E-Paper
Sign In
How UAE-made AI-powered robot can pick delicate strawberries, help during labour shortage
MBZUAI’s strawberry-picking robot works 24/7 — outpacing human workers limited to eight-hour shifts and relieving them of back strain
- PUBLISHED: Thu 27 Mar 2025, 6:49 PM
A robot can clean your house, drive your car and even deliver your food. Now, a robot could even pick delicate strawberries in a field, easing the burden on farmers.
The Mohamed Bin Zayed University of Artificial Intelligence’s robotics lab, Professor Dezhen Song, Deputy Department Chair of Robotics and Professor of Robotics demonstrated a strawberry-picking robot whose ‘brains’ were designed by the university’s robotics programme.
Recommended For You
The robotic harvester could offer a significant leap forward in agricultural automation, blending artificial intelligence with advanced robotics and precision agriculture technologies.
Stay up to date with the latest news. Follow KT on WhatsApp Channels.
The robot is fitted with high-resolution cameras and sensors that analyse plants in real-time, identifying ripe strawberries based on parameters such as colour, size, and shape. “The camera senses where it is and allows the robot to figure out how to grasp it… from which angle to grasp it.”
Advanced machine learning algorithms ensure accurate identification of ripe fruits while ignoring unripe or damaged ones, optimising yield and reducing waste.
One of the standout features of the robot is its adaptive grasping mechanism, which allows it to handle strawberries with remarkable precision. “You can see it’s not a rigid mount, so every time you put it back to a different position, you have to be able to know where it is on the fly to formulate a good grasping plane,” explained Professor Song while demonstrating the robot in action.
Once a ripe strawberry is identified, a robotic arm with a highly sensitive gripper gently picks the fruit without causing any damage. “This is a touch sensor, we call it a tactile sensor, it gives the robot the ability to know how hard to close the gripper without damaging the strawberry.
“We did a lot of tests on our own - how much force can be applied to the strawberry without damaging it, so we transferred that knowledge to the robot,” he added. In the future, the robot’s gripper can be easily replaced to match the size of other crops.
Unlike traditional factory robots that follow pre-programmed motions in fixed environments, the strawberry-picking robot dynamically adapts to its surroundings. “Not only is it not fixed, but also you have to control the contact force, so you don’t destroy the object.” The robot integrates active perception with its manipulative capabilities, allowing it to adjust its position or grip based on environmental factors such as light conditions, obstructions, or plant movement caused by wind, to ensure optimal efficiency in diverse agricultural settings.
Professor Song emphasised that the automation of farming is becoming increasingly necessary due to the global decline in human labour willing to take on agricultural work. “Generally speaking, in the precision or agricultural domain, there's a lack of human labour. So, using robots to replace humans is a trend, internationally, not just in developed countries, but also in developing countries. This is hard labour for people. If you ever go to the strawberry field to pick it for a day, you're going to feel your back is going to be hurting you.”
He explained that one of the key benefits of robotic farming is cost reduction. “We want to replace this type of job by using robots to make it healthier for humans. Also, to improve efficiency, reduce cost, and make them cheaper, because a lot of the cost of strawberries is associated with labour costs, not just the growing cost. So, if we can reduce that cost, it will be cheaper.”
Beyond reducing labour costs, Professor Song noted that agricultural robots could also enhance sustainability by limiting the need for harmful chemicals. “In general sense, if we do precision agriculture with robots, we can reduce the usage of chemicals and make the robot take care of the plants better.” He added that their team has also worked on robotic weeding operations. “We did the weeding operation without using chemicals, so you can do agriculture not only cheaper but also healthier and environmentally friendly as well.”
The professor highlighted the demographic shift in farming populations as another factor driving the need for automation. “Even in both China and the US, which are the largest markets in the world, the farmers' average age is over 53 years old. Nobody wants to do a farming job. Developing countries are facing a similar problem. A lot of young people, especially millennials, don't want to go to a farm field to work on this type of hard labour job.”
MBZUAI is currently in discussions with agricultural partners for potential commercialization of the harvesting robot. According to Professor Song, the next step would be to establish a startup that will focus on taking the technology to the market. The impact of MBZUAI’s strawberry-picking robot is already generating interest beyond academic circles. Professor Song recounted an interaction with a woman who had travelled from Cyprus specifically to see the robot. “She came here and asked, ‘Can I buy one for my husband? He is a strawberry farmer, and he really needs this,’” he shared. The strong demand from real farmers underscores the potential of robotics in transforming agriculture globally. “I can imagine an army of this (robot) working at strawberry farms everywhere in the world – not just in Abu Dhabi. That's my vision.’
The economic advantage of the robot is particularly evident when compared to the UK farming market, where labour shortages and rising wages are pressing concerns. He explained that a robotic solution could help farmers cut down on labour costs while ensuring continuous, efficient harvesting.
Unlike human workers, the robot does not require breaks, sick days, or seasonal recruitment, making it a potentially more sustainable and cost-effective alternative in the long run. “The robot can pick a strawberry, one strawberry per nine seconds, that cost is equivalent to the UK human labour for a whole day, eight-hour shift, but this robot can do 24-hour,” the professor concluded.
MBZUAI is constantly collaborating with government entities and major agritech companies to drive AI-driven advancements in agriculture. Through its Memorandum of Understanding with ADQ, the university engages in research and development projects with companies like Silal, focusing on predictive analytics to enhance crop yields and strengthen food security.
