In this study, we propose a compact autonomous roadside weeding robot, R4 (Ritsumeikan Road-weed Removal Robot), aimed at reducing human labor in road infrastructure maintenance. Conventional weeding robots are often designed for agricultural or horticultural applications and are unsuitable for operation in roadside environments where drivable areas are limited. The proposed robot has a compact structure with a width of 603 mm, depth of 512 mm, and height of 368 mm, minimizing the risk of contact with vehicles. A low-cost autonomous navigation system was developed using a single RGB camera to determine a path along the roadside independent of lighting and pavement conditions. Furthermore, a lightweight weed detection algorithm was developed to deploy the weeding mechanism only when weeds are present, enabling safe operation without human intervention. The experiments conducted in both indoor and outdoor environments demonstrated the stability of the autonomous navigation system, efficiency of the weed detection algorithm, and practical applicability of the proposed robot.

1. [1] N. Kheradmandi and V. Mehranfar, “A critical review and comparative study on image segmentation-based techniques for pavement crack detection,” Construct. Build Mater., Vol.321, Article No.126162, 2022. https://doi.org/10.1016/j.conbuildmat.2021.126162
2. [2] T. Cao, Y. Wang, and S. Liu, “Pavement Crack Detection Based on 3D Edge Representation and Data Communication with Digital Twins,” IEEE Trans. Intell. Transp. Syst., Vol.24, No.7, pp. 7697-7706, 2023. https://doi.org/10.1109/TITS.2022.3194013
3. [3] B. Goffin, N. Banthia, and N. Yonemitsu, “Use of infrared thermal imaging to detect corrosion of epoxy coated and uncoated rebar in concrete,” Construct. Build Mater., Vol.263, Article No.120162, 2020. https://doi.org/10.1016/j.conbuildmat.2020.120162
4. [4] E. Ichi and S. Dorafshan, “Effectiveness of infrared thermography for delamination detection in reinforced concrete bridge decks,” Autom. Constr., Vol.142, Article No.104523, 2022. https://doi.org/10.1016/j.autcon.2022.104523
5. [5] S. M. Blunt, “Trees and pavements—Are they compatible?,” Arboric. J., Vol.31, No.2, pp. 73-80, 2008. https://doi.org/10.1080/03071375.2008.9747522
6. [6] F. Giuliani, F. Autelitano, E. Degiovanni, and A. Montepara, “DEM modelling analysis of tree root growth in street pavements,” Int. J. Pavement Eng., Vol.18, No.1, pp. 1-10, 2017. https://doi.org/10.1080/10298436.2015.1019495
7. [7] P.-R. Hirt, J. Holtkamp, L. Hoegner, Y. Xu, and U. Stilla, “Occlusion detection of traffic signs by voxel-based ray tracing using highly detailed models and MLS point clouds of vegetation,” Int. J. Appl. Earth Obs. Geoinform., Vol.114, Article No.103017, 2022. https://doi.org/10.1016/j.jag.2022.103017
8. [8] R. Li, J. Yu, F. Li, R. Yang, Y. Wang, and Z. Peng, “Automatic bridge crack detection using Unmanned aerial vehicle and Faster R-CNN,” Construct. Build Mater., Vol.362, Article No.129659, 2023. https://doi.org/10.1016/j.conbuildmat.2022.129659
9. [9] J. Zhu, Y. Wu, and T. Ma, “Multi-Object Detection for Daily Road Maintenance Inspection with UAV Based on Improved YOLOv8,” IEEE Trans. Intell. Transp. Syst., Vol.25, No.11, pp. 16548-16560, 2024. https://doi.org/10.1109/TITS.2024.3437770
10. [10] M. Saska, M. Hess, and K. Schilling, “Efficient airport snow shoveling by applying autonomous multi-vehicle formations,” Proc. of 2008 IEEE Int. Conf. Robot. Autom, pp. 1684-1690, 2008. https://doi.org/10.1109/ROBOT.2008.4543443
