{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T08:17:34Z","timestamp":1781079454729,"version":"3.54.1"},"reference-count":27,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,11,9]],"date-time":"2023-11-09T00:00:00Z","timestamp":1699488000000},"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. Neurorobot."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Behavioral Cloning (BC) is a common imitation learning method which utilizes neural networks to approximate the demonstration action samples for task manipulation skill learning. However, in the real world, the demonstration trajectories from human are often sparse and imperfect, which makes it challenging to comprehensively learn directly from the demonstration action samples. Therefore, in this paper, we proposes a streamlined imitation learning method under the terse geometric representation to take good advantage of the demonstration data, and then realize the manipulation skill learning of assembly tasks.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We map the demonstration trajectories into the geometric feature space. Then we align the demonstration trajectories by Dynamic Time Warping (DTW) method to get the unified data sequence so we can segment them into several time stages. The Probability Movement Primitives (ProMPs) of the demonstration trajectories are then extracted, so we can generate a lot of task trajectories to be the global strategy action samples for training the neural networks. Notalby, we regard the current state of the assembly task as the via point of the ProMPs model to get the generated trajectories, while the time point of the via point is calculated according to the probability model of the different time stages. And we get the action of the current state according to the target position of the next time state. Finally, we train the neural network to obtain the global assembly strategy by Behavioral Cloning.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We applied the proposed method to the peg-in-hole assembly task in the simulation environment based on Pybullet + Gym to test its task skill learning performance. And the learned assembly strategy was also executed on a real robotic platform to verify the feasibility of the method further.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>According to the result of the experiment, the proposed method achieves higher success rates compared to traditional imitation learning methods while exhibiting reasonable generalization capabilities. It shows that the ProMPs under geometric representation can help the BC method make better use of the demonstration trajectory and thus better learn the task skills.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fnbot.2023.1320251","type":"journal-article","created":{"date-parts":[[2023,11,13]],"date-time":"2023-11-13T06:18:35Z","timestamp":1699856315000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":10,"title":["Peg-in-hole assembly skill imitation learning method based on ProMPs under task geometric representation"],"prefix":"10.3389","volume":"17","author":[{"given":"Yajing","family":"Zang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pengfei","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fusheng","family":"Zha","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Guo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chao","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lining","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2023,11,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2874","DOI":"10.1109\/TITS.2022.3227738","article-title":"Modeling human driving behavior through generative adversarial imitation learning","volume":"24","author":"Bhattacharyya","year":"2023","journal-title":"IEEE Trans. Intell. Transp. Syst"},{"key":"B2","doi-asserted-by":"publisher","first-page":"7863","DOI":"10.1109\/TNNLS.2021.3088947","article-title":"Complex robotic manipulation via graph-based hindsight goal generation","volume":"33","author":"Bing","year":"","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B3","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1109\/MRA.2022.3204237","article-title":"Simulation to real: learning energy-efficient slithering gaits for a snake-like robot","volume":"29","author":"Bing","year":"","journal-title":"IEEE Robot. Autom. Mag"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TNNLS.2023.3270298","article-title":"Meta-reinforcement learning in nonstationary and nonparametric environments","author":"Bing","year":"","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B5","doi-asserted-by":"publisher","first-page":"3476","DOI":"10.1109\/TPAMI.2022.3185549","article-title":"Meta-reinforcement learning in non-stationary and dynamic environments","volume":"45","author":"Bing","year":"","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B6","doi-asserted-by":"publisher","first-page":"2759","DOI":"10.1109\/TIE.2022.3172754","article-title":"Solving robotic manipulation with sparse reward reinforcement learning via graph-based diversity and proximity","volume":"70","author":"Bing","year":"","journal-title":"IEEE Trans. Ind. Electron"},{"key":"B7","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1109\/ICARM.2017.8273217","article-title":"\u201cRobot learning from multiple demonstrations with dynamic movement primitive\u201d","volume-title":"2017 2nd International Conference on Advanced Robotics and Mechatronics (ICARM)","author":"Chen","year":"2017"},{"key":"B8","doi-asserted-by":"crossref","first-page":"5862","DOI":"10.1109\/IROS40897.2019.8968282","article-title":"\u201cTransferable trial-minimizing progressive peg-in-hole model,\u201d","volume-title":"2019 