Autonomous tomato harvesting robot using YOLO model

Shahbaz, Muhammad Sikander (author), Aqeel, Muhammad (author), Iqbal, Muhammad (author), Sargana, Hussain Mahmood (author), Gull, Areesha (author)

Manually harvesting tomatoes is costly and time-consuming. This work proposes a tomato-harvesting robot to solve the time lapse. This work centers on the motion integration of a four-wheel steering system for robots. A collection of 1268 photos of ripe and unripe tomatoes was assembled in this work. By applying image preprocessing methods such as rotation, flipping, and resizing, we have increased and improved the size and quality of the dataset. The two primary components of our study are location detection and tomato identification. The former uses a focal length stereo vision cam-era to estimate tomatoes’ depth and stream live video in real time. We compute the 3D position and angle of the robotic arm through servo motors to reach target points using inverse kinematics. With 94 % harvesting efficiency, our system harvests ripe tomatoes in less than 8 seconds and stores them in a storage tank. For the detection and classification of ripe and un-ripe tomatoes, YOLOv5s, YOLOv8s, YOLOv9s, and YOLOv10s were used. Accuracy, precision, and recall were used to evaluate model performance; YOLOv8s is found to have the best evalu-ation metrics.

• Language: English
• Keywords: inverse kinematics, stereo vision, tomato harvesting robot, YOLO
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: BF - Biotechnical Faculty
• Publication status: Published
• Publication version: Version of Record
• Year: 2026
• Number of pages: 12 str.
• Numbering: Vol. 122, no. 3
• PID: 20.500.12556/RUL-189406
• UDC: 635
• ISSN on article: 1854-1941
• DOI: 10.14720/aas.2026.122.3.21385
• COBISS.SI-ID: 293853443
• Publication date in RUL: 06.10.2026

Record is a part of a journal

• Title: Acta agriculturae Slovenica
• Publisher: Biotehniška fakulteta, Oddelek za agronomijo in Oddelek za zootehniko, Biotehniška fakulteta, Oddelek za agronomijo in Oddelek za zootehniko, Univerza v Ljubljani, Biotehniška fakulteta, Založba Univerze v Ljubljani
• ISSN: 1854-1941
• COBISS.SI-ID: 219581184

Licences

• License: CC BY 4.0, Creative Commons Attribution 4.0 International
• Link: http://creativecommons.org/licenses/by/4.0/
• Description: This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.

Secondary language

• Language: Slovenian
• Title: Avtonomni robot za obiranje paradižnika z uporabo modela YOLO
• Keywords: inverzna kinematika, stereo vid, robot za obiranje paradižnikov