11. [11] J. Wu, L. Liu, Q. Wu, J. He, and W. Li, “Autonomous Robotic System for Snow Removal in Large-Scale Photovoltaic Arrays,” IEEE Access, Vol.13, pp. 141547-141559, 2025. https://doi.org/10.1109/ACCESS.2025.3597534
12. [12] Y. Iwano, A. Tanaka, and K. Iizuka, “Development of a Flail-Type Mowing System,” J. Robot. Mechatron., Vol.37, No.3, pp. 555-562, 2025. https://doi.org/10.20965/jrm.2025.p0555
13. [13] D. He, H. Li, and J. Yin, “LIGO: A Tightly Coupled LiDAR-Inertial-GNSS Odometry Based on a Hierarchy Fusion Framework for Global Localization with Real-Time Mapping,” IEEE Trans. Robot., Vol.41, pp. 1224-1244, 2025. https://doi.org/10.1109/TRO.2025.3530298
14. [14] W. Yuan, Z. Li, and C.-Y. Su, “RGB-D sensor-based visual SLAM for localization and navigation of indoor mobile robot,” Proc. of 2016 Int. Conf. Adv. Robot. Mechatron., pp. 82-87, 2016. https://doi.org/10.1109/ICARM.2016.7606899
15. [15] T. Shan, B. Englot, C. Ratti, and D. Rus, “LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mapping,” Proc. of 2021 IEEE Int. Conf. Robot. Autom., pp. 5692-5698, 2021. https://doi.org/10.1109/ICRA48506.2021.9561996
16. [16] U. Franke, D. Gavrila, S. Gorzig, F. Lindner, F. Puetzold, and C. Wohler, “Autonomous driving goes downtown,” IEEE Intell. Syst. Appl., Vol.13, No.6, pp. 40-48, 1998. https://doi.org/10.1109/5254.736001
17. [17] M. Aly, “Real time detection of lane markers in urban streets,” Proc. of 2008 IEEE Intell. Veh. Symp., pp. 7-12, 2008. https://doi.org/10.1109/ivs.2008.4621152
18. [18] M. Bojarski, D. Del Testa, D. Dworakowski, B. Firner, B. Flepp, P. Goyal, L. D. Jackel, M. Monfort, U. Muller, J. Zhang, X. Zhang, J. Zhao, and K. Zieba, “End to End Learning for Self-Driving Cars,” arXiv preprint, arXiv:1604.07316, 2016. https://doi.org/10.48550/arXiv.1604.07316
19. [19] V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,” IEEE Trans. Pattern Anal. Mach. Intell., Vol.39, No.12, pp. 2481-2495, 2017. https://doi.org/10.1109/tpami.2016.2644615
20. [20] X. Fan, T. Sun, X. Chai, and J. Zhou, “YOLO-WDNet: A lightweight and accurate model for weeds detection in cotton field,” Comput. Electron. Agric., Vol.225, Article No.109317, 2024. https://doi.org/10.1016/j.compag.2024.109317
21. [21] K. Wang, X. Hu, H. Zheng, M. Lan, C. Liu, Y. Liu, L. Zhong, H. Li, and S. Tan, “Weed detection and recognition in complex wheat fields based on an improved YOLOv7,” Front. Plant Sci., Vol.15, Article No.1372237, 2024. https://doi.org/10.3389/fpls.2024.1372237
22. [22] A. Olsen, D. A. Konovalov, B. Philippa, P. Ridd, J. C. Wood, J. Johns, W. Banks, B. Girgenti, O. Kenny, J. Whinney et al., “DeepWeeds: A multiclass weed species image dataset for deep learning,” Scientific Reports, Vol.9, No.1, Article No.2058, 2019. https://doi.org/10.1038/s41598-018-38343-3
23. [23] S. Haug and J. Ostermann, “A crop/weed field image dataset for the evaluation of computer vision based precision agriculture tasks,” Proc. of 2014 Eur. Conf. Comput. Vis., pp. 105-116, 2014. https://doi.org/10.1007/978-3-319-16220-1_8
24. [24] Y. Matsushita, K. K. Gupta, Y. Fujii, D. T. Tran, and J.-H. Lee, “Development of an Autonomous Weeding Robot for Roadside Weeds,” Proc. of 2025 Int. Conf. Ubiquitous Robot., pp. 444-450, 2025. https://doi.org/10.1109/UR65550.2025.11078083
25. [25] R. Wallace, A. Stentz, C. Thorpe, H. Moravec, W. Whittaker, and T. Kanade, “First Results in Robot Road-Following,” Proc. of 9th Int. Jt. Conf. Artif. Intell., Vol.2, pp. 1089-1095, 1985.