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Ding","year":"2019"},{"key":"B9","doi-asserted-by":"crossref","first-page":"7821","DOI":"10.1109\/ICRA.2019.8793780","article-title":"\u201cImitating human search strategies for assembly,\u201d","volume-title":"2019 International Conference on Robotics and Automation (ICRA)","author":"Ehlers","year":"2019"},{"key":"B10","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1177\/0278364919846363","article-title":"Kernelized movement primitives","volume":"38","author":"Huang","year":"2019","journal-title":"Int. J. Rob. Res"},{"key":"B11","doi-asserted-by":"publisher","first-page":"189","DOI":"10.1108\/AA-10-2018-0164","article-title":"Peg-in-hole assembly based on master-slave coordination for a compliant dual-arm robot","volume":"40","author":"Huang","year":"2020","journal-title":"Assem. Autom"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1016\/j.rcim.2019.101863","article-title":"Reinforcement learning based on movement primitives for contact tasks","author":"Kim","year":"2020","journal-title":"Robot. Comput. Integr. Manuf"},{"key":"B13","doi-asserted-by":"publisher","first-page":"582","DOI":"10.1109\/TRO.2019.2959445","article-title":"Making sense of vision and touch: learning multimodal representations for contact-rich tasks","volume":"36","author":"Lee","year":"2020","journal-title":"IEEE Trans. Robot"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1025","DOI":"10.1109\/TIV.2022.3198678","article-title":"Driver behavioral cloning for route following in autonomous vehicles using task knowledge distillation","volume":"8","author":"Li","year":"2023","journal-title":"IEEE Trans. Intell. Veh"},{"key":"B15","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1016\/j.neucom.2021.12.043","article-title":"Supervised assisted deep reinforcement learning for emergency voltage control of power systems","volume":"475","author":"Li","year":"2022","journal-title":"Neurocomputing"},{"key":"B16","doi-asserted-by":"publisher","first-page":"895","DOI":"10.1108\/AA-04-2020-0053","article-title":"Learning peg-in-hole assembly using cartesian dmps with feedback mechanism","volume":"40","author":"Liu","year":"2020","journal-title":"Assem. Autom"},{"key":"B17","doi-asserted-by":"crossref","first-page":"3080","DOI":"10.1109\/ICRA.2019.8793506","article-title":"\u201cReinforcement learning on variable impedance controller for high-precision robotic assembly,\u201d","volume-title":"2019 International Conference on Robotics and Automation (ICRA)","author":"Luo","year":"2019"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1573","DOI":"10.1007\/s10514-017-9648-7","article-title":"Using probabilistic movement primitives in robotics","volume":"42","author":"Paraschos","year":"2018","journal-title":"Auton. Robots"},{"key":"B19","doi-asserted-by":"publisher","first-page":"6299","DOI":"10.1109\/TIE.2017.2682002","article-title":"Compliance-based robotic peg-in-hole assembly strategy without force feedback","volume":"64","author":"Park","year":"2017","journal-title":"IEEE Trans. Ind. Electron"},{"key":"B20","doi-asserted-by":"crossref","first-page":"1024","DOI":"10.1109\/IROS.2015.7353496","article-title":"\u201cLearning optimal controllers in human-robot cooperative transportation tasks with position and force constraints,\u201d","volume-title":"2015 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Rozo","year":"2015"},{"key":"B21","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1109\/TRO.2016.2540623","article-title":"Learning physical collaborative robot behaviors from human demonstrations","volume":"32","author":"Rozo","year":"2016","journal-title":"IEEE Trans. Robot"},{"key":"B22","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1109\/TRO.2018.2878355","article-title":"Learning task priorities from demonstrations","volume":"35","author":"Silv\u00e9rio","year":"2019","journal-title":"IEEE Trans. Robot"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.robot.2022.104264","article-title":"Goal-aware generative adversarial imitation learning from imperfect demonstration for robotic cloth manipulation","author":"Tsurumine","year":"2022","journal-title":"Rob. Auton. Syst"},{"key":"B24","first-page":"84","article-title":"\u201c3d analysis of contact in peg-hole insertion,\u201d","volume-title":"Proceedings of the 1997 IEEE International Symposium on Assembly and Task Planning (ISATP'97) - Towards Flexible and Agile Assembly and Manufacturing","author":"Tsuruoka","year":"1997"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1153","DOI":"10.1109\/TII.2018.2826064","article-title":"A learning framework of adaptive manipulative skills from human to robot","volume":"15","author":"Yang","year":"2019","journal-title":"IEEE Trans Ind. Inform"},{"key":"B26","doi-asserted-by":"publisher","first-page":"3478","DOI":"10.1109\/LRA.2023.3261759","article-title":"Geometric-feature representation based pre-training method for reinforcement learning of peg-in-hole tasks","volume":"8","author":"Zang","year":"2023","journal-title":"IEEE Robot. Autom. Lett"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIE.2023.3299051","article-title":"A learning-based two-stage method for submillimeter insertion tasks with only visual inputs","author":"Zhao","year":"2023","journal-title":"IEEE Trans. Ind. Electr"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1320251\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,13]],"date-time":"2023-11-13T06:18:41Z","timestamp":1699856321000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1320251\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,9]]},"references-count":27,"alternative-id":["10.3389\/fnbot.2023.1320251"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2023.1320251","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,9]]},"article-number":"1320251"}}