A cutting-edge AI system is revolutionizing the way fruit farmers predict harvest sizes, making crop yields more efficient, sustainable, and profitable. This innovative tool, developed by researchers at the National Robotarium in collaboration with scientific partners in Chile and Spain, accurately estimates the number of flowers on fruit trees using images taken with a standard smartphone. By recognizing patterns and features such as the edges and shapes of petals, even when overlapping or partially obscured, the AI system provides precise yield forecasts months in advance.

When tested on peach orchards in Catalonia, Spain, the AI demonstrated a 90% accuracy in predicting flower counts, a significant improvement over current manual methods which can have error rates of 30-50%. This accuracy allows growers to optimize water use, allocate resources more efficiently, and better plan for harvesting and distribution logistics. This is particularly crucial given that agriculture accounts for 65% of the world’s fresh water usage, with nearly half of it being wasted, and approximately 45% of fruit and vegetables lost annually in the global supply chain.

Researchers from the National Robotarium, the UK’s center for robotics and AI based in Edinburgh, will validate the AI’s predictions against the actual peach harvest in September 2024. If effective, this approach could be adapted for other crops like apples, pears, and cherries, benefiting fruit growers across Britain, Europe, and beyond.
Supported by the Data-Driven Innovation initiative, with funding from the UK and Scottish Governments, the National Robotarium aims to make Edinburgh the data capital of Europe. Dr. Fernando Auat Cheein, associate professor in robotics and autonomous systems at the National Robotarium, emphasized the ease with which the AI integrates with traditional farming practices. Farmers appreciate the simplicity and accuracy of the flower counting AI, which helps them make informed decisions about crop management, resource optimization, and environmental impact reduction.
This research, involving collaboration with Chilean universities, positions the National Robotarium at the forefront of agricultural innovation, using AI and robotics to tackle real-world challenges and create tangible benefits for farmers and the environment.