{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T13:20:15Z","timestamp":1775913615206,"version":"3.50.1"},"reference-count":44,"publisher":"Cambridge University Press (CUP)","issue":"11","license":[{"start":{"date-parts":[[2024,10,16]],"date-time":"2024-10-16T00:00:00Z","timestamp":1729036800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.cambridge.org\/core\/terms"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Robotica"],"published-print":{"date-parts":[[2024,11]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>This study proposes a novel hybrid learning approach for developing a visual path-following algorithm for industrial robots. The process involves three steps: data collection from a simulation environment, network training, and testing on a real robot. The actor network is trained using supervised learning for 500 epochs. A semitrained network is then obtained at the <jats:inline-formula><jats:alternatives><jats:inline-graphic xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" mime-subtype=\"png\" xlink:href=\"S026357472400170X_inline1.png\"\/><jats:tex-math>\n$250^{th}$\n<\/jats:tex-math><\/jats:alternatives><\/jats:inline-formula> epoch. This network is further trained for another 250 epochs using reinforcement learning methods within the simulation environment. Networks trained with supervised learning (500 epochs) and the proposed hybrid learning method (250 epochs each of supervised and reinforcement learning) are compared. The hybrid learning approach achieves a significantly lower average error (30.9 mm) compared with supervised learning (39.3 mm) on real-world images. Additionally, the hybrid approach exhibits faster processing times (31.7 s) compared with supervised learning (35.0 s). The proposed method is implemented on a KUKA Agilus KR6 R900 six-axis robot, demonstrating its effectiveness. Furthermore, the hybrid approach reduces the total power consumption of the robot\u2019s motors compared with the supervised learning method. These results suggest that the hybrid learning approach offers a more effective and efficient solution for visual path following in industrial robots compared with traditional supervised learning.<\/jats:p>","DOI":"10.1017\/s026357472400170x","type":"journal-article","created":{"date-parts":[[2024,10,16]],"date-time":"2024-10-16T09:46:19Z","timestamp":1729071979000},"page":"3888-3903","source":"Crossref","is-referenced-by-count":3,"title":["Hybrid learning-based visual path following for an industrial robot"],"prefix":"10.1017","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5448-8281","authenticated-orcid":false,"given":"Mustafa Can","family":"Bingol","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8142-1146","authenticated-orcid":false,"given":"Omur","family":"Aydogmus","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"56","published-online":{"date-parts":[[2024,10,16]]},"reference":[{"key":"S026357472400170X_ref7","doi-asserted-by":"crossref","first-page":"497","DOI":"10.1109\/TIV.2019.2919476","article-title":"Vision-based autonomous path following using a human driver control model with reliable input-feature value estimation","volume":"4","author":"Okamoto","year":"2019","journal-title":"IEEE Trans. Intell. Veh."},{"key":"S026357472400170X_ref42","unstructured":"[42] Schulman, J. , Wolski, F. , Dhariwal, P. , Radford, A. and Klimov, O. , Proximal policy optimization algorithms (2017)."},{"key":"S026357472400170X_ref8","doi-asserted-by":"crossref","first-page":"104956","DOI":"10.1016\/j.autcon.2023.104956","article-title":"Robot morphology evolution for automated HVAC system inspections using graph heuristic search and reinforcement learning","volume":"153","author":"Duan","year":"2023","journal-title":"Automat. Constr."