{"id":"arxiv:2610.39076","kind":"preprint","title":"Tactus: Tactile & Pose Sensing Glove for Natural Dexterity Data Collection","authors":["Li Guan","Rachael Granberry","Sahar Aseeri","Sahar Rostami","Taylor Niehues","Sai Dhanooshi Marri","Mia Huang","Dingtian Zhang","Ziyi Kou","Jason Pursel","Eric Whitmire","Vatsal Mehta","Ashish Deshpande","Tristan Trutna","Reed Belcher","Emelia Black","Julian Blake","Roger Boldu","Sean Bonaventure","Einar Boyovich","Jordan Brennan","Elizabeth Casillas","David Christensen","Tyler Christensen","Nicholas Colonnese"],"author_count":104,"year":2026,"published":"2026-10-29","venue":null,"cited_by":null,"doi":null,"arxiv_id":"2610.39076","nct_id":null,"source_url":"https://arxiv.org/abs/2610.39076","archive_url":"https://space-kitty.com/papers/arxiv:2610.39076","updated":"2026-10-29","pmid":null,"pmcid":null,"openalex_id":null,"pdf_url":"https://arxiv.org/pdf/2610.39076","abstract":"Collecting naturalistic human manipulation demonstrations with accurate contact force and hand pose is essential for robot dexterous manipulation policy and foundational world model learning - yet no existing wearable platform provides both calibrated fingertip force sensing and occlusion-resilient hand tracking in a form factor that preserves natural manipulation behavior. We present Tactus, a textile-integrated sensing glove that co-optimizes three design pillars: (1) a thin, conformal 3D-knit form factor (1.5 mm fingertip stackup) designed to minimize interference with natural manipulation, (2) strategically placed capacitive tactile sensors at manipulation-relevant regions informed by empirical contact studies, and (3) IMU-fused pose tracking that maintains accuracy through the hand-object occlusions where manipulation data is most valuable. Tactus includes a documented calibration pipeline for cross-glove consistency and per-session calibration for subject-dependent correction. To enable objective comparison, we introduce standardized benchmarking protocols for tactile force regression and hand pose estimation, and apply them to Tactus and two commercially available tactile gloves - to our knowledge the first quantitative cross-platform evaluation of this kind. Tactus resolves a minimum detectable force of 0.59 N at 95% confidence, 1.6 - 2.7x lower than the commercial baselines, and with 0.87 cross-participant consistency, while our pose tracking sustains 11.0 mm fingertip keypoint relative error during manipulation, when occlusion is heaviest. We show that these combined signals support object property estimation, slip detection, and representative recording of hand-object interaction - yielding demonstration data with the fidelity required for contact-aware manipulation policies and world models.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":true,"retracted":false,"peer_reviewed":false,"categories":["cs.RO","cs.AI","cs.AR","cs.LG"],"details":null,"provenance":"arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv.","topics":[{"slug":"robotics-embodied","name":"Robot Learning & Embodied AI","domain_slug":"ai"}],"cited_in_dossiers":[]}