HiTac-WAM：一种用于接触丰富型机器人操作的分层触觉世界动作模型
HiTac-WAM: A Hierarchical Tactile World Action Model for Contact-Rich Robot Manipulation
- Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）
- ImprintX Robotics（ImprintX机器人公司）
- Beijing Academy of Artificial Intelligence（北京人工智能研究院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出HiTac-WAM这一分层触觉世界动作模型，通过分层结构建模触觉状态的物理依赖关系，在机器人接触操作任务中提升了预测精度与操作成功率。
AI中文摘要：
世界动作模型可联合预测未来的视觉观测与动作，而现有的触觉感知变体通常将未来触觉表示为图像或潜在流，未对组织触觉状态的物理依赖关系进行分层建模。本文提出HiTac-WAM，一种分层触觉世界动作模型，可在执行前为每个候选动作块预测一系列未来触觉状态。该预测分解为接触状态、3D变形场和滑移风险，以有向分层结构组织，其中每个下游阶段都以来自前序阶段的停止梯度信号为条件。有向注意力掩码允许触觉查询关注每个候选的视频-动作上下文，同时防止视频和动作查询关注触觉令牌。在规划阶段，HiTac-WAM利用触觉预测和任务进度估计对候选动作块进行排名；在执行阶段，保留选定的触觉预测作为参考，当预测与观测到的触觉状态存在持续差异时，触发修正重规划。HiTac-WAM实现了0.921的平均接触F1值；在匹配的训练预算下，与仅变形预测器相比，该有向分层结构将3D位移L2误差降低了17.6%，与仅滑移预测器相比，将滑移AUPRC提高了60.4%。在芯片抓取、黑板擦除和USB插入任务中，由分层预测引导的选择将平均真实机器人成功率从31.1%提升至61.1%，而完整系统的成功率达到72.2%。
英文摘要：
World action models jointly predict future visual observations and actions, whereas existing tactile-aware variants typically represent future touch as an image or latent stream without modeling the physical dependencies that organize tactile states hierarchically. We present HiTac-WAM, a hierarchical tactile world action model that forecasts a sequence of future tactile states for each candidate action chunk before execution. The forecast factorizes into contact state, a 3D deformation field, and slip risk, organized as a directed hierarchy in which each downstream stage is conditioned on stop-gradient signals from preceding stages. A directed attention mask allows tactile queries to attend to the video-action context of each candidate while preventing video and action queries from attending to tactile tokens. For planning, HiTac-WAM ranks candidate action chunks using tactile forecasts and task-progress estimates. For execution, the selected tactile forecast is retained as a reference; persistent discrepancies between predicted and observed tactile states trigger corrective replanning. HiTac-WAM achieves a mean contact F1 of 0.921; under matched training budgets, the directed hierarchy reduces 3D displacement L2 error by 17.6% relative to the deformation-only predictor and improves slip AUPRC by 60.4% relative to the slip-only predictor. Across chip grasping, blackboard erasing, and USB insertion, selection guided by the hierarchical forecasts increases the average real-robot success rate from 31.1% to 61.1%, while the full system attains 72.2%.
