{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T10:02:14Z","timestamp":1777716134147,"version":"3.51.4"},"reference-count":78,"publisher":"SAGE Publications","issue":"13","license":[{"start":{"date-parts":[[2025,5,15]],"date-time":"2025-05-15T00:00:00Z","timestamp":1747267200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"name":"Marie Sk\u0142odowska-Curie Actions-Doctoral Networks","award":["101119614"],"award-info":[{"award-number":["101119614"]}]},{"name":"Fellowship from Vidyasirimedhi Institute of Science and Technology"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of Robotics Research"],"published-print":{"date-parts":[[2025,11]]},"abstract":"<jats:p>Continual locomotion learning faces four challenges: incomprehensibility, sample inefficiency, lack of knowledge exploitation, and catastrophic forgetting. Thus, this work introduces growable online locomotion learning under multicondition (GOLLUM), which exploits the interpretability feature to address the aforementioned challenges. GOLLUM has two dimensions of interpretability: layer-wise interpretability for neural control function encoding and column-wise interpretability for robot skill encoding. With this interpretable control structure, GOLLUM utilizes neurogenesis to unsupervisely increment columns (ring-like networks); each column is trained separately to encode and maintain a specific primary robot skill. GOLLUM also transfers the parameters to new skills and supplements the learned combination of acquired skills through another neural mapping layer added (layer-wise) with online supplementary learning. On a physical hexapod robot, GOLLUM successfully acquired multiple locomotion skills (e.g., walking, slope climbing, and bouncing) autonomously and continuously within an hour using a simple reward function. Furthermore, it demonstrated the capability of combining previous learned skills to facilitate the learning process of new skills while preventing catastrophic forgetting. Compared to state-of-the-art locomotion learning approaches, GOLLUM is the only approach that addresses the four challenges above mentioned without human intervention. It also emphasizes the potential exploitation of interpretability to achieve autonomous lifelong learning machines.<\/jats:p>","DOI":"10.1177\/02783649251336385","type":"journal-article","created":{"date-parts":[[2025,5,15]],"date-time":"2025-05-15T14:50:52Z","timestamp":1747320652000},"page":"2156-2180","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":2,"title":["Growable and interpretable neural control with online continual learning for autonomous lifelong locomotion learning machines"],"prefix":"10.1177","volume":"44","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7997-745X","authenticated-orcid":false,"given":"Arthicha","family":"Srisuchinnawong","sequence":"first","affiliation":[{"name":"The University of Southern Denmark"},{"name":"Vidyasirimedhi Institute of Science and Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Poramate","family":"Manoonpong","sequence":"additional","affiliation":[{"name":"The University of Southern Denmark"},{"name":"Vidyasirimedhi Institute of Science and Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2025,5,15]]},"reference":[{"key":"e_1_3_5_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2005.04.026"},{"key":"e_1_3_5_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2019.12.012"},{"key":"e_1_3_5_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10846-020-01162-8"},{"key":"e_1_3_5_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3218167"},{"key":"e_1_3_5_6_1","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.ade2256"},{"key":"e_1_3_5_7_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1004402"},{"key":"e_1_3_5_8_1","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.5605-05.2006"},{"key":"e_1_3_5_9_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14422"},{"key":"e_1_3_5_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCDS.2023.3250393"},{"key":"e_1_3_5_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jksuci.2022.05.001"},{"key":"e_1_3_5_12_1","doi-asserted-by":"publisher","DOI":"10.1609\/aimag.v40i2.2850"},{"key":"e_1_3_5_13_1","doi-asserted-by":"crossref","unstructured":"Feber