{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:52:48Z","timestamp":1776279168025,"version":"3.50.1"},"reference-count":38,"publisher":"American Association for the Advancement of Science (AAAS)","issue":"113","funder":[{"DOI":"10.13039\/100018693","name":"Horizon Europe","doi-asserted-by":"publisher","award":["101070596"],"award-info":[{"award-number":["101070596"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"name":"EU project DARKO","award":["H2020 ICT-46-2020"],"award-info":[{"award-number":["H2020 ICT-46-2020"]}]},{"name":"EU project DARKO","award":["H2020 ICT-46-2020"],"award-info":[{"award-number":["H2020 ICT-46-2020"]}]}],"content-domain":{"domain":["www.science.org"],"crossmark-restriction":true},"short-container-title":["Sci. Robot."],"published-print":{"date-parts":[[2026,4,15]]},"abstract":"<jats:p>Teaching robots new skills should be as natural as showing rather than programming. Learning from demonstration (LfD) moves toward this goal by allowing users to guide a robot or sketch a desired motion, enabling learning without writing a line of code. However, most LfD methods remain tied to the robot they were trained on. Changes in morphology, different link lengths, joint orientations, or limits often break the learned behavior, making retraining unavoidable. Here, we introduce a framework that endows robots with kinematic intelligence: an internal understanding of their own joint limits, singularities, and connectivity. Instead of correcting for these constraints after learning, we embedded them directly into the control policy from the outset. The approach takes one or multiple demonstrations, extracts a globally stable dynamical system, and produces behaviors that remain valid across robots with different kinematic structures. Our method is grounded in a comprehensive analytical classification of noncuspidal three-revolute (3R) robots, which form the building blocks of many commercial robots. This classification enables a joint space policy that preserves user intent and adapts to robot-specific constraints. We validated the framework on diverse simulated and real robots, both redundant and nonredundant, with varied link geometries and joint configurations. The demonstrated skill executes safely and consistently across robots without retuning, thereby achieving cross-robot skill transfer.<\/jats:p>","DOI":"10.1126\/scirobotics.aea1995","type":"journal-article","created":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T17:58:12Z","timestamp":1776275892000},"update-policy":"https:\/\/doi.org\/10.34133\/aaas_crossmark","source":"Crossref","is-referenced-by-count":0,"title":["Demonstrate once, execute on many: Kinematic intelligence for cross-robot skill transfer"],"prefix":"10.1126","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4625-7963","authenticated-orcid":true,"given":"Sthithpragya","family":"Gupta","sequence":"first","affiliation":[{"name":"Learning Algorithms and Systems Laboratory, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne, Lausanne 1015, Switzerland."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9331-2390","authenticated-orcid":true,"given":"Durgesh Haribhau","family":"Salunkhe","sequence":"additional","affiliation":[{"name":"Learning Algorithms and Systems Laboratory, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne, Lausanne 1015, Switzerland."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7076-8010","authenticated-orcid":true,"given":"Aude","family":"Billard","sequence":"additional","affiliation":[{"name":"Learning Algorithms and Systems Laboratory, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne, Lausanne 1015, Switzerland."}]}],"member":"221","reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-100819-063206"},{"key":"e_1_3_2_3_2","doi-asserted-by":"crossref","unstructured":"E. Gribovskaya A. Billard \u201cCombining dynamical systems control and programming by demonstration for teaching discrete bimanual coordination tasks to a humanoid robot\u201d in 2008 3rd ACM\/IEEE International Conference on Human-Robot Interaction (HRI) (Association for Computing Machinery 2008) pp. 33\u201340.","DOI":"10.1145\/1349822.1349828"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/TOH.2013.54"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2015.2495003"},{"key":"e_1_3_2_6_2","doi-asserted-by":"crossref","unstructured":"A. Pervez A. Ali J.