},{"key":"S026357472400170X_ref22","doi-asserted-by":"crossref","first-page":"110631","DOI":"10.1016\/j.oceaneng.2022.110631","article-title":"Soft actor\u2013critic based active disturbance rejection path following control for unmanned surface vessel under wind and wave disturbances,\u201d","volume":"247","author":"Zheng","year":"2022","journal-title":"Ocean Engineering"},{"key":"S026357472400170X_ref34","doi-asserted-by":"crossref","first-page":"109725","DOI":"10.1016\/j.measurement.2021.109725","article-title":"A deep q-network for robotic odor\/gas source localization: Modeling, measurement and comparative study","volume":"183","author":"Chen","year":"2021","journal-title":"Measurement"},{"key":"S026357472400170X_ref28","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1016\/j.neucom.2020.04.059","article-title":"A trajectory planning method for robot scanning system uuuusing mask r-cnn for scanning objects with unknown model,\u201d","volume":"404","author":"Yang","year":"2020","journal-title":"Neurocomputing"},{"key":"S026357472400170X_ref6","doi-asserted-by":"crossref","first-page":"749","DOI":"10.1109\/TRO.2009.2017140","article-title":"Qualitative vision-based path following","volume":"25","author":"Chen","year":"2009","journal-title":"IEEE Trans. Robot."},{"key":"S026357472400170X_ref4","doi-asserted-by":"crossref","first-page":"1646","DOI":"10.1109\/TMECH.2020.3026994","article-title":"Dynamic path correction of an industrial robot using a distance sensor and an ADRC controller","volume":"26","author":"Khaled","year":"2020","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"S026357472400170X_ref24","doi-asserted-by":"crossref","first-page":"6189","DOI":"10.1016\/j.ifacol.2020.12.1709","article-title":"Multidimensional path tracking with global least squares solution","volume":"53","author":"Handler","year":"2020","journal-title":"IFAC-PapersOnLine"},{"key":"S026357472400170X_ref26","doi-asserted-by":"crossref","first-page":"102540","DOI":"10.1016\/j.mechatronics.2021.102540","article-title":"A novel 3d path following control framework for robots performing surface finishing tasks","volume":"76","author":"Wen","year":"2021","journal-title":"Mechatronics"},{"key":"S026357472400170X_ref11","doi-asserted-by":"crossref","first-page":"888","DOI":"10.1017\/S0263574721000886","article-title":"Multi-objective optimal trajectory planning for manipulators in the presence of obstacles","volume":"40","author":"Zhang","year":"2022","journal-title":"Robotica"},{"key":"S026357472400170X_ref38","doi-asserted-by":"crossref","unstructured":"[38] Kamali, K. , Bonev, I. A. and Desrosiers, C. , \u201cReal-Time Motion Planning for Robotic Teleoperation Using Dynamic-Goal Deep Reinforcement Learning,\u201d In 2020 17th Conference on Computer and Robot Vision (CRV) (2020) pp. 182\u2013189.","DOI":"10.1109\/CRV50864.2020.00032"},{"key":"S026357472400170X_ref39","doi-asserted-by":"crossref","unstructured":"[39] Liu, Z. and Kubota, N. , \u201cHybrid Learning Approach Based on Multi-Objective BehaviorCoordination for Multiple Robots,\u201d In 2007 International Conference on Mechatronics and Automation (2007) pp. 204\u2013209.","DOI":"10.1109\/ICMA.2007.4303541"},{"key":"S026357472400170X_ref41","doi-asserted-by":"crossref","first-page":"736","DOI":"10.1016\/j.jmsy.2020.08.010","article-title":"Attitude data-based deep hybrid learning architecture for intelligent fault diagnosis of multi-joint industrial robots","volume":"61","author":"Long","year":"2021","journal-title":"Journal of Manufacturing Systems"},{"key":"S026357472400170X_ref20","doi-asserted-by":"crossref","first-page":"106682","DOI":"10.1016\/j.cie.2020.106682","article-title":"An analytical and a deep learning model for solving the inverse kinematic problem of an industrial parallel robot","volume":"151","author":"Toquica","year":"2021","journal-title":"Computers and Industrial Engineering"},{"key":"S026357472400170X_ref12","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1017\/S0263574723001807","article-title":"Collision avoidance trajectory planning for a dual-robot system: Using a modified APF method","volume":"42","author":"Yang","year":"2024","journal-title":"Robotica"},{"key":"S026357472400170X_ref17","doi-asserted-by":"crossref","first-page":"359","DOI":"10.1108\/IR-09-2019-0180","article-title":"Practical application of a safe human-robot interaction software,","volume":"47","author":"Bingol","year":"2020","journal-title":"Industrial Robot: The International Journal of Robotics Research and Application"},{"key":"S026357472400170X_ref5","doi-asserted-by":"crossref","first-page":"102702","DOI":"10.1016\/j.rcim.2023.102702","article-title":"A novel 3D vision-based robotic welding path extraction method for complex intersection curves","volume":"87","author":"Geng","year":"2024","journal-title":"Robot. Comput.-Integr. Manuf."