{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T22:45:26Z","timestamp":1774305926065,"version":"3.50.1"},"reference-count":20,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T00:00:00Z","timestamp":1760659200000},"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. Artif. Intell."],"abstract":"<jats:p>With the advancement of autonomous driving technology, efficient and safe lane-keeping has become one of the core issues in this field. Currently, Deep Reinforcement Learning (DRL) methods still face challenges such as low training efficiency, slow algorithm convergence, and a tendency to fall into local optima when addressing lane-keeping issues. To address these challenges, a Prioritized Experience Replay (PER) mechanism designed to adapt to the learning process of the Twin Delayed Deep Deterministic Policy Gradient (TD3) is proposed, referred to as PER-TD3, to enhance the learning efficiency and lane-keeping performance of the vehicle in this work. It adjusts the probability of a selected sample by utilizing the difference between the predicted <jats:italic>Q<\/jats:italic> value and the true <jats:italic>Q<\/jats:italic> value to assign priority to different samples. By prioritizing samples with higher errors, the algorithm can correct biases in decision-making more quickly, especially when the vehicle deviates from its lane. In addition, introducing a probabilistic sampling mechanism helps to enhance the diversity of samples, ensuring high-frequency playback of high-value experiences, and enabling vehicles to learn accurate and stable lane-keeping strategies in a shorter period. Validation experiments on the TORCS platform demonstrate that the proposed framework can effectively solve the problem of unbalanced training, which is common in DRL, enhances training sample quality, accelerates algorithm convergence, and ultimately improves driving performance while ensuring safety.<\/jats:p>","DOI":"10.3389\/frai.2025.1688764","type":"journal-article","created":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T12:31:47Z","timestamp":1760704307000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Adaptive enhancements of autonomous lane keeping via advanced PER-TD3 framework"],"prefix":"10.3389","volume":"8","author":[{"given":"Xiting","family":"Peng","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinyan","family":"Liang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaoyu","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haibo","family":"Yang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weimin","family":"Lei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,10,17]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"1531","DOI":"10.1109\/TMECH.2017.2716838","article-title":"Longitudinal jerk estimation of driver intentions for advanced driver assistance systems","volume":"22","author":"Bisoffi","year":"2017","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"409","DOI":"10.1117\/12.317473","article-title":"The Argo autonomous vehicle\u2019s vision and control systems","volume":"3","author":"Broggi","year":"1999","journal-title":"Int. J. Intell. Control Syst."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"322","DOI":"10.12720\/jait.15.3.322-329","article-title":"Deep learning-based lane-keeping assist system for self-driving cars using transfer learning and fine tuning","volume":"15","author":"Hong","year":"2024","journal-title":"J. Adv. Inf. Technol."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1832","DOI":"10.1049\/iet-its.2019.0045","article-title":"Lane-keeping control based on an improved artificial potential method and coordination of steering braking systems","volume":"13","author":"Hongbo","year":"2019","journal-title":"IET Intell. Transp. Syst."},{"key":"ref5","first-page":"012020","article-title":"Exploration strategy improved DDPG for lane keeping tasks in autonomous driving","volume-title":"Journal of Physics: Conference Series","author":"Hua","year":"2022"},{"key":"ref6","doi-asserted-by":"crossref","first-page":"8248","DOI":"10.1109\/ICRA.2019.8793742","article-title":"Learning to drive in a day","volume-title":"2019 international conference on robotics and automation (ICRA)","author":"Kendall","year":"2019"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"4909","DOI":"10.1109\/TITS.2021.3054625","article-title":"Deep reinforcement learning for autonomous driving: a survey","volume":"23","author":"Kiran","year":"2021","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"3129","DOI":"10.1109\/TITS.2018.2873101","article-title":"Synthesis of robust lane keeping systems: impact of controller and design parameters on system performance","volume":"20","author":"Lee","year":"2018","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1459","DOI":"10.1016\/j.conengprac.2011.08.005","article-title":"Nested PID steering control for lane keeping in autonomous vehicles","volume":"19","author":"Marino","year":"2011","journal-title":"Control. Eng. Pract."},{"key":"ref10","doi-asserted-by":"crossref","first-page":"3518","DOI":"10.1109\/GLOBECOM54140.2023.10437047","article-title":"LK-TDDQN: a lane keeping transfer double deep Q network framework for autonomous vehicles","volume-title":"GLOBECOM 2023-2023 IEEE global communications conference","author":"Peng","year":"2023"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1007\/s42154-021-00151-3","article-title":"End-to-end autonomous driving through dueling double deep Q-network","volume":"4","author":"Peng","year":"2021","journal-title":"Auto. Innov."},{"key":"ref12","doi-asserted-by":"crossref","first-page":"300","DOI":"10.1109\/LARS\/SBR\/WRE54079.2021.9605436","article-title":"Addressing lane keeping and intersections using deep conditional reinforcement learning","volume-title":"2021 Latin American robotics symposium (LARS), 2021 Brazilian symposium on robotics (SBR), and 2021 workshop on robotics in education (WRE)","author":"Silva","year":"2021"},{"key":"ref13","doi-asserted-by":"crossref","DOI":"10.4271\/2020-01-0728","article-title":"Lane keeping assist for an autonomous vehicle based on deep reinforcement learning","author":"Wang","year":"2020"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"11","DOI":"10.56578\/ataiml030102","article-title":"Adaptive lane keeping assistance system with integrated driver intent and lane departure warning","volume":"3","author":"Wei","year":"2024","journal-title":"Acadlore Trans. Mach. Learn."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1109\/TITS.2008.922874","article-title":"The heterogeneous systems integration design and implementation for lane keeping on a vehicle","volume":"9","author":"Wu","year":"2008","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref16","first-page":"2","article-title":"Torcs, the open racing car simulator","volume":"4","author":"Wymann","year":"2000","journal-title":"Software"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1093\/tse\/tdaa009","article-title":"Self-learning drift control of automated vehicles beyond handling limit after rear-end collision","volume":"2","author":"Yin","year":"2020","journal-title":"Transp. Saf. Environ."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/MRA.2021.3115980","article-title":"Prioritized experience replay-based deep q learning: multiple-reward architecture for highway driving decision making","volume":"28","author":"Yuan","year":"2021","journal-title":"IEEE Robot. Autom. Mag."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"626","DOI":"10.1007\/s12204-021-2352-y","article-title":"Stochastic model predictive control approach to autonomous vehicle lane keeping","volume":"26","author":"Zhang","year":"2021","journal-title":"J. Shanghai Jiaotong Univ. (Sci.)"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"3017","DOI":"10.1109\/TMECH.2023.3236245","article-title":"Driver centriclane-keeping assistance system design: a noncertainty-equivalent neuro-adaptive control approach","volume":"28","author":"Zhou","year":"2023","journal-title":"IEEE\/ASME Trans. Mechatron."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1688764\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T12:31:49Z","timestamp":1760704309000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1688764\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,17]]},"references-count":20,"alternative-id":["10.3389\/frai.2025.1688764"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1688764","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,17]]},"article-number":"1688764"}}