Working In Agriculture Is Tough – Why Haven’t Robots Already Taken Over?
Farming can be back-breaking work. And things like fruit-picking can cause an alarming number of health problems: from pesticide-related illnesses and skin diseases to respiratory conditions. Wouldn’t it be better for machines to do the work?
Undoubtedly. But unfortunately robotics in agriculture is hard. Plants, trees and animals are organic, so the robots have to work on farms where every piece of crop is different and every animal is different. Crops grow, animals move, and they’re both delicate and difficult to control. For example, picking berries is very dexterous work requiring nimble hands and awareness of damaging both fruit and plant. On top of this, every berry is a slightly different size, a different level of ripeness and covered by a differing number of leaves.
Dr Matthew Howard, a senior lecturer in robotics at King’s College, London, is researching how to use robotics to overcome the challenges of large fresh produce growers. Speaking at a recent Agri-Tech East event, he explained that because every plant is different, no-one can teach a robot every single scenario. So his research tries to ‘generalise’ this learning. For example, teach a robot how to pick up a few strawberries of different shapes and sizes, then present it with a strawberry it’s never seen before, and let the robot fill in the gaps in its knowledge to pick it up.
As the robots learn, we can learn from them. Digital technologies that capture, process and interpret data are already transforming agriculture, enabling the sector to become more productive and efficient than ever before. PA’s recent research explored the current landscape and the best ways to speed up progress. We think new collaborations between the different organisations involved will be the answer. That includes big names in the machine and equipment and agriscience sectors and technology companies and start-ups.
Making use of robots and data mining is critical for farmers to cut through the complexity, make things easier on themselves physically, and make sense of lots of important information – to help feed the world. Research like Matthew’s may still be years away from being adopted in the mainstream, but it’s important that we keep doing it and it’s important we accelerate how that research is taken to farmers.