AnchorDP3：面向机器人操作的3D可供性引导稀疏扩散策略
AnchorDP3: 3D Affordance Guided Sparse Diffusion Policy for Robotic Manipulation
AI总结：
提出双臂机器人操作扩散策略框架AnchorDP3，通过语义分割、任务条件编码和可供性锚定关键位姿扩散三项创新，在RoboTwin基准测试中实现98.7%平均成功率，有望消除人类示范实现自主策略生成。
摘要：
我们提出了AnchorDP3，一个用于双臂机器人操作的扩散策略框架，在高度随机化的环境中达到了最先进的性能。AnchorDP3集成了三项关键创新：（1）仿真器监督的语义分割，使用渲染的真值在点云中显式分割任务关键物体，提供强大的可供性先验；（2）任务条件的特征编码器，轻量级模块处理每个任务的增强点云，通过共享的基于扩散的动作专家实现高效的多任务学习；（3）具有全状态监督的可供性锚定关键位姿扩散，用稀疏的、具有几何意义的动作锚点（即直接锚定到可供性的预抓取位姿、抓取位姿等关键位姿）替代密集的轨迹预测，极大简化了预测空间；动作专家被强制同时预测机器人关节角度和末端执行器位姿，利用几何一致性加速收敛并提升准确性。在大规模程序化生成的仿真数据上训练后，AnchorDP3在RoboTwin基准测试中，面对物体、杂乱度、桌面高度、光照和背景的极端随机化，跨多样化任务实现了98.7%的平均成功率。该框架与RoboTwin的虚实转换管线结合时，有望仅凭场景和指令实现完全可部署的视觉运动策略的自主生成，彻底消除学习操作技能中的人类示范。
英文摘要：
We present AnchorDP3, a diffusion policy framework for dual-arm robotic manipulation that achieves state-of-the-art performance in highly randomized environments. AnchorDP3 integrates three key innovations: (1) Simulator-Supervised Semantic Segmentation, using rendered ground truth to explicitly segment task-critical objects within the point cloud, which provides strong affordance priors; (2) Task-Conditioned Feature Encoders, lightweight modules processing augmented point clouds per task, enabling efficient multi-task learning through a shared diffusion-based action expert; (3) Affordance-Anchored Keypose Diffusion with Full State Supervision, replacing dense trajectory prediction with sparse, geometrically meaningful action anchors, i.e., keyposes such as pre-grasp pose, grasp pose directly anchored to affordances, drastically simplifying the prediction space; the action expert is forced to predict both robot joint angles and end-effector poses simultaneously, which exploits geometric consistency to accelerate convergence and boost accuracy. Trained on large-scale, procedurally generated simulation data, AnchorDP3 achieves a 98.7% average success rate in the RoboTwin benchmark across diverse tasks under extreme randomization of objects, clutter, table height, lighting, and backgrounds. This framework, when integrated with the RoboTwin real-to-sim pipeline, has the potential to enable fully autonomous generation of deployable visuomotor policies from only scene and instruction, totally eliminating human demonstrations from learning manipulation skills.
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。
