{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T08:01:03Z","timestamp":1771920063162,"version":"3.50.1"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T00:00:00Z","timestamp":1771891200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Robot. AI"],"abstract":"<jats:p>Achieving robust, dexterous manipulation in unstructured environments remains a central challenge in robotics, particularly for continuous, contact-rich tasks like cleaning. While motion primitives can also be learned directly in full joint space, a compact, synergy-based representation provides a shared latent coordinate system that simplifies interpretation, modulation, and cross-task composition. We adopt a data-driven framework for representing and reproducing dexterous manipulation trajectories, using cleaning motions as a test bed. To model these movements, we combine Principal Component Analysis (PCA) with Probabilistic Movement Primitives (ProMPs), leveraging hand synergies. While the PCA and ProMP combination itself is established, our focus in this study, is on the cleaning use case and on the compositional generalization across tasks. PCA, applied in joint space, provides a compact, low-dimensional synergy space for coordinated finger movements, while the ProMPs encode the time-varying structure and variability of trajectories within this space. We first recorded a kinematic dataset of human cleaning motions with 20 degrees of freedom (DOF) haptic exoskeleton gloves across thirteen tasks and learn one ProMP per five selected training tasks in the PCA space. This dataset is then used as a basis to learn cleaning motions using the PCA + ProMPs. We demonstrate the ability of the learned primitives to reconstruct and reproduce kinematic patterns in simulation (Shadow Hand) and successfully deploy them on a physical robotic hand (Aeon Robotics). These results indicate that motion primitives, when grounded in synergy-informed coordinates, can generalize beyond grasping to encode and modulate contact-rich dexterous manipulation skills. Moreover, a library of the five task-specific ProMPs compositionally approximates trajectories from eight unseen cleaning tasks, with nearest-expert selection outperforming convex blends and Product-of-Experts combinations.<\/jats:p>","DOI":"10.3389\/frobt.2026.1725261","type":"journal-article","created":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T05:19:20Z","timestamp":1771910360000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Generalization of finger-joint kinematics for cleaning tasks"],"prefix":"10.3389","volume":"13","author":[{"given":"Clara","family":"Pham","sequence":"first","affiliation":[{"name":"Institute for Robotics and Process Control, Technische Universit\u00e4t Braunschweig","place":["Braunschweig, Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jan-Philipp","family":"Tauscher","sequence":"additional","affiliation":[{"name":"Institute for Computer Graphics, Technische Universit\u00e4t Braunschweig","place":["Braunschweig, Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Colin","family":"Groth","sequence":"additional","affiliation":[{"name":"Immersive Computing Lab, New York University Tandon School of Engineering","place":["New York, NY, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jochen J.","family":"Steil","sequence":"additional","affiliation":[{"name":"Institute for Robotics and Process Control, Technische Universit\u00e4t Braunschweig","place":["Braunschweig, Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,24]]},"reference":[{"key":"B1","first-page":"2043","article-title":"Generalization of human grasping for multi-fingered robot hands","volume-title":"Proceedings of the 2012 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Amor","year":"2012"},{"key":"B2","doi-asserted-by":"crossref","DOI":"10.1109\/ICHR.2009.5379530","article-title":"Latent spaces for dynamic movement primitives","volume-title":"Proceedings of the IEEE-RAS International Conference on Humanoid Robots (Humanoids)","author":"Bitzer","year":"2009"},{"key":"B3","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2019.00891","article-title":"Contactdb: analyzing and predicting grasp contact via thermal imaging","volume-title":"IEEE\/CVF Conference On Computer Vision and Pattern Recognition (CVPR)","author":"Brahmbhatt","year":"2019"},{"key":"B4","doi-asserted-by":"crossref","DOI":"10.1109\/IROS.2013.6696414","article-title":"On improving the extrapolation capability of task-parameterized movement models","volume-title":"Proceedings of the IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Calinon","year":"2013"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1299","DOI":"10.1017\/s0263574720001186","article-title":"Motion adaptation based on learning the manifold of task and dynamic movement primitive parameters","volume":"39","author":"Cohen","year":"2021","journal-title":"Robotica"},{"key":"B6","doi-asserted-by":"publisher","first-page":"602","DOI":"10.1109\/TRO.2018.2808924","article-title":"Dimensionality reduction for dynamic movement primitives and application to bimanual manipulation of clothes","volume":"34","author":"Colom\u00e9","year":"2018","journal-title":"IEEE Trans. Robotics"},{"key":"B7","doi-asserted-by":"crossref","first-page":"794","DOI":"10.1109\/HUMANOIDS.2014.7041454","article-title":"Dimensionality reduction for probabilistic movement primitives","volume-title":"2014 IEEE-RAS International Conference on Humanoid Robots (IEEE)","author":"Colom\u00e9","year":"2014"},{"key":"B8","doi-asserted-by":"publisher","first-page":"193","DOI":"10.3389\/fbioe.2019.00193","article-title":"Principal component analysis reveals the proximal to distal pattern in vertical jumping is governed by two functional degrees of freedom","volume":"7","author":"Cushion","year":"2019","journal-title":"Front. Bioeng. Biotechnol."