Napredne navigacijske tehnike in postopki umetne inteligence: Ključne strategije ekipe Carbonite na tekmovanju Field Robot Event 2023
DOI:
https://doi.org/10.18690/agricsci.22.1-2.2Ključne besede:
kmetijska robotika, precizno kmetijstvo, trajnost, umetna inteligencaPovzetek
Eden od pristopov za reševanje trenutnih izzivov v kmetijstvu, ki jih povzročajo podnebne spremembe, naraščajoče svetovno prebivalstvo in izguba biotske raznovrstnosti, je precizno kmetijstvo, pri katerem igrajo robotski sistemi ključno vlogo. Tekmovanje poljskih robotov Field Robot Event (FRE) 2023 je zato izzvalo študentske ekipe, da razvijejo in izboljšajo avtonomne kmetijske robote. V tem prispevku predstavljamo izboljšave našega poljskega robota »Carbonite«, ki ga razvijamo na Schülerforschungszentrum (SFZ) Südwürttemberg. Naša lahka in kompaktna zasnova robota, podprta z naprednim in učinkovitim navigacijskim algoritmom, je omogočila hitro premikanje robota po polju. Poleg tega pa smo uvedli novo razvit sistem za ciljno in natančno uporabo vode, gnojil in herbicidov, ki temelji na naprednem algoritmu za zaznavanje prisotnosti rastlin, kar preprečuje nepotrebno porabo virov. Prav tako smo pripravili postopke umetne inteligence za prepoznavanje objektov, ki temeljijo na modelih You Only Look Once (YOLO), kar robotu omogoča ustrezno odzivanje glede na vrsto ovire. Carbonite je tako osvojil prvo mesto tako v nalogi navigacije kot tudi v nalogi obdelave rastlin, k čemur je prispevala lahka zasnova in posledično visoka hitrost premikanja robota, vse to pa je pripomoglo k skupni zmagi skupine na FRE 2023.
Prenosi
Literatura
1. ASUSTeK Computer Inc. (n.d.). EA-AC87—Support. Retrieved March 29, 2025, from https://www.asus.com/supportonly/eaac87/helpdesk_knowledge/
2. Benenson, R., Popov, S., & Ferrari, V. (2019). Large-scale interactive object segmentation with human annotators. arXiv:1903.10830. https://doi.org/10.48550/arXiv.1903.10830
3. Bjelonic, M. (2016). Leggedrobotics/darknet_ros [C++]. Robotic Systems Lab - Legged Robotics at ETH Zürich. https://github.com/leggedrobotics/darknet_ros
4. Bochkovskiy, A. (2013). AlexeyAB/darknet [C]. https://github.com/AlexeyAB/darknet
5. Bosch Sensortec GmbH. (n.d.). Smart Sensor BNO055. Bosch Sensortec. Retrieved from https://www.bosch-sensortec.com/products/smart-sensor-systems/bno055/
6. Cardinale, B. J., Duffy, J. E., Gonzalez, A., Hooper, D. U., Perrings, C., Venail, P., Narwani, A., Mace, G. M., Tilman, D., Wardle, D. A., Kinzig, A. P., Daily, G. C., Loreau, M., Grace, J. B., Larigauderie, A., Srivastava, D. S., & Naeem, S. (2012). Biodiversity loss and its impact on humanity. Nature, 486(7401), 59–67. https://doi.org/10.1038/nature11148
7. Field Robot Event. (2022, October 21). Field Robot Event 2023 in Slovenia! – Field Robot Event. https://fieldrobot.nl/event/index.php/2022/10/21/field-robot-event-goes-to-slovenia/
8. Field Robot Event. (2023a). FRE2023-RESULTS_FINAL.pdf. Field Contest 13.06.2023 – 15.06.2023 FRE 2023 Results. Retrieved from: https://fieldrobot.nl/event/wp-content/uploads/2023/06/FRE2023-RESULTS_FINAL.pdf
9. Field Robot Event. (2023b). Task 1 Navigation – Field Robot Event. Retrieved from: https://fieldrobot.nl/event/index.php/contest-hybrid/tasks-h/
10. Field Robot Event. (2023c). Task 2 treating (spraying) the plants – Field Robot Event. Retrieved from: https://fieldrobot.nl/event/index.php/contest-hybrid/task-h1/
11. Field Robot Event. (2023d). Task 3 sensing and recognizing possible obstacles – Field Robot Event. Retrieved from: https://fieldrobot.nl/event/index.php/contest-hybrid/task-h2/
12. Field Robot Event (2023e). Task 4 Static and dynamic obstacles – Field Robot Event. https://fieldrobot.nl/event/index.php/contest-hybrid/task-h3/
13. Field Robot Event (2023f). Task 5 Freestyle – Field Robot Event. Retrieved from: https://fieldrobot.nl/event/index.php/task-5-freestyle/
14. Foley, J. A., Ramankutty, N., Brauman, K. A., Cassidy, E. S., Gerber, J. S., Johnston, M., Mueller, N. D., O’Connell, C., Ray, D. K., West, P. C., Balzer, C., Bennett, E. M., Carpenter, S. R., Hill, J., Monfreda, C., Polasky, S., Rockström, J., Sheehan, J., Siebert, S., Tilman, D., & Zaks, D. P. M. (2011). Solutions for a cultivated planet. Nature, 478(7369), 337–342. https://doi.org/10.1038/nature10452