-H. Ryu D. Lee \u201cNovel learning from demonstration approach for repetitive teleoperation tasks\u201d in 2017 IEEE World Haptics Conference (WHC) (IEEE 2017) pp. 60\u201365.","DOI":"10.1109\/WHC.2017.7989877"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1177\/02783649241273565"},{"key":"e_1_3_2_8_2","doi-asserted-by":"crossref","unstructured":"R. R. Ma A. Spiers A. M. Dollar \u201cM2 gripper: Extending the dexterity of a simple underactuated gripper\u201d in Advances in Reconfigurable Mechanisms and Robots II X. Ding X. Kong J. S. Dai Eds. (Springer International Publishing 2016) pp. 795\u2013805.","DOI":"10.1007\/978-3-319-23327-7_68"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.robot.2008.10.024"},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSMCB.2006.886952"},{"key":"e_1_3_2_11_2","doi-asserted-by":"crossref","unstructured":"A. Shaw J. Lee J. Park \u201cConstrained dynamic movement primitives for collision avoidance in novel environments\u201d in 2023 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE 2023) pp. 3672\u20133679.","DOI":"10.1109\/IROS55552.2023.10341839"},{"key":"e_1_3_2_12_2","doi-asserted-by":"crossref","unstructured":"L. Yang M. Fischer D. Kragic \u201cEnhancing learning from demonstration with DLS-IK and ProMPs\u201d in 2024 29th International Conference on Automation and Computing (ICAC) (IEEE 2024) pp. 1\u20136.","DOI":"10.1109\/ICAC61394.2024.10718791"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1177\/02783649211040561"},{"key":"e_1_3_2_14_2","unstructured":"L. Bakker V. Koltun M. Toussaint \u201cTamedPUMA: Safe and stable imitation learning with geometric fabrics\u201d in Proceedings of the 7th Annual Learning for Dynamics and Control (L4DC) Conference (Proceedings of Machine Learning Research 2025)."},{"key":"e_1_3_2_15_2","unstructured":"B. Trabucco M. Phielipp G. Berseth \u201cAnyMorph: Learning transferable polices by inferring agent morphology\u201d in Proceedings of the 39th International Conference on Machine Learning (Proceedings of Machine Learning Research 2022) pp. 21677\u201321691."},{"key":"e_1_3_2_16_2","unstructured":"C. Sferrazza D.-M. Huang F. Liu J. Lee P. Abbeel \u201cBody transformer: Leveraging robot embodiment for policy learning\u201d in Proceedings of the 8th Annual Conference on Robot Learning (Proceedings of Machine Learning Research 2024) pp. 3407\u20133424."},{"key":"e_1_3_2_17_2","doi-asserted-by":"crossref","unstructured":"H. Klein N. Jaquier A. Meixner T. Asfour \u201cA Riemannian take on human motion analysis and retargeting\u201d in 2022 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE 2022) pp. 5210\u20135217.","DOI":"10.1109\/IROS47612.2022.9982127"},{"key":"e_1_3_2_18_2","doi-asserted-by":"crossref","unstructured":"N. Makondo B. Rosman O. Hasegawa \u201cKnowledge transfer for learning robot models via local procrustes analysis\u201d in 2015 IEEE-RAS 15th International Conference on Humanoid Robots (Humanoids) (IEEE 2015) pp. 1075\u20131082.","DOI":"10.1109\/HUMANOIDS.2015.7363502"},{"key":"e_1_3_2_19_2","doi-asserted-by":"crossref","unstructured":"P. Pastor H. Hoffmann T. Asfour S. Schaal \u201cLearning and generalization of motor skills by learning from demonstration\u201d in 2009 IEEE International Conference on Robotics and Automation (IEEE 2009) pp. 763\u2013768.","DOI":"10.1109\/ROBOT.2009.5152385"},{"key":"e_1_3_2_20_2","doi-asserted-by":"crossref","unstructured":"S. Schaal J. Peters J. Nakanishi A. Ijspeert \u201cLearning movement primitives\u201d in Robotics Research. The Eleventh International Symposium P. Dario R. Chatila Eds. (Springer 2005) pp. 561\u2013572.","DOI":"10.1007\/11008941_60"},{"key":"e_1_3_2_21_2","doi-asserted-by":"crossref","unstructured":"M. Ewerton G. Maeda G. Kollegger J. Wiemeyer J. Peters \u201cIncremental imitation learning of context-dependent motor skills\u201d in 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids) (IEEE 2016) pp. 351\u2013358.","DOI":"10.1109\/HUMANOIDS.2016.7803300"},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","unstructured":"D. H. Salunkhe D. Chablat P. Wenger \u201cTrajectory planning issues in cuspidal commercial robots\u201d in 2023 IEEE International Conference on Robotics and Automation (ICRA) (IEEE 2023) pp. 7426\u20137432.","DOI":"10.1109\/ICRA48891.2023.10161444"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1177\/02783649241293481"},{"key":"e_1_3_2_24_2","doi-asserted-by":"crossref","unstructured":"P. Wenger El J. Omri \u201cChanging posture for cuspidal robot manipulators\u201d in Proceedings of IEEE International Conference on Robotics and Automation (IEEE 1996) pp. 3173\u20133178.","DOI":"10.1109\/ROBOT.1996.509195"},{"key":"e_1_3_2_25_2","doi-asserted-by":"crossref","unstructured":"P. Wenger \u201cDesign of cuspidal and non-cuspidal robot manipulators\u201d in Proceedings of International Conference on Robotics and Automation (IEEE 1997) pp. 2172\u20132177.","DOI":"10.1109\/ROBOT.1997.619284"},{"key":"e_1_3_2_26_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2004.829467"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2025.3623011"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1016\/0094-114X(94)00043-K"},{"key":"e_1_3_2_29_2","doi-asserted-by":"crossref","unstructured":"P. Borrel A. Liegeois \u201cA study of multiple manipulator inverse kinematic solutions with applications to trajectory planning and workspace determination\u201d in Proceedings of the 1986 IEEE International Conference on Robotics and Automation (IEEE 1986) pp. 1180\u20131185.","DOI":"10.1109\/ROBOT.1986.1087554"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1115\/1.2826976"},{"key":"e_1_3_2_31_2","unstructured":"D. Paganelli \u201cTopological analysis of singularity loci for serial and parallel manipulators \u201d thesis Universita di Bologna Bologna Italy (2008)."},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mechmachtheory.2024.105884"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mechmachtheory.2024.105824"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2026.3666134"},{"key":"e_1_3_2_35_2","doi-asserted-by":"crossref","unstructured":"A. Dragan K. Lee S. Srinivasa \u201cLegibility and predictability of robot motion\u201d in Proceedings of the 8th ACM\/IEEE International Conference on Human-Robot Interaction HRI \u201913 (IEEE 2013) pp. 301\u2013308.","DOI":"10.1109\/HRI.2013.6483603"},{"key":"e_1_3_2_36_2","doi-asserted-by":"crossref","unstructured":"J. Geldenbott K. Leung \u201cLegible and proactive robot planning for prosocial human-robot interactions\u201d in 2024 IEEE International Conference on Robotics and Automation (ICRA) (IEEE 2024) pp.13397\u201313403.","DOI":"10.1109\/ICRA57147.2024.10611294"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1115\/1.4055677"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mechmachtheory.2022.104729"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2011.2159412"}],"container-title":["Science Robotics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.science.org\/doi\/pdf\/10.1126\/scirobotics.aea1995","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T17:58:21Z","timestamp":1776275901000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.science.org\/doi\/10.1126\/scirobotics.aea1995"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,15]]},"references-count":38,"journal-issue":{"issue":"113","published-print":{"date-parts":[[2026,4,15]]}},"alternative-id":["10.1126\/scirobotics.aea1995"],"URL":"https:\/\/doi.org\/10.1126\/scirobotics.aea1995","relation":{},"ISSN":["2470-9476"],"issn-type":[{"value":"2470-9476","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,15]]},"assertion":[{"value":"2025-06-30","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-17","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-15","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"eaea1995"}}