},{"key":"S026357472400170X_ref9","doi-asserted-by":"crossref","first-page":"106941","DOI":"10.1016\/j.engappai.2023.106941","article-title":"A deep reinforcement learning algorithm to control a two-wheeled scooter with a humanoid robot","volume":"126","author":"Baltes","year":"2023","journal-title":"Eng. Appl. Artif. Intell."},{"key":"S026357472400170X_ref23","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1016\/j.matcom.2021.10.028","article-title":"Intelligent controller for nonholonomic wheeled mobile robot: A fuzzy path following combination","volume":"193","author":"Mondal","year":"2022","journal-title":"Mathematics and Computers in Simulation"},{"key":"S026357472400170X_ref37","unstructured":"[37] Bingol, M. C. , \u201cProximal policy optimization-based path generation of an industrial robot,\u201d International Conference on Engineering, Natural and Applied Science 2021 (ICENAS\u201921) (2021) pp. 322\u2013328."},{"key":"S026357472400170X_ref31","doi-asserted-by":"crossref","first-page":"107603","DOI":"10.1016\/j.cie.2021.107603","article-title":"Deep learning-based optimization for motion planning of dual-arm assembly robots","volume":"160","author":"Ying","year":"2021","journal-title":"Computers and Industrial Engineering"},{"key":"S026357472400170X_ref18","doi-asserted-by":"crossref","first-page":"103903","DOI":"10.1016\/j.engappai.2020.103903","article-title":"Performing predefined tasks using the human\u2013robot interaction on speech recognition for an industrial robot","volume":"95","author":"Bingol","year":"2020","journal-title":"Engineering Applications of Artificial Intelligence"},{"key":"S026357472400170X_ref14","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1017\/S0263574723001546","article-title":"Obstacle avoidance path planning of 6-DOF robotic arm based on improved A* algorithm and artificial potential field method","volume":"42","author":"Tang","year":"2024","journal-title":"Robotica"},{"key":"S026357472400170X_ref21","doi-asserted-by":"crossref","first-page":"110265","DOI":"10.1016\/j.oceaneng.2021.110265","article-title":"A robust model predictive path following controller for an autonomous underwater vehicle","volume":"244","author":"Rath","year":"2022","journal-title":"Ocean Engineering"},{"key":"S026357472400170X_ref25","doi-asserted-by":"crossref","first-page":"102130","DOI":"10.1016\/j.rcim.2021.102130","article-title":"A visual path-following learning approach for industrial robots using drl","volume":"71","author":"Maldonado-Ramirez","year":"2021","journal-title":"Robotics and Computer-Integrated Manufacturing"},{"key":"S026357472400170X_ref30","doi-asserted-by":"crossref","first-page":"102262","DOI":"10.1016\/j.rcim.2021.102262","article-title":"Development of a vision based pose estimation system for robotic machining and improving its accuracy using lstm neural networks and sparse regression","volume":"74","author":"Bilal","year":"2022","journal-title":"Robotics and Computer-Integrated Manufacturing"},{"key":"S026357472400170X_ref32","doi-asserted-by":"crossref","first-page":"104985","DOI":"10.1016\/j.compag.2019.104985","article-title":"Double-dqn based path smoothing and tracking control method for robotic vehicle navigation","volume":"166","author":"Zhang","year":"2019","journal-title":"Computers and Electronics in Agriculture"},{"key":"S026357472400170X_ref43","unstructured":"[43] Lillicrap, T. P. , Hunt, J. J. , Pritzel, A. , Heess, N. , Erez, T. , Tassa, Y. , Silver, D. and Wierstra, D. , Continuous control with deep reinforcement learning (2015)."},{"key":"S026357472400170X_ref2","doi-asserted-by":"crossref","first-page":"3410","DOI":"10.1109\/TMECH.2023.3267980","article-title":"A data-driven approach for online path correction of industrial robots using modified flexible dynamics model and disturbance state observer","volume":"28","author":"Lin","year":"2023","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"S026357472400170X_ref10","doi-asserted-by":"crossref","first-page":"2026","DOI":"10.1017\/S0263574724000766","article-title":"Multi-objective trajectory planning for industrial robots using a hybrid optimization approach","volume":"42","author":"Chettibi","year":"2024","journal-title":"Robotica"},{"key":"S026357472400170X_ref3","doi-asserted-by":"crossref","first-page":"107410","DOI":"10.1016\/j.compag.2022.107410","article-title":"Navigation and control development for a four-wheel-steered mobile orchard robot using model-based