{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T09:25:10Z","timestamp":1765445110525,"version":"3.46.0"},"reference-count":21,"publisher":"Emerald","issue":"6","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,11,25]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>The safety and traversability of tiny robots in longitudinal ditch-like terrain pose significant challenges because of wheel-based structures and vehicle height limitations. This paper aims to provide a reinforcement learning approach that enables a robot\u2019s four wheels to traverse longitudinal ditch-like terrain by leaping over it using only the conventional LIDAR sensor carried by the robot.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design\/methodology\/approach<\/jats:title>\n                    <jats:p>In this paper, a reinforcement learning architecture is proposed, which can find a path sequence suitable for the four wheels of a robot in the grid map built by LiDAR. First, an elevation grid map is established, and the standing points of the robot are selected from the grid map according to the length and width of the robot\u2019s four wheels and the height of the chassis. Then, the state quantity, including the four-wheel position sequence, is established. A comprehensive reward function is designed, including the robot\u2019s roll angle and the minimum distance from the chassis to the ground. Finally, an augmented learning robot\u2019s four-wheel trajectory planning algorithm aiming at safety and passability is established.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Findings<\/jats:title>\n                    <jats:p>The experimental results show that the proposed reinforcement learning method allows small robots to find suitable paths for their undercarriage structure in naturally formed longitudinal ditch-like rugged terrain. The centroid offset of the robot is less than 20 degrees in the planned path, and the overall operation is relatively stable. The minimum distance between the road surface and the chassis is 1.32\u2009cm, which is 66% of the height of the robot chassis, ensuring the safety and passability of the robot.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Research limitations\/implications<\/jats:title>\n                    <jats:p>The method in this paper is based on the condition that the road can be accurately built in a digital model. If the actual factors of the road are challenging to consider and describe digitally, the four-wheel trajectory planning method, which does not rely on an accurate digital map model, and the influence of four-wheel suspension structure and parameters on four-wheel trajectory planning, should be studied.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Practical implications<\/jats:title>\n                    <jats:p>This research result can effectively enhance the application environment of wheeled robots and can effectively improve the actual ability of robots in field investigation, exploration, rescue and other aspects. The technology included can serve the development and application of robot products working in special environments, bringing technical optimization and economic benefits to related enterprises.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Social implications<\/jats:title>\n                    <jats:p>This paper proposes a trajectory planning method based on reinforcement learning for small-wheeled robots to traverse longitudinal gullies by stepping. This method provides a new way to solve problems such as poor safety and poor passability of wheeled robots in rugged environments. It provides a theoretical basis for the path planning of robots in more complex environments. It has important significance for the research of robots in military, rescue, exploration and other fields.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Originality\/value<\/jats:title>\n                    <jats:p>This study presents a new path-planning method for small-wheeled robots, which has advantages over the existing methods for dealing with robot safety and passability problems under rough terrain.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1108\/ir-07-2024-0343","type":"journal-article","created":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T23:57:25Z","timestamp":1742860645000},"page":"833-842","source":"Crossref","is-referenced-by-count":0,"title":["Reinforcement learning for generating four-track trajectory planning to adapt to longitudinal rugged terrain"],"prefix":"10.1108","volume":"52","author":[{"given":"Zhiguang","family":"Liu","sequence":"first","affiliation":[{"name":"Tianjin Chengjian University School of Control and Mechanical Engineering, , Tianjin,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiahui","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tianjin Chengjian University School of Control and Mechanical Engineering, , Tianjin,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hainan","family":"Yang","sequence":"additional","affiliation":[{"name":"Tianjin Chengjian University School of Control and Mechanical Engineering, , Tianjin,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jian","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tianjin Chengjian University School of Control and Mechanical Engineering, , Tianjin,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fei","family":"Yu","sequence":"additional","affiliation":[{"name":"Hebei University of Technology School of Mechanical Engineering, , Tianjin,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","published-online":{"date-parts":[[2025,3,26]]},"reference":[{"issue":"3","key":"2025121104225077200_ref001","doi-asserted-by":"crossref","first-page":"2096","DOI":"10.1109\/LRA.2024.3354629","article-title":"Speed planning based on terrain-aware constraint reinforcement learning in rugged environments","volume":"9","author":"Andong","year":"2024","journal-title":"IEEE Robotics and Automation Letters"},{"issue":"6","key":"2025121104225077200_ref002","doi-asserted-by":"crossref","first-page":"843","DOI":"10.1007\/s10514-021-10013-w","article-title":"Navigating by touch: haptic monte Carlo localization via geometric sensing and terrain classificatio","volume":"45","author":"Buchanan","year":"2021","journal-title":"Autonomous Robots"},{"issue":"8","key":"2025121104225077200_ref003","doi-asserted-by":"publisher","DOI":"10.1007\/s10845-023-02197-y","article-title":"Predictive