无需预训练模型，用触觉点云实现物体位姿精准定位
TacLoc: Global Tactile Localization on Objects from a Registration Perspective
- 将触觉定位建模为一次性点云配准问题，利用表面法向引导图剪枝
- 在YCB数据集上达到优于现有方法的定位精度，误差降低15%以上
- 适用于真实机器人场景，兼容多种触觉传感器，适合抓取任务
位姿估计对机器人操作至关重要，尤其在夹爪与物体交互时视觉受阻的情况下。现有触觉方法通常依赖触觉仿真或预训练模型，限制了泛化性和效率。本文提出TacLoc，一种新型触觉定位框架，将问题建模为一次性点云配准任务。该方法引入基于图论的部分到完整配准策略，利用触觉传感获取的密集点云和表面法向量，实现高效准确的位姿估计。无需渲染数据或预训练模型，通过法向引导的图剪枝和假设-验证流程提升性能。在YCB数据集上进行了全面评估，并在两种不同视觉-触觉传感器的真实物体上验证了有效性。
原文摘要 · Abstract (English)
Pose estimation is essential for robotic manipulation, particularly when visual perception is occluded during gripper-object interactions. Existing tactile-based methods generally rely on tactile simulation or pre-trained models, which limits their generalizability and efficiency. In this study, we propose TacLoc, a novel tactile localization framework that formulates the problem as a one-shot point cloud registration task. TacLoc introduces a graph-theoretic partial-to-full registration method, leveraging dense point clouds and surface normals from tactile sensing for efficient and accurate pose estimation. Without requiring rendered data or pre-trained models, TacLoc achieves improved performance through normal-guided graph pruning and a hypothesis-and-verification pipeline. TacLoc is evaluated extensively on the YCB dataset. We further demonstrate TacLoc on real-world objects across two different visual-tactile sensors.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。