design","volume":"202","author":"Raikwar","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"S026357472400170X_ref35","doi-asserted-by":"crossref","first-page":"108372","DOI":"10.1016\/j.ymssp.2021.108372","article-title":"Robotic seam tracking system combining convolution filter and deep reinforcement learning","volume":"165","author":"Zou","year":"2022","journal-title":"Mechanical Systems and Signal Processing"},{"key":"S026357472400170X_ref27","volume-title":"Development of Artificial Intelligence-Based Self-Programmable Robot Software Compatible with Industry 4.0 using Human-Robot Interaction PhD thesis","author":"Bingol","year":"2020"},{"key":"S026357472400170X_ref1","unstructured":"[1] International Federation of Robotics (IFR), Homepage, https:\/\/ifr.org, accessed October 05, 2024."},{"key":"S026357472400170X_ref13","doi-asserted-by":"crossref","first-page":"265","DOI":"10.1017\/S0263574723001479","article-title":"End-effector path tracking of a 14 DOF rover manipulator system in CG-space framework","volume":"42","author":"Katiyar","year":"2024","journal-title":"Robotica"},{"key":"S026357472400170X_ref40","doi-asserted-by":"crossref","unstructured":"[40] Desouky, S. F. and Schwartz, H. M. , \u201cA Novel Hybrid Learning Technique Applied to a Self-Learning Multi-Robot System,\u201d In 2009 IEEE International Conference on Systems, Man and Cybernetics (2009) pp. 2616\u20132623.","DOI":"10.1109\/ICSMC.2009.5346111"},{"key":"S026357472400170X_ref44","unstructured":"[44] Schulman, J. , Levine, S. , Abbeel, P. , Jordan, M. and Moritz, P. , \u201cTrust region policy optimization,\u201d International Conference on Machine Learning (2015) pp. 1889\u20131897."},{"key":"S026357472400170X_ref19","doi-asserted-by":"crossref","first-page":"29299","DOI":"10.1016\/j.jclepro.2021.129299","article-title":"A transfer-learning based energy consumption modeling method for industrial robots","volume":"325","author":"Yan","year":"2021","journal-title":"Journal of Cleaner Production"},{"key":"S026357472400170X_ref15","doi-asserted-by":"crossref","first-page":"521","DOI":"10.1016\/j.isatra.2021.11.019","article-title":"A deep transferable motion-adaptive fault detection method for industrial robots using a residual\u2013convolutional neural network","volume":"128","author":"Oh","year":"2021","journal-title":"ISA Transactions"},{"key":"S026357472400170X_ref29","doi-asserted-by":"crossref","first-page":"103857","DOI":"10.1016\/j.robot.2021.103857","article-title":"Where is my hand? deep hand segmentation for visual self-recognition in humanoid robots","volume":"145","author":"Almeida","year":"2021","journal-title":"Robotics and Autonomous Systems"},{"key":"S026357472400170X_ref33","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1016\/j.neunet.2020.12.001","article-title":"Modular deep reinforcement learning from reward and punishment for robot navigation","volume":"135","author":"Wang","year":"2021","journal-title":"Neural Networks"},{"key":"S026357472400170X_ref36","doi-asserted-by":"crossref","first-page":"107605","DOI":"10.1016\/j.asoc.2021.107605","article-title":"A multi-robot path-planning algorithm for autonomous navigation using meta-reinforcement learning based on transfer learning,\u201d","volume":"110","author":"Wen","year":"2021","journal-title":"Applied Software Computing"},{"key":"S026357472400170X_ref16","doi-asserted-by":"crossref","first-page":"102228","DOI":"10.1016\/j.rcim.2021.102228","article-title":"Application of generalized frequency response functions and improved convolutional neural network to fault diagnosis of heavy-duty industrial robot","volume":"73","author":"Chen","year":"2022","journal-title":"Robotics and Computer-Integrated Manufacturing"}],"container-title":["Robotica"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S026357472400170X","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,30]],"date-time":"2025-01-30T12:49:10Z","timestamp":1738241350000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S026357472400170X\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,16]]},"references-count":44,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2024,11]]}},"alternative-id":["S026357472400170X"],"URL":"https:\/\/doi.org\/10.1017\/s026357472400170x","relation":{},"ISSN":["0263-5747","1469-8668"],"issn-type":[{"value":"0263-5747","type":"print"},{"value":"1469-8668","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,16]]}}}