A team of researchers from Harper Adams University in the UK say they are the first in the world to complete a harvest – from sowing and fertilizing all the way through to the actual harvest of the crop – using robotic technology and autonomous vehicles.
The project, called Hands Free Hectare, was conducted on a farm in the village of Edgmond in the UK.
The team successfully harvested 4.5 metric tons of spring barley, just short of their goal to get 5 metric tons.
The team worked with Precision Decisions, a precision farming services company, which also partially funded the project.
They completed the task with basic, commercially-available agriculture machines and open-source software typically used to guide hobbyists’ drones. They equipped the farming machines with actuators, electronics and robot technology, which enabled them to control the machines without a human operator required to be present in the field.
The first step was making everything radio controlled, and they then progressed into pre-programming all the necessary actions to allow for an autopilot system. The machines followed GPS navigation to targeted points where they carried out the specific, pre-programmed tasks.
Drones were used to monitor the field’s health and take samples with specially-made grippers attached to the drone. The samples let the team know if the crop was ready for harvest or not.
Every step of the process – fertilization, sample collection, and the harvest – were all done by the autonomous vehicles, without any human intervention in the field.
"It's not about putting people out of jobs; instead changing the job they do. The tractor driver won't be physically in the tractor driving up and down a field. Instead, they will be a fleet manager and agricultural analyst, looking after a number of farming robots and meticulously monitoring the development of their crops,” says Kit Franklin, researcher at Harper Adams.
The team used smaller agriculture equipment than most farms currently do – noting that the use of smaller machines could improve soil and plant health.
Jonathan Gill, mechatronics researcher at Harper Adams University, explained in a release on the project that the variability of weather and precipitation causes farmers to adjust their spraying and harvest schedules. As a result, machines have increased in size over the years so that the work can be completed in a shorter time frame. But bigger machines mean weaker accuracy in distributing fertilizers and herbicides and degrade the overall health of the soil.
A fleet of smaller machines would overcome these challenges, according to Gill.
The team used a 25-year-old combine harvester machine, a small Japanese tractor designed for use in rice paddies, and a basic sprayer boom and seed drill.
The software proved successful in steering the tractor’s wheel, turning spray nozzles on and off, raising and lowering the drill, and navigating the combine.
“This project aimed to prove that there’s no technological reason why a field can’t be farmed without humans working the land directly now and we’ve done that. We set-out to identify the opportunities for farming and to prove that it’s possible to autonomously farm the land, and that’s been the great success of the project,” says Martin Abell, mechatronics researcher for the industry lead at Precision Decisions. “We achieved this on a low budget compared to other projects looking at creating autonomous farming vehicles. The whole project cost less than £200k, funded by Precision Decisions and Innovate UK. We used machinery that was readily available for farmers to buy; open source technology; and an autopilot from a drone for the navigation system.”
The first harvest was a success overall, but the team did experience some hurdles that can be addressed in future work. For example, the rows of crops weren’t perfectly straight, which led to some loss. Gill explained instances where the software would approach a large rock, and navigate around the object instead of going right through it.
The team intends to produce a “Hands Free Hectare” beer using the spring barley that was harvested during the project. They also plan to repeat the experiment with a winter crop.