https://phys.org/news/2025-12-robocrop-robots-tomatoes.html Phys.org Topics * Week's top * Latest news * Unread news * Subscribe [ ] Science X Account [ ] [ ] Sign In Sign in with Forget Password? Not a member? Sign up Learn more * Nanotechnology * Physics * Earth * Astronomy & Space * Chemistry * Biology * Other Sciences * Medical Xpress Medicine * Tech Xplore Technology * * share this! * 23 * Tweet * Share * Email 1. Home 2. Biology 3. Biotechnology 1. Home 2. Biology 3. Agriculture * * * --------------------------------------------------------------------- December 8, 2025 The GIST RoboCrop: Teaching robots how to pick tomatoes by Osaka Metropolitan University edited by Robert Egan [Robert] Robert Egan associate editor Meet our editorial team Behind our editorial process Editors' notes This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility: fact-checked trusted source proofread RoboCrop: Teaching robots how to pick tomatoes The left image shows the tomato-picking robot and camera. The right image shows a 'robot-eye view' of the tomatoes. Red represents mature fruits, green indicates immature fruits, and blue indicates selected harvesting targets. Credit: Osaka Metropolitan University In the agricultural sector, labor shortages are increasing the need for automated harvesting using robots. However, some fruits, like tomatoes, are tricky to harvest. Tomatoes typically bear fruit in clusters, requiring robots to pick the ripe ones while leaving the rest on the vine, demanding advanced decision-making and control capabilities. How robots learn to pick tomatoes To teach robots how to become tomato pickers, Osaka Metropolitan University Assistant Professor Takuya Fujinaga, Graduate School of Engineering, programmed them to evaluate the ease of harvesting for each tomato before attempting to pick it. Fujinaga's new model uses image recognition paired with statistical analysis to evaluate the optimal approach direction for each fruit. The system involves image processing/vision of the fruit, its stems, and whether it is concealed behind another part of the plant. These factors inform robot control decisions and help it choose the best approach. The findings are published in Smart Agricultural Technology . Shifting from recognition to harvest-ease The model represents a shift in focus from the traditional 'detection /recognition' model to what Fujinaga calls a 'harvest-ease estimation'. "This moves beyond simply asking 'can a robot pick a tomato?' to thinking about 'how likely is a successful pick?', which is more meaningful for real-world farming," he explained. When tested, Fujinaga's new model demonstrated an 81% success rate, far above predictions. Notably, about a quarter of the successes were tomatoes that were successfully harvested from the right or left side that had previously failed to be harvested by a front approach. This suggested that the robot changed its approach direction when it initially struggled to pick the fruit. Implications for the future of farming Ultimately, Fujinaga's research highlights the nuance involved in fruit-picking for robots with factors including fruit clustering, stem geometry, background leaves, and occlusion all being important. "This research establishes 'ease of harvesting' as a quantitatively evaluable metric, bringing us one step closer to the realization of agricultural robots that can make informed decisions and act intelligently," he said. Fujinaga sees a future where robots will be able to independently determine whether crops are ready for harvest. "This is expected to usher in a new form of agriculture where robots and humans collaborate," he explained. "Robots will automatically harvest tomatoes that are easy to pick, while humans will handle the more challenging fruits." More information: Takuya Fujinaga, Realizing an intelligent agricultural robot: An analysis of the ease of tomato harvesting, Smart Agricultural Technology (2025). DOI: 10.1016/ j.atech.2025.101538 Provided by Osaka Metropolitan University Citation: RoboCrop: Teaching robots how to pick tomatoes (2025, December 8) retrieved 10 December 2025 from https://phys.org/news/ 2025-12-robocrop-robots-tomatoes.html This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only. --------------------------------------------------------------------- Explore further Farm robot autonomously navigates and harvests among raised beds --------------------------------------------------------------------- 23 shares * Facebook * Twitter * Email Feedback to editors * Featured * Last Comments * Popular Used cooking oil yields super strong glue and recyclable plastics 4 hours ago 1 Analysis of Diplodocus dinosaur scales reveals possible speckled color patterning 5 hours ago 0 Searchable Bronze Age site database could help answer key questions about ancient Anatolia 5 hours ago 0 Ancient Roman officers may have kept pet monkeys to highlight their status 7 hours ago 0 Neutrino observatories show promise for detecting light dark matter 8 hours ago 7 --------------------------------------------------------------------- [gif] Roman urbanism was bad for health, new study confirms 