J Szadkowski R Faigl J (2022) Gait adaptation after leg amputation of hexapod walking robot without sensory feedback. In: Artificial neural networks and machine learning\u2013ICANN 2022: 31st international conference on artificial neural networks proceedings part III Bristol UK 6\u20139 September 2022 656\u2013667. Springer.","DOI":"10.1007\/978-3-031-15934-3_54"},{"key":"e_1_3_5_14_1","doi-asserted-by":"publisher","DOI":"10.3390\/e26010093"},{"key":"e_1_3_5_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSE.2023.3269081"},{"key":"e_1_3_5_16_1","unstructured":"Glanois C Weng P Zimmer M et al. (2021) A survey on interpretable reinforcement learning. ArXiv Preprint arXiv:2112.13112."},{"key":"e_1_3_5_17_1","volume-title":"Digital Image Processing","author":"Gonzalez RC","year":"2009","unstructured":"Gonzalez RC (2009) Digital Image Processing. Chennai, India: Pearson Education India."},{"key":"e_1_3_5_18_1","doi-asserted-by":"publisher","DOI":"10.1242\/jeb.138438"},{"key":"e_1_3_5_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCDS.2023.3315513"},{"key":"e_1_3_5_20_1","doi-asserted-by":"publisher","DOI":"10.3389\/fnbot.2023.1127642"},{"key":"e_1_3_5_21_1","unstructured":"Hiraoka T Imagawa T Hashimoto T et al. (2021) Dropout q-functions for doubly efficient reinforcement learning. ArXiv Preprint arXiv:2110.02034."},{"key":"e_1_3_5_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3119127"},{"key":"e_1_3_5_23_1","doi-asserted-by":"crossref","unstructured":"Homchanthanakul J Ngamkajornwiwat P Teerakittikul P et al. (2019) Neural control with an artificial hormone system for energy-efficient compliant terrain locomotion and adaptation of walking robots. In: 2019 IEEE\/RSJ international conference on intelligent robots and systems (IROS) Macau China 03\u201308 November 2019 5475\u20135482. IEEE.","DOI":"10.1109\/IROS40897.2019.8968580"},{"key":"e_1_3_5_24_1","unstructured":"Hu K O\u2019Connor P (2018) Learning a representation map for robot navigation using deep variational autoencoder. ArXiv Preprint arXiv:1807.02401."},{"key":"e_1_3_5_25_1","doi-asserted-by":"crossref","unstructured":"Isele D Cosgun A (2018) Selective experience replay for lifelong learning. In: Proceedings of the AAAI conference on artificial intelligence New Orleans LA 2\u20137 February 2018.","DOI":"10.1609\/aaai.v32i1.11595"},{"key":"e_1_3_5_26_1","volume-title":"Identification of Behaviors in an Agent\u2019s Phase Space","author":"Jaeger H","year":"1995","unstructured":"Jaeger H (1995) Identification of Behaviors in an Agent\u2019s Phase Space. Princeton, NJ: Citeseer."},{"key":"e_1_3_5_27_1","unstructured":"Kalashnikov D Varley J Chebotar Y et al. (2021) Mt-opt: continuous multi-task robotic reinforcement learning at scale. ArXiv Preprint arXiv:2104.08212."},{"key":"e_1_3_5_28_1","doi-asserted-by":"publisher","DOI":"10.1613\/jair.1.13673"},{"key":"e_1_3_5_29_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00452-0"},{"key":"e_1_3_5_30_1","doi-asserted-by":"publisher","DOI":"10.1177\/0278364911426178"},{"key":"e_1_3_5_31_1","doi-asserted-by":"crossref","unstructured":"Lechner M Hasani R Zimmer M et al. (2019) Designing worm-inspired neural networks for interpretable robotic control. In: 2019 international conference on robotics and automation (ICRA) Montreal QC 20\u201324 May 2019 87\u201394. IEEE.","DOI":"10.1109\/ICRA.2019.8793840"},{"key":"e_1_3_5_32_1","unstructured":"Lee D Ahn D Kim T et al. (2018) Viterbi-based pruning for sparse matrix with fixed and high index compression ratio. In: International conference on learning representations Vancouver BC 30 April\u20133 May 2018."},{"key":"e_1_3_5_33_1","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.abc5986"},{"key":"e_1_3_5_34_1","doi-asserted-by":"crossref","unstructured":"Lele AS Fang Y Ting J et al. (2020) Learning to walk: spike based reinforcement learning for hexapod robot central pattern generation. In: 2020 2nd IEEE international conference on artificial intelligence circuits and systems (AICAS) Genova Italy 31 August 2020\u20132 September 2020 208\u2013212. IEEE.","DOI":"10.1109\/AICAS48895.2020.9073987"},{"key":"e_1_3_5_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10489-020-01786-1"},{"key":"e_1_3_5_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2024.3388842"},{"key":"e_1_3_5_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3236386.3241340"},{"key":"e_1_3_5_38_1","doi-asserted-by":"publisher","DOI":"10.24425\/mms.2019.130562"},{"key":"e_1_3_5_39_1","unstructured":"Mansard N Aycard O Koike C (2005) Hierarchy of behaviors. In: Proceedings of the IEEE international conference on robotics and biomimetics Barcelona Spain 18\u201322 April 2005."