{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T06:05:03Z","timestamp":1790229903507,"version":"4.1.0"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:00:00Z","timestamp":1790208000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Science Committee of the Ministry of Higher Education and Science of the Republic of Kazakhstan","award":["AP26103285"],"award-info":[{"award-number":["AP26103285"]}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Autonomous robotic manipulation in modern manufacturing requires control policies that simultaneously achieve precision, adaptation, operational safety, and multi-task capability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This paper proposes a Transformer-Based Reinforcement Learning Network (TRL-Net) that combines temporal state encoding, task-conditioned representation, actor-critic reinforcement learning, domain-randomized simulation, and safety-constrained action projection. The evaluation is limited to ten simulated industrial manipulation tasks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The reported point estimates are a 98.6% task success rate, 0.81 mm average positioning error, 0.0% collision rate in the evaluated simulation episodes, 0.42 m\/s\u00b3 trajectory jerk, and 4.2 ms policy inference time.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These findings are consistent with improved temporal control and multi-task performance relative to the reported baselines. However, the archived experimental record does not contain real-robot trials, component-wise ablations, seed-level dispersion, complete hyperparameters, or hardware specifications. The conclusions are therefore restricted to the reported simulation setting, and the safety layer is interpreted as enforcing configured simulator constraints rather than guaranteeing safe operation on physical industrial hardware.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1922657","type":"journal-article","created":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T05:44:49Z","timestamp":1790228689000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A transformer-based reinforcement learning framework for autonomous training of industrial robotic manipulators"],"prefix":"10.3389","volume":"9","author":[{"given":"Balnur","family":"Kenjayeva","sequence":"first","affiliation":[{"name":"International University of Tourism and Hospitality","place":["Turkistan, Kazakhstan"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alibek","family":"Galimov","sequence":"additional","affiliation":[{"name":"Joldasbekov Institute of Mechanics and Engineering","place":["Almaty, Kazakhstan"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,24]]},"reference":[{"key":"ref1","first-page":"3135","volume-title":"Scalable multi-agent Reinforcement Learning Framework for multi-Machine Tending","author":"Abdalwhab","year":"2026"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"14721","DOI":"10.1038\/s41598-026-50418-0","article-title":"A novel intelligent hybrid reinforcement learning framework for autonomous decision making in complex health cognitive systems","volume":"16","author":"Abdullah","year":"2026","journal-title":"Sci. Rep."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1434","DOI":"10.1109\/TASE.2025.3646982","article-title":"Autonomous driving at unsignalized intersections: a review of decision-making challenges and reinforcement learning-based solutions","volume":"23","author":"Al-Sharman","year":"2025","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"31422","DOI":"10.1109\/JIOT.2024.3418352","article-title":"Online fault diagnosis of industrial robot using IoRT and hybrid deep learning techniques: an experimental approach","volume":"11","author":"Bilal","year":"2024","journal-title":"IEEE Internet Things J."},{"key":"ref5","volume-title":"RT-1: Robotics Transformer for real-World Control at scale","author":"Brohan","year":"2023"},{"key":"ref6","first-page":"15084","article-title":"Decision transformer: reinforcement learning via sequence modeling","volume":"34","author":"Chen","year":"2021","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"10233","DOI":"10.1109\/TII.2023.3240585","article-title":"Transformer-based imitative reinforcement learning for multirobot path planning","volume":"19","author":"Chen","year":"2023","journal-title":"IEEE Trans. Industr. Inform."},{"key":"ref8","first-page":"336","author":"Chu","year":"2025"},{"key":"ref9","first-page":"193","author":"Elseiagy","year":"2026"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"29233","DOI":"10.1038\/s41598-025-12742-9","article-title":"Human-robot interaction using retrieval-augmented generation and fine-tuning with transformer neural networks in industry 5.0","volume":"15","author":"Fazlollahtabar","year":"2025","journal-title":"Sci. Rep."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"14563","DOI":"10.1109\/TIE.2025.3581265","article-title":"Object recognition and obstacle avoidance for multimobile robots via Swin transformer and deep reinforcement learning","volume":"72","author":"Fu","year":"2025","journal-title":"IEEE Trans. Ind. Electron."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"114293","DOI":"10.1016\/j.engappai.2026.114293","article-title":"Transformer-based fault-tolerant control for fixed-wing aerial vehicles using offline reinforcement learning","volume":"171","author":"Giral","year":"2026","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"11088","DOI":"10.1109\/TITS.2025.3560379","article-title":"A survey on deep reinforcement learning applications in autonomous systems: applications, open challenges, and future directions","volume":"26","author":"Govinda","year":"2025","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"2301.04104","DOI":"10.48550\/arXiv.2301.04104","article-title":"Mastering diverse domains through world models","author":"Hafner","year":"2023"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"8580","DOI":"10.1109\/TPAMI.2024.3408271","article-title":"On transforming reinforcement learning with transformers: the development trajectory","volume":"46","author":"Hu","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1832","DOI":"10.1109\/TITS.2023.3312453","article-title":"Goal-guided transformer-enabled reinforcement learning for efficient autonomous navigation","volume":"25","author":"Huang","year":"2023","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref17","first-page":"1273","article-title":"Offline reinforcement learning as one big sequence modeling problem","volume":"34","author":"Janner","year":"2021","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref18","first-page":"1","author":"Kaigom","year":"2023"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"6799","DOI":"10.1109\/LRA.2026.3681156","article-title":"Self-supervised multisensory Pretraining for contact-rich robot reinforcement learning","volume":"11","author":"Krohn","year":"2026","journal-title":"IEEE Rob. Autom. Lett."