reinforcement learning: map-less navigation method for mobile robot","volume":"35","author":"Dobriborsci","year":"2023","journal-title":"Journal of Intelligent Manufacturing"},{"issue":"5","key":"2025121104225077200_ref004","doi-asserted-by":"crossref","first-page":"3402","DOI":"10.1109\/TRO.2023.3275384","article-title":"Perceptive locomotion through nonlinear model-predictive control","volume":"39","author":"Grandia","year":"2023","journal-title":"IEEE Transactions on Robotics"},{"key":"2025121104225077200_ref005","doi-asserted-by":"crossref","DOI":"10.1109\/LRA.2020.3011912","volume-title":"Deep Reinforcement Learning for Safe Local Planning of a Ground Vehicle in Unknown Rough Terrain","author":"Josef","year":"2020"},{"issue":"15","key":"2025121104225077200_ref006","doi-asserted-by":"crossref","first-page":"1","DOI":"10.3390\/s24154884","article-title":"Improved a* algorithm for mobile robots under rough terrain based on ground trafficability model and ground ruggedness model","volume":"24","author":"Liu","year":"2024","journal-title":"Sensors"},{"issue":"10","key":"2025121104225077200_ref007","doi-asserted-by":"crossref","first-page":"803","DOI":"10.1002\/rob.20204","article-title":"Scan registration for autonomous mining vehicles using 3D-NDT","volume":"24","author":"Magnusson","year":"2007","journal-title":"Journal of Field Robotics"},{"key":"2025121104225077200_ref008","first-page":"1","article-title":"Gem: online globally consistent dense elevation map for unstructured terrain","volume":"70","author":"Pan","year":"2020","journal-title":"IEEE Transactions on Instrumentation and Measurement"},{"issue":"2","key":"2025121104225077200_ref009","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1177\/0278364906075165","article-title":"An efficient extension to elevation maps for outdoor terrain mapping and loop closing","volume":"26","author":"Pfaff","year":"2007","journal-title":"The International Journal of Robotics Research"},{"key":"2025121104225077200_ref010","doi-asserted-by":"crossref","first-page":"3271","DOI":"10.1109\/TRO.2024.3400947","article-title":"Variable wheelbase control of wheeled mobile robots with Worm-Inspired creeping gait strategy","volume":"40","author":"Qi","year":"2024","journal-title":"IEEE Transactions on Robotics"},{"key":"2025121104225077200_ref011","article-title":"Enhancing maneuverability in a variable wheelbase wheeled mobile robot through dynamic steering curvature control","author":"Qi","year":"2024","journal-title":"Journal of Field Robotics"},{"issue":"7","key":"2025121104225077200_ref012","doi-asserted-by":"crossref","first-page":"1","DOI":"10.3390\/math12070992","article-title":"Transferability of multi-objective neuro-fuzzy motion controllers: towards cautious and courageous motion behaviors in rugged terrains","volume":"12","author":"Salih","year":"2024","journal-title":"Mathematics"},{"issue":"11","key":"2025121104225077200_ref013","doi-asserted-by":"crossref","first-page":"2592","DOI":"10.1017\/S0263574715000235","article-title":"Occupancy-elevation grid: an alternative approach for robotic mapping and navigation","volume":"34","author":"Souza","year":"2016","journal-title":"Robotica"},{"issue":"3","key":"2025121104225077200_ref014","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1007\/s43154-021-00056-3","article-title":"Planetary surface mobility and exploration: a review","volume":"2","author":"Thoesen","year":"2021","journal-title":"Current Robotics Reports"},{"key":"2025121104225077200_ref015","first-page":"345","article-title":"Current state of space robots","author":"Wang","year":"2021","journal-title":"Space Robotics"},{"key":"2025121104225077200_ref016","first-page":"1348","article-title":"Effect of compliance on ground adaptability of crawler mobile robots with Sub-crawlers","author":"Watanabe","year":"2020"},{"key":"2025121104225077200_ref017","doi-asserted-by":"crossref","DOI":"10.1109\/LRA.2021.3126904","volume-title":"Control of Rough Terrain Vehicles Using Deep Reinforcement Learning","author":"Wiberg","year":"2022"},{"key":"2025121104225077200_ref018","doi-asserted-by":"crossref","DOI":"10.1016\/j.robot.2024.104731","article-title":"Sim-to-real transfer of active suspension control using deep reinforcement learning","volume":"179","author":"Wiberg","year":"2024","journal-title":"Robotics and Autonomous Systems"},{"issue":"1","key":"2025121104225077200_ref019","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1109\/TRO.2022.3187818","article-title":"Efficient path planning in narrow passages for robots with ellipsoidal components","volume":"39","author":"Ruan","year":"2022","journal-title":"IEEE Transactions on Robotics"},{"issue":"1\/4","key":"2025121104225077200_ref020","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1007\/s10846-016-0362-z","article-title":"A review of global path planning methods for occupancy grid maps regardless of obstacle density","volume":"84","author":"Tsardoulias","year":"2016","journal-title":"Journal of Intelligent & Robotic Systems"},{"issue":"3","key":"2025121104225077200_ref021","doi-asserted-by":"crossref","first-page":"75","DOI":"10.14257\/ijsh.2014.8.3.07","article-title":"A multiple mobile robots path planning algorithm based on A-star and Dijkstra algorithm","volume":"8","author":"Zhang","year":"2014","journal-title":"International Journal of Smart Home"}],"container-title":["Industrial Robot: the international journal of robotics research and application"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IR-07-2024-0343\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/ir\/article-pdf\/52\/6\/833\/10834338\/ir-07-2024-0343en.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/www.emerald.com\/ir\/article-pdf\/52\/6\/833\/10834338\/ir-07-2024-0343en.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T09:23:01Z","timestamp":1765444981000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.emerald.com\/ir\/article\/52\/6\/833\/1253064\/Reinforcement-learning-for-generating-four-track"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,26]]},"references-count":21,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,11,25]]}},"URL":"https:\/\/doi.org\/10.1108\/ir-07-2024-0343","relation":{},"ISSN":["0143-991X","1758-5791"],"issn-type":[{"type":"print","value":"0143-991X"},{"type":"electronic","value":"1758-5791"}],"subject":[],"published":{"date-parts":[[2025,3,26]]}}}