},{"key":"e_1_3_5_40_1","doi-asserted-by":"crossref","unstructured":"Margolis GB Yang G Paigwar K et al. (2022) Rapid locomotion via reinforcement learning. ArXiv Preprint arXiv:2205.02824.","DOI":"10.15607\/RSS.2022.XVIII.022"},{"key":"e_1_3_5_41_1","doi-asserted-by":"crossref","unstructured":"Nahrendra IMA Yu B Myung H (2023) Dreamwaq: learning robust quadrupedal locomotion with implicit terrain imagination via deep reinforcement learning. In: 2023 IEEE international conference on robotics and automation (ICRA) London UK 29 May\u20132 June 2023 5078\u20135084. IEEE.","DOI":"10.1109\/ICRA48891.2023.10161144"},{"key":"e_1_3_5_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/3583558"},{"key":"e_1_3_5_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2019.01.012"},{"key":"e_1_3_5_44_1","doi-asserted-by":"crossref","unstructured":"Pasemann F Hild M Zahedi K (2003) So (2)-networks as neural oscillators. In: Proceedings part I computational methods in neural modeling: 7th international work-conference on artificial and natural neural networks IWANN Menorca Spain 3\u20136 June 2003 144\u2013151. Springer.","DOI":"10.1007\/3-540-44868-3_19"},{"key":"e_1_3_5_45_1","unstructured":"Ribeiro J Melo FS Dias J (2019) Multi-task learning and catastrophic forgetting in continual reinforcement learning. ArXiv Preprint arXiv:1909.10008."},{"key":"e_1_3_5_46_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12065-014-0109-3"},{"key":"e_1_3_5_47_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-019-0048-x"},{"key":"e_1_3_5_48_1","unstructured":"Rudin N Hoeller D Reist P et al. (2022) Learning to walk in minutes using massively parallel deep reinforcement learning. In: Faust A Hsu D Neumann G (eds) Proceedings of the 5th conference on robot learning proceedings of machine learning research London UK 8\u201311 November 2021 91\u2013100. PMLR. URL: https:\/\/proceedings.mlr.press\/v164\/rudin22a.html"},{"key":"e_1_3_5_49_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00505-4"},{"key":"e_1_3_5_50_1","doi-asserted-by":"crossref","unstructured":"Schilling M Konen K Korthals T (2020) Modular deep reinforcement learning for emergent locomotion on a six-legged robot. In: In: 2020 8th IEEE RAS\/EMBS international conference for biomedical robotics and biomechatronics (BioRob) New York NY 29 November 2020\u201301 December 2020 946\u2013953. IEEE.","DOI":"10.1109\/BioRob49111.2020.9224332"},{"key":"e_1_3_5_51_1","unstructured":"Schulman J Wolski F Dhariwal P et al. (2017) Proximal policy optimization algorithms. ArXiv Preprint arXiv:1707.06347."},{"key":"e_1_3_5_52_1","unstructured":"Schwarz J Czarnecki W Luketina J et al. (2018) Progress & compress: a scalable framework for continual learning. In: International conference on machine learning Stockholm Sweden 10\u201315 July 2018 4528\u20134537. PMLR."},{"key":"e_1_3_5_53_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2009.12.004"},{"key":"e_1_3_5_54_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-47443-w"},{"key":"e_1_3_5_55_1","doi-asserted-by":"crossref","unstructured":"Smith L Kew JC Peng XB et al. (2022a) Legged robots that keep on learning: fine-tuning locomotion policies in the real world. In: 2022 international conference on robotics and automation (ICRA) Philadelphia PA 23\u201327 May 2022 1593\u20131599. IEEE.","DOI":"10.1109\/ICRA46639.2022.9812166"},{"key":"e_1_3_5_56_1","doi-asserted-by":"crossref","unstructured":"Smith L Kostrikov I Levine S (2022b) A walk in the park: learning to walk in 20\u00a0minutes with model-free reinforcement learning. ArXiv Preprint arXiv:2208.07860.","DOI":"10.15607\/RSS.2023.XIX.056"},{"key":"e_1_3_5_57_1","doi-asserted-by":"crossref","unstructured":"Srisuchinnawong A Shao D Ngamkajornwiwat P et al. (2019) Neural