26. [26] X. Zhao, W. Ding, Y. An, Y. Du, T. Yu, M. Li, M. Tang, and J. Wang, “Fast segment anything,” arXiv preprint, arXiv:2306.12156, 2023. https://doi.org/10.48550/arXiv.2306.12156
27. [27] A. Howard, M. Sandler, B. Chen, W. Wang, L.-C. Chen, M. Tan et al., “Searching for mobilenetv3,” Proc. of 2019 IEEE/CVF Int. Conf. Comput. Vis., pp. 1314-1324, 2019. https://doi.org/10.1109/iccv.2019.00140
28. [28] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein et al., “Imagenet large scale visual recognition challenge,” Int. J. Comput. Vis., Vol.115, No.3, pp. 211-252, 2015. https://doi.org/10.1007/s11263-015-0816-y
29. [29] R. Khanam and M. Hussain, “Yolov11: An overview of the key architectural enhancements,” arXiv preprint, arXiv:2410.17725, 2024. https://doi.org/10.48550/arXiv.2410.17725
30. [30] K. Han, Y. Wang, Q. Tian, J. Guo, C. Xu, and C. Xu, “Ghostnet: More features from cheap operations,” Proc. of 2020 IEEE/CVF Conf. Comput. Vis. Pattern Recognit., pp. 1580-1589, 2020. https://doi.org/10.1109/cvpr42600.2020.00165
31. [31] C. Rother, V. Kolmogorov, and A. Blake, ““GrabCut” interactive foreground extraction using iterated graph cuts,” ACM Trans. Graph., Vol.23, No.3, pp. 309-314, 2004. https://doi.org/10.1145/1015706.1015720
32. [32] T.-Y. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollár, and C. L. Zitnick, “Microsoft coco: Common objects in context,” 2014 Eur. Conf. Comput. Vis., pp. 740-755, 2014. https://doi.org/10.1007/978-3-319-10602-1_48
33. [33] “Farm Wise.” https://farmwise.io/ [Accessed October 13, 2025]
34. [34] “Robot Lawn mowers.” https://www.husqvarna.com/us/robotic-lawn-mowers/ [Accessed October 13, 2025]
35. [35] “ARA Field Sprayer.” https://ecorobotix.com/crop-care/ara-field-sprayer/ [Accessed October 13, 2025]
36. [36] “See & Spray.” https://www.deere.com/en/sprayers/see-spray/ [Accessed October 13, 2025]
37. [37] “LASERWEEDER G2.” https://carbonrobotics.com/laserweeder-g2 [Accessed October 13, 2025]
38. [38] “The Rover for Weeding.” https://romi-project.eu/tools/weeding/ [Accessed October 13, 2025]
39. [39] “FarmDroid FD20.” https://farmdroid.com/products/farmdroid-fd20/ [Accessed October 13, 2025]
40. [40] “Landroid robot lawnmower by Worx.” https://eu.worx.com/en/landroid/ [Accessed October 13, 2025]
41. [41] “ultralytics ultralytics.” https://github.com/ultralytics/ultralytics [Accessed October 13, 2025]