},{"key":"ref20","first-page":"1","author":"Kumar","year":"2026"},{"key":"ref21","first-page":"1084","author":"Li","year":"2025"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"2564","DOI":"10.1109\/TIE.2025.3600552","article-title":"ERL-fill: an emotion-aware reinforcement learning framework with staged Pretraining for gas\u2013solid flow filling control","volume":"73","author":"Ma","year":"2025","journal-title":"IEEE Trans. Ind. Electron."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"891","DOI":"10.32604\/cmes.2025.068615","article-title":"Three-dimensional trajectory planning for robotic manipulators using model predictive control and point cloud optimization","volume":"145","author":"Momynkulov","year":"2025","journal-title":"Comput. Model. Eng. Sci."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"1766","DOI":"10.1038\/s41598-025-28904-8","article-title":"Software-defined self-learning control system for industrial robots by using reinforcement learning","volume":"16","author":"Moon","year":"2026","journal-title":"Sci. Rep."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"3803","DOI":"10.1109\/ICRA.2018.8460528","article-title":"Sim-to-real transfer of robotic control with dynamics randomization","volume":"2018","author":"Peng","year":"2018","journal-title":"IEEE Int. Conf. Robot. Autom."},{"key":"ref26","first-page":"1","author":"Puang","year":"2024"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s10845-026-02831-5","article-title":"Adaptive multimodal fusion-driven reinforcement learning for robotic peg-in-hole assembly","author":"Qu","year":"2026","journal-title":"J. Intell. Manuf."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s44379-025-00050-y","article-title":"Physical AI: bridging the Sim-to-real divide toward embodied, ethical, and autonomous intelligence","volume":"2","author":"Ray","year":"2026","journal-title":"Mach. Learn. Comput. Sci. Eng."},{"key":"ref29","first-page":"205","author":"Shridhar","year":"2023"},{"key":"ref30","first-page":"09544070251408897","article-title":"Machine learning techniques enabling advanced autonomous vehicle navigation","author":"Shrivastava","year":"2026","journal-title":"Proc. Inst. Mech. Eng., Part D: J. Automob. Eng."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"100015","DOI":"10.1016\/j.cacaie.2026.100015","article-title":"Energy-aware reinforcement learning for robotic manipulation of articulated components in infrastructure operation and maintenance","volume":"46","author":"Tao","year":"2026","journal-title":"Comput. Aided Civ. Inf. Eng."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"4915","DOI":"10.1109\/LRA.2021.3070252","article-title":"Recovery RL: safe reinforcement learning with learned recovery zones","volume":"6","author":"Thananjeyan","year":"2021","journal-title":"IEEE Rob. Autom. Lett."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1109\/IROS.2017.8202133","article-title":"Domain randomization for transferring deep neural networks from simulation to the real world","volume":"2017","author":"Tobin","year":"2017","journal-title":"IEEE\/RSJ Int. Conf. Intell. Robot. Syst."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"3801","DOI":"10.1007\/s00170-025-16144-z","article-title":"Transformer-based path planning for single-arm and dual-arm robots in dynamic environments","volume":"139","author":"Wang","year":"2025","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"1446","DOI":"10.1109\/TMECH.2022.3219115","article-title":"Model-based reinforcement learning control of electrohydraulic position servo systems","volume":"28","author":"Yao","year":"2023","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1109\/TMECH.2024.3390129","article-title":"Model-data hybrid driven control of hydraulic Euler-Lagrange systems","volume":"30","author":"Yao","year":"2025","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"2673","DOI":"10.1109\/TMECH.2023.3336070","article-title":"Data-driven control of hydraulic manipulators by reinforcement learning","volume":"29","author":"Yao","year":"2024","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1109\/TFUZZ.2025.3585934","article-title":"Multi-agent fuzzy reinforcement learning with LLM for cooperative navigation of endovascular robotics","volume":"34","author":"Yao","year":"2026","journal-title":"IEEE Trans. Fuzzy Syst."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"763","DOI":"10.1631\/FITEE.2300548","article-title":"Transformer in reinforcement learning for decision-making: a survey","volume":"25","author":"Yuan","year":"2024","journal-title":"Front. Inf. Technol. Electron. Eng."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"125064","DOI":"10.1016\/j.eswa.2024.125064","article-title":"Cwt-vit: a time\u2013frequency representation and vision transformer-based framework for automated robotic surgical skill assessment","volume":"258","author":"Zhang","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"168732","DOI":"10.1109\/ACCESS.2025.3613729","article-title":"Uncertainty-aware multi-modal learning for autonomous pallet handling in dynamic warehouse scenarios","volume":"13","author":"Zhao","year":"2025","journal-title":"IEEE Acc."},{"key":"ref42","doi-asserted-by":"publisher","first-page":"35432","DOI":"10.1109\/JIOT.2024.3436110","article-title":"TranDRL: a transformer-driven deep reinforcement learning enabled prescriptive maintenance framework","volume":"11","author":"Zhao","year":"2024","journal-title":"IEEE Internet Things J."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"1646336","DOI":"10.3389\/fnbot.2025.1646336","article-title":"End-to-end robot intelligent obstacle avoidance method based on deep reinforcement learning with spatiotemporal transformer architecture","volume":"19","author":"Zhou","year":"2025","journal-title":"Front. Neurorobot."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1922657\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T05:44:50Z","timestamp":1790228690000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1922657\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,24]]},"references-count":43,"alternative-id":["10.3389\/frai.2026.1922657"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1922657","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,24]]},"article-number":"1922657"}}