control for gait generation and adaptation of a gecko robot. In: 2019 19th international conference on advanced robotics (ICAR) Belo Horizonte Brazil 2\u20136 December 2019 468\u2013473. IEEE.","DOI":"10.1109\/ICAR46387.2019.8981580"},{"key":"e_1_3_5_58_1","doi-asserted-by":"publisher","DOI":"10.3389\/fncir.2021.743101"},{"key":"e_1_3_5_59_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10846-020-01285-y"},{"key":"e_1_3_5_60_1","first-page":"1205","article-title":"Adaptive bipedal robot walking on industrial pipes under neural multimodal locomotion control: toward robotic out-pipe inspection","volume":"29","author":"Srisuchinnawong A","year":"2023","unstructured":"Srisuchinnawong A, Phongaksorn K, Ausrivong W, et al. (2023) Adaptive bipedal robot walking on industrial pipes under neural multimodal locomotion control: toward robotic out-pipe inspection. IEEE 29: 1205\u20131216.","journal-title":"IEEE"},{"key":"e_1_3_5_61_1","doi-asserted-by":"publisher","DOI":"10.1515\/revneuro-2017-0102"},{"key":"e_1_3_5_62_1","unstructured":"Stulp F Sigaud O (2012) Policy improvement methods: between black-box optimization and episodic reinforcement learning. https:\/\/hal.science\/hal-00738463v1"},{"key":"e_1_3_5_63_1","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton RS","year":"2018","unstructured":"Sutton RS, Barto AG (2018) Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press."},{"key":"e_1_3_5_64_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00444-0"},{"key":"e_1_3_5_65_1","unstructured":"Thor M Larsen JC Manoonpong P (2018) Morf\u2014modular robot framework. In: Proceedings of 2nd international youth conference of bionic engineering (IYCBE) Odense Denmark 7\u20139 November 2018 pp. 21\u201323."},{"key":"e_1_3_5_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3016523"},{"key":"e_1_3_5_67_1","doi-asserted-by":"publisher","DOI":"10.3389\/fncir.2021.743888"},{"key":"e_1_3_5_68_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2020.08.006"},{"key":"e_1_3_5_69_1","doi-asserted-by":"publisher","DOI":"10.1111\/jne.12911"},{"key":"e_1_3_5_70_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3055499"},{"key":"e_1_3_5_71_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3367329"},{"key":"e_1_3_5_72_1","doi-asserted-by":"publisher","DOI":"10.1002\/cne.23705"},{"key":"e_1_3_5_73_1","unstructured":"Xie A Harrison J Finn C (2020) Deep reinforcement learning amidst lifelong non-stationarity. ArXiv Preprint arXiv:2006.10701."},{"key":"e_1_3_5_74_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature24056"},{"key":"e_1_3_5_75_1","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.abb2174"},{"key":"e_1_3_5_76_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-023-00701-w"},{"key":"e_1_3_5_77_1","doi-asserted-by":"crossref","unstructured":"Zenker S Aksoy EE Goldschmidt D et al. (2013) Visual terrain classification for selecting energy efficient gaits of a hexapod robot. In: 2013 IEEE\/ASME international conference on advanced intelligent mechatronics Wollongong NSW 9\u201312 July 2013 577\u2013584. IEEE.","DOI":"10.1109\/AIM.2013.6584154"},{"key":"e_1_3_5_78_1","unstructured":"Zhang H Shi D Zhuang Z et al. (2023) Rsg: fast learning adaptive skills for quadruped robots by skill graph. ArXiv Preprint arXiv:2311.06015."},{"key":"e_1_3_5_79_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-024-05949-6"}],"container-title":["The International Journal of Robotics Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649251336385","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/02783649251336385","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649251336385","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:17:27Z","timestamp":1777457847000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/02783649251336385"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,15]]},"references-count":78,"journal-issue":{"issue":"13","published-print":{"date-parts":[[2025,11]]}},"alternative-id":["10.1177\/02783649251336385"],"URL":"https:\/\/doi.org\/10.1177\/02783649251336385","relation":{},"ISSN":["0278-3649","1741-3176"],"issn-type":[{"value":"0278-3649","type":"print"},{"value":"1741-3176","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,15]]}}}