1 minute ago [gif] Reddit field experiment examines what distinguishes lurkers from power users 1 hour ago [gif] Mitochondrial enzyme's atomic-level structure reveals how it processes RNA 1 hour ago [gif] A new reference brain could make the clonal raider ant a go-to model species for neuroscience 1 hour ago [gif] Spending less can deliver more climate-friendly nutrition 2 hours ago [gif] New circoviruses discovered in pilot whales and orcas from the North Atlantic 2 hours ago [gif] Lunar soil analyses reveal how space weathering shapes the moon's ultraviolet reflectance 2 hours ago [gif] Descriptions of mollusks in the Global South are still, for the most part, the result of 'parachute science' 2 hours ago [gif] Slow changes in radio scintillation can nudge pulsar timing by billionths of a second 2 hours ago [gif] Einstein's theory comes wrapped up with a bow: Astronomers spot star 'wobbling' around black hole 3 hours ago --------------------------------------------------------------------- Relevant PhysicsForums posts Influenza A H5, an avian flu, first time in a human Nov 30, 2025 Magnetoreception in Animals Nov 30, 2025 PFAS and Power Lines Cause Cancer? Nov 25, 2025 Viral (RNA and DNA) polymerases, inhibitors and antiviral drugs Nov 17, 2025 Mold in Plastic Water Bottles? What does it eat? Nov 10, 2025 Who knows if Scotopic Sensitivity Syndrome is real or not? Nov 10, 2025 More from Biology and Medical --------------------------------------------------------------------- * Related Stories [gif] Farm robot autonomously navigates and harvests among raised beds Apr 16, 2025 [gif] Using fake raspberries to train robots how to pick real ones Jun 26, 2023 [gif] Team develops a system of robots that use teamwork to pick fruit and transport it autonomously Mar 16, 2023 [gif] Robotic harvester uses AI vision and soft grippers to pick hidden strawberries Sep 4, 2025 [gif] Silicone raspberry used to train harvesting robots Mar 30, 2022 [gif] Harvest smarter, not harder: Machine learning meets tomato farming Jan 23, 2025 * Recommended for you [gif] More yield through heterosis: Researchers decode gene interaction behind hybrid vigor 3 hours ago [gif] Modified bacterial transport system imports artificial amino acids for efficient designer protein creation 6 hours ago [gif] Melatonin wakes up plants, stimulating growth and boosting stress tolerance Dec 9, 2025 [gif] Simplified lab process produces potent nerve-blocking molecules found in shellfish Dec 9, 2025 [gif] Microneedle system delivers biofertilizer directly into plants, boosting growth with less waste Dec 9, 2025 [gif] Implantable sensor uses engineered bacteria for wireless molecular tracking Dec 8, 2025 Load comments (0) Get Instant Summarized Text (Gist) A new robotic model for tomato harvesting uses image recognition and statistical analysis to estimate the ease of picking each fruit, considering factors like clustering, stem geometry, and occlusion. This approach achieved an 81% success rate, with robots adapting their picking direction for difficult tomatoes, supporting more efficient and intelligent automated harvesting. This summary was automatically generated using LLM. Full disclaimer Let us know if there is a problem with our content Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form. For general feedback, use the public comments section below (please adhere to guidelines). Please select the most appropriate category to facilitate processing of your request [-- please select one -- ] [ ] [ ] [ ] [ ] [ ] Your message to the editors [ ] Your email (optional, only if you'd like a response) [ ] Send Feedback Thank you for taking time to provide your feedback to the editors. Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages. E-mail the story RoboCrop: Teaching robots how to pick tomatoes Your friend's email [ ] Your email [ ] [ ] I would like to subscribe to Science X Newsletter. Learn more Your name [ ] Note Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Phys.org in any form. [ ] [ ] [ ] [ ] [ ] [ ] [ ] Your message [ ] Send Newsletter sign up Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties. [ ] Subscribe More information Privacy policy Donate and enjoy an ad-free experience We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account. Remove ads Maybe later Medical Xpress Medical Xpress Medical research advances and health news Tech Xplore Tech Xplore The latest engineering, electronics and technology advances Science X Science X The most comprehensive sci-tech news coverage on the web Newsletters [ ] Subscribe Science X Daily and the Weekly Email Newsletter are free features that allow you to receive your favorite sci-tech news updates in your email inbox Follow us * * * * * * * * Top * Home * Search * Mobile version * Help * FAQ * About * Contact * Support us * Science X Account * Archive * News wire * Android app * iOS app * RSS feeds * Push notification (c) Phys.org 2003 - 2025 powered by Science X Network Privacy policy Terms of use E-mail newsletter [ ] Subscribe Follow us * * * * It appears that you are currently using Ad Blocking software. What are the consequences? x Quantcast