15. Gil, G., Casagrande, D. E., Cortés, L. P., & Verschae, R. (2023). Why the low adoption of robotics in the farms? Challenges for the establishment of commercial agricultural robots. Smart Agricultural Technology, 3, 100069. https://doi.org/10.1016/j.atech.2022.100069
16. Intel Corporation. (n.d.). Introducing the Intel® RealSenseTM Depth Camera D455. Intel® RealSenseTM Depth and Tracking Cameras. Retrieved from: https://www.intelrealsense.com/depth-camera-d455/
17. Keenso. (n.d.). Mini-Wasserpumpe, 12 V DC, 6 W, Tauchpumpe, ohne Bürste, energiesparend, für Aquarium, Brunnen, kleine Fischteiche, Solarsystem: Amazon.de: Haustier. Retrieved from: https://www.amazon.de/Submersible-Without-Energy-Aquarium-Fountain/dp/B07VGQ8KJV
18. Kuznetsova, A., Rom, H., Alldrin, N., Uijlings, J., Krasin, I., Pont-Tuset, J., Kamali, S., Popov, S., Malloci, M., Kolesnikov, A., Duerig, T., & Ferrari, V. (2020). The open images dataset V4: Unified image classification, object detection, and visual relationship detection at scale. International Journal of Computer Vision, 128(7), 1956–1981. https://doi.org/10.1007/s11263-020-01316-z
19. Lin, T.-Y., Maire, M., Belongie, S., Bourdev, L., Girshick, R., Hays, J., Perona, P., Ramanan, D., Zitnick, C. L., & Dollár, P. (2015). Microsoft COCO: Common Objects in Context. arXiv:1405.0312. https://doi.org/10.48550/arXiv.1405.0312
20. Malhi, G. S., Kaur, M., & Kaushik, P. (2021). Impact of climate change on agriculture and its mitigation strategies: A review. Sustainability, 13(3), 1318. https://doi.org/10.3390/su13031318
21. Mayer, J., Fauser, K., Schupp, J., & Locher, L. (2022). Carbonite–Team Carbonite. Proceedings of the 18th Field Robot Event 2021, 51–57. Retrieved from: https://www.fieldrobot.com/event/wp-content/uploads/2022/02/Proceedings_FRE2021.pdf
22. Nelson, G. C., Rosegrant, M. W., Koo, J., Robertson, R. D., Sulser, T., Zhu, T., Ringler, C., Msangi, S., Palazzo, A., Batka, M., Magalhaes, M., Valmonte-Santos, R., Ewing, M., & Lee, D. R. (2009). Climate change: Impact on agriculture and costs of adaptation. International Food Policy Research Institute. https://doi.org/10.2499/0896295354
23. NVIDIA Corporation. (n.d.). Jetson AGX Xavier Developer Kit by NVIDIA | 945-82972-0045-000. Arrow.Com.
Retrieved March 29, 2025, from https://www.arrow.com/en/products/945-82972-0045-000/nvidia
24. Open Robotics. (2024a, January 9). actionlib—ROS Wiki. Retrieved from: https://wiki.ros.org/actionlib
25. Open Robotics. (2024b, August 31). noetic—ROS Wiki. https://wiki.ros.org/noetic
26. OpenAI. (n.d.). Overview—OpenAI API. Retrieved from: https://platform.openai.com
27. Redmon, J. (2013, 2016). Darknet: Open Source Neural Networks in C.
28. Redmon, J., & Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv:1804.02767. https://doi.org/10.48550/arXiv.1804.02767
29. Robitronic Electronic Ges.m.b.H. (n.d.). Robitronic Platinium Brushless Motor 1/8 10.5T. Robitronic RC Car Online Shop - Power for Winners. Retrieved from: https://shop.robitronic.com/en/robitronic-platinium-brushless-motor-r03204
30. Rusu, R. B., & Cousins, S. (2011). 3D is here: Point Cloud Library (PCL). IEEE International Conference on Robotics and Automation (ICRA), Shanghai, China. Retrieved from: https://github.com/PointCloudLibrary/pcl
31. SICK AG. (n.d.). TiM571-2050101—TiM | SICK. Retrieved from: https://www.sick.com/us/en/catalog/products/lidar-and-radar-sensors/lidar-sensors/tim/tim571-2050101/p/p412444
32. The Qt Company. (n.d.). Qt 5.15. Retrieved from: https://doc.qt.io/qt-5/
33. Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2022). YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv:2207.02696. Retrieved from: https://doi.org/10.48550/arXiv.2207.02696
34. Worldsemi. (n.d.). WS2812B.pdf. Retrieved from: https://cdn-shop.adafruit.com/datasheets/WS2812B.pdf
Prenosi
Objavljeno
Številka
Rubrika
Licenca
Avtorske pravice (c) 2025 Samuel Mannchen, Jonas Mayer, Janis Lion Schőnegg, Klara Fauser
To delo je licencirano pod Creative Commons Priznanje avtorstva-Nekomercialno-Brez predelav 4.0 mednarodno licenco.