},{"key":"B9","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1016\/j.clinbiomech.2004.01.005","article-title":"Pca in studying coordination and variability: a tutorial","volume":"19","author":"Daffertshofer","year":"2004","journal-title":"Clin. Biomech."},{"key":"B10","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1016\/j.gaitpost.2006.01.007","article-title":"Biomechanical features of gait waveform data associated with knee osteoarthritis: an application of principal component analysis","volume":"25","author":"Deluzio","year":"2007","journal-title":"Gait and Posture"},{"key":"B11","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52729.2023.01244","article-title":"Arctic: a dataset for dexterous bimanual hand-object manipulation","volume-title":"IEEE\/CVF conference on computer vision and pattern recognition (CVPR)","author":"Fan","year":"2023"},{"key":"B12","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1109\/thms.2015.2470657","article-title":"The grasp taxonomy of human grasp types","volume":"46","author":"Feix","year":"2015","journal-title":"IEEE Trans. Human-Machine Syst."},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01627","article-title":"Gigahands: a massive annotated dataset of bimanual hand activities","author":"Fu","year":"2025","journal-title":"Prepr. CVPR 2025 Highlight"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2403.00929","article-title":"Prime: scaffolding manipulation tasks with behavior primitives for data-efficient imitation learning","author":"Gao","year":"2024","journal-title":"IEEE Robotics Automation Lett."},{"key":"B15","doi-asserted-by":"publisher","first-page":"825","DOI":"10.1109\/tro.2013.2252251","article-title":"Mapping synergies from human to robotic hands with dissimilar kinematics: an approach in the object domain","volume":"29","author":"Gioioso","year":"2013","journal-title":"IEEE Trans. Robotics"},{"key":"B16","doi-asserted-by":"publisher","first-page":"904","DOI":"10.1109\/70.544773","article-title":"On the stability of grasped objects","volume":"12","author":"Howard","year":"1996","journal-title":"IEEE Trans. Robotics Automation"},{"key":"B17","article-title":"The representation of objects for grasping","volume":"8","author":"Iberall","year":"1986","journal-title":"Proc. Annu. Meet. Cognitive Sci. Soc."},{"key":"B18","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1162\/NECO_a_00393","article-title":"Dynamical movement primitives: learning attractor models for motor behaviors","volume":"25","author":"Ijspeert","year":"2013","journal-title":"Neural Comput."},{"key":"B19","doi-asserted-by":"crossref","first-page":"1094","DOI":"10.1007\/978-3-642-04898-2_455","article-title":"Principal component analysis","volume-title":"International Encyclopedia of Statistical Science","author":"Jolliffe","year":"2011"},{"key":"B20","doi-asserted-by":"publisher","first-page":"20150202","DOI":"10.1098\/rsta.2015.0202","article-title":"Principal component analysis: a review and recent developments","volume":"374","author":"Jolliffe","year":"2016","journal-title":"Philosophical Transactions Royal Society A Math. Phys. Eng. Sci."},{"key":"B21","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1007\/s10846-019-01072-4","article-title":"Cleaning tasks knowledge transfer between heterogeneous robots: a deep learning approach","volume":"98","author":"Kim","year":"2020","journal-title":"J. Intelligent and Robotic Syst."},{"key":"B23","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1007\/s10846-014-0100-3","article-title":"Model predictive motion control based on generalized dynamical movement primitives","volume":"77","author":"Krug","year":"2015","journal-title":"J. Intelligent and Robotic Syst."},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1109\/HUMANOIDS47582.2021.9555788","article-title":"The kit bimanual manipulation dataset","author":"Krebs","year":"2021","journal-title":"Tech. Report, KIT Whole-Body Hum. Motion Database"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2647","DOI":"10.1109\/tcsvt.2021.3057992","article-title":"Survey on mapping human hand motion to robotic hands for teleoperation","volume":"32","author":"Li","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"B24","volume-title":"Robotic Table Wiping via Reinforcement Learning and Whole-Body Trajectory Optimization","author":"Lew","year":"2022"},{"key":"B26","volume-title":"Cleanupbench: Embodied Sweeping and Grasping Benchmark","author":"Li","year":"2025"},{"key":"B27","doi-asserted-by":"crossref","first-page":"2075","DOI":"10.1109\/IROS.2013.6696638","article-title":"Mapping human to robot motion with functional anthropomorphism for teleoperation and telemanipulation with robot arm hand systems","volume-title":"2013 IEEE\/RSJ International Conference on Intelligent Robots and Systems","author":"Liarokapis","year":"2013"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1152\/jn.2001.86.6.2896","article-title":"Hand