-phase prior injection framework for fine-grained bimanual manipulation. First, a physics-grounded flow-matching loss maps the implicit clean-action estimate at every noise level through differentiable forward kinematics (FK) and supervises the resulting Tool Center Point (TCP) pose and inter-arm relative pose, with a bounded noise-level weighting that compensates gradient decay. Second, we introduce a global prediction query group distilled from a frozen DINOv3 teacher through future-frame feature prediction, together with a local interaction query supervised by Grounding DINO at the attention level. Both teachers are removed at inference, and the deployment pipeline is that of a standard joint-space flow policy, adding only the learned query pathway. Simulation and real-robot experiments on long-horizon bimanual tasks show that the two priors are complementary across different deployment settings. The full model excels in in-distribution deployment, while the distilled query branch is preferable under strong visual randomization.

作者：Zeyuan Ren, Guanxun Zhao, Qi Li, Hao Wang, Wenlong Liao, Keyi Luo, Xijing Zhang, Zhiyuan Zhang, Jingye Cui, Qibin Gu

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.23123：GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation访问 2026-10-21