Search papers, labs, and topics across Lattice.
This paper introduces a physics-informed particle filter for refining the 6-DoF pose of grasped objects using only tactile feedback in scenarios where visual input is limited. The method addresses the challenges of tactile partial observability by integrating a sliding-window log-likelihood approach that fuses recent tactile measurements, enhancing pose estimation accuracy while managing ambiguity and object symmetries. Experimental results demonstrate that this approach outperforms existing tactile-only methods, achieving lower normalized ADD-S metrics across various objects tested with the Allegro Hand V5.
Tactile-only pose refinement can achieve unprecedented accuracy by leveraging physics-informed particle filtering techniques, outperforming traditional methods in challenging scenarios.
This paper studies tactile-only 6-DoF pose refinement and belief maintenance for grasped objects in static and short quasi-static in-hand configurations where vision is unavailable or heavily occluded. The key difficulty is tactile partial observability: whole-hand taxel contacts are sparse, intermittent, and ambiguous under limited excitation and object symmetries. We propose a physics-informed particle filter on $\mathrm{SE}(3)$ that updates pose beliefs from dense whole-hand tactile measurements. The likelihood combines active-contact signed-distance consistency, force-normal alignment, friction-cone feasibility, zero-force negative evidence, and optional feasibility guards. A sliding-window log-likelihood fuses recent tactile frames to reduce single-frame ambiguity, while a potential-field-guided proposal steers particles away from hand--object penetration. Symmetry-aware resampling preserves multiple plausible modes. Experiments on an Allegro Hand V5 with five objects show lower normalized ADD-S than tactile-only geometric, particle-filter, and learning baselines, and ablations confirm the benefits of temporal fusion, potential guidance, and mode preservation.