synergies during reach-to-grasp","author":"Mason","year":"2001","journal-title":"J. Neurophysiology"},{"key":"B29","doi-asserted-by":"crossref","DOI":"10.1109\/IROS.2010.5651049","article-title":"Learning stylistic dynamic movement primitives from multiple demonstrations","volume-title":"IROS","author":"Matsubara","year":"2010"},{"key":"B30","article-title":"Probabilistic movement primitives","volume":"26","author":"Paraschos","year":"2013","journal-title":"Adv. Neural Inf. Process. Syst. (NeurIPS)"},{"key":"B31","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1007\/s10514-017-9648-7","article-title":"Using probabilistic movement primitives in robotics","volume":"42","author":"Paraschos","year":"","journal-title":"Aut. Robots"},{"key":"B32","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1080\/01691864.2018.1437674","article-title":"Probabilistic movement primitives under unknown system dynamics","volume":"32","author":"Paraschos","year":"","journal-title":"Adv. Robot."},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.4172\/2155-9538.1000163","article-title":"Linear and nonlinear kinematic synergies in the grasping hand","volume":"5","author":"Patel","year":"2015","journal-title":"J. Bioeng. and Biomed. Sci."},{"key":"B34","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1016\/j.robot.2017.03.017","article-title":"Learning adaptive dressing assistance from human demonstration","volume":"93","author":"Pignat","year":"2017","journal-title":"Robotics Aut. Syst."},{"key":"B35","doi-asserted-by":"crossref","DOI":"10.1109\/ICRA.2015.7139390","article-title":"Extracting low-dimensional control variables for movement primitives","volume-title":"ICRA","author":"Rueckert","year":"2015"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1007\/s002210000420","article-title":"Force synergies for multifingered grasping","author":"Santello","year":"2000","journal-title":"Exp. Brain Res."},{"key":"B37","doi-asserted-by":"publisher","first-page":"10105","DOI":"10.1523\/JNEUROSCI.18-23-10105.1998","article-title":"Postural hand synergies for tool use","volume":"18","author":"Santello","year":"1998","journal-title":"J. Neuroscience"},{"key":"B38","doi-asserted-by":"publisher","first-page":"180101","DOI":"10.1038\/sdata.2018.101","article-title":"Human grasping database for activities of daily living with depth, color and kinematic data streams","volume":"5","author":"Saudabayev","year":"2018","journal-title":"Sci. Data"},{"key":"B39","doi-asserted-by":"publisher","first-page":"1133","DOI":"10.1177\/02783649231201196","article-title":"Dynamic movement primitives in robotics: a tutorial survey","volume":"42","author":"Saveriano","year":"","journal-title":"Int. J. Robotics Res."},{"key":"B40","doi-asserted-by":"publisher","first-page":"104510","DOI":"10.1016\/j.robot.2023.104510","article-title":"Learning stable robotic skills on riemannian manifolds","volume":"169","author":"Saveriano","year":"","journal-title":"Robotics Aut. Syst."},{"key":"B41","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1007\/4-431-31381-8_23","article-title":"Dynamic movement primitives\u2013a framework for motor control in humans and humanoid robotics","volume-title":"Adaptive Motion of Animals and Machines","author":"Schaal","year":"2006"},{"key":"B42","article-title":"Senseglove_ros: ros interfaces for senseglove","volume-title":"GitHub organization: Adjuvo. Maintainer: Akshay R. Menon. Commit USED: e5e9a58","author":"SenseGlove","year":"2025"},{"key":"B43","first-page":"4119","article-title":"Kinematic synergy primitives for human-like grasp motion generation","volume-title":"Proceedings of the 2024 IEEE International Conference on Robotics and Automation (ICRA)","author":"Starke","year":"2024"},{"key":"B44","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-58548-8_34","article-title":"Grab: a dataset of whole-body human grasping of objects","volume-title":"European Conference on Computer Vision (ECCV)","author":"Taheri","year":"2020"},{"key":"B45","first-page":"13","article-title":"Dimensionality reduction: a comparative review","volume":"10","author":"van der Maaten","year":"2009","journal-title":"J. Machine Learning Research"},{"key":"B46","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52733.2024.00050","article-title":"Oakink2: a dataset of bimanual hands-object manipulation in complex task completion","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Zhan","year":"2024"},{"key":"B47","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/TCDS.2018.2868921","article-title":"Cooperative manipulation for a mobile dual-arm robot using sequences of dynamic movement primitives","volume":"12","author":"Zhao","year":"2020","journal-title":"IEEE Trans. Cognitive Dev. Syst."}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1725261\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T05:19:25Z","timestamp":1771910365000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1725261\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,24]]},"references-count":47,"alternative-id":["10.3389\/frobt.2026.1725261"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2026.1725261","relation":{},"ISSN":["2296-9144"],"issn-type":[{"value":"2296-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,24]]},"article-number":"1725261"}}