November 4, 2025
A unique AI-powered agriculture project aimed at helping grain farmers manage black grass has been successfully completed.
Bosch and its partners have created a comprehensive and innovative technology package in which a high-tech camera mounted on the sprayer boom can detect black grass (Alopecrus myosilloides) Individual growth stages of cereal crops and each weed.
This unique precision technology ensures that the right amount of herbicide is spot applied to specific field areas, the researchers explained.
They also highlighted the positive outcomes brought about by this technology. First, herbicide use is optimized through spot spraying and is not applied evenly across the field, which saves farmers money.
Secondly, black grass will no longer be a problem and its infestation will decrease over time. Central to this project was the use of AI to identify where black grass grows and the stage of development of each weed.
“A truly collaborative project”
Various steps were required to complete the project. The original purpose was to see if black grass could be identified. Rothamsted Research in a laboratory setting helped Bosch cameras train their cameras to detect black grass by photographing it from a constant height over long periods of time as it grows.
Bosch then trained an AI algorithm to apply this knowledge to images from cameras driven in the field. The objective was to identify areas where black grass grows. Importantly, the camera does not ask the sprayer to target specific plants. Instead, ask the sprayer to target areas where you can see black grass.
As the project progressed, Chafer Machinery’s crop sprayer setup was adjusted to include an increased number of cameras and different boom heights on each side of the sprayer, for example. This continued development eventually resulted in a fully equipped boom with 28 cameras.
Peter Frankland, application engineer at Bosch, said: “This was a truly collaborative project. The farming part of the process was managed by BASF Digital Farming, which used its xarvio digital farming solutions platform to generate a customized map for use in the scarab sprayer to determine the dose and type of herbicide.”
“Rothermsted has given us an understanding of the types of cultivation that farmers use and some ways to solve black grass problems without using crop protection. We worked together to ensure the best possible results and tackled the various challenges that arose.”
“For example, in discussions with Rothamsted Research about the impact of evolving agricultural technology, we found that more farmers are adopting no-till or no-till farming methods, making weed control an even more important issue.”
Frankland added that unlike tillage, which buries blackgrass seeds and prevents them from germinating, these practices leave the seeds close to the surface. “There’s a lot of science behind this process,” he concluded.
Rothamsted Research now has an additional way to check your project’s test results. The researchers counted black grass occurrences using precisely localized quadrants. Measure each quadrant consistently to ensure black grass is reduced. One of the more challenging aspects of the project was training the AI.
Mohammad Qasem, AI expert at Bosch, explained how: “Initially, we didn’t have anything to benchmark this against. As the project progressed, the model became more mature and was able to detect invisible black grass on images with a high level of accuracy.”
“In total, we scanned around 5,000 images covering different seasons and where black grass grows among different crops, including barley as well as wheat. Each time the sprayer passed over a field, a large number of images were generated, so it took several days to download them all.
“We then used coding to clean up the images, another code to convert the images to the appropriate file format, and the final code to train the AI algorithm. This was a great project and we were able to deliver with high accuracy, which is a key measure.”
“Enabling key innovations for the UK market”
Bharath Jayakumar, director of global key accounts at Bosch UK, explained that the project began in 2021.
He said: “First we looked for a suitable consortium partner and together we applied to Innovate UK. We secured funding from DEFRA and Innovate UK to develop a local use case for the ONE SMART SPRAY weed management system developed by Bosch and BASF Digital Farming.”
“As a small team in the UK, we have been working with Rothamsted Research, Chafer Machinery and BASF Digital Farming to test technology to scan and identify black grass in wheat.
“Peter’s expertise in machine integration was critical here. As the project’s AI expert, Muhammad analyzed, cleaned and annotated the scanned images and trained machine learning models to identify weeds between crops.”
Scientists have reached an accuracy of the model that allows control and reduction of black grass over time. “This project represents an opportunity in the UK to use government funding to deliver key innovations for the UK market. It also supports the development of our workforce and this project is in line with the objectives of the Bosch Mobility UK strategy to grow in markets and sectors outside of our core business areas,” Jayakumar concluded.
Bosch was the leading partner in the research and development project consortium, consisting of renowned experts from BASF/xarvio Digital Farming Solutions, Chafer Machinery and Rothamsted Research.
Together they secured a grant of £1,452,614 from DEFRA and the UK innovation agency Innovate UK for the three-year project under the Agricultural Innovation Program – Small R&D Partnership Project.
Funding for the Blackgrass project was awarded as part of DEFRA’s Agricultural Innovation Program and UKRI’s Food Production Transformation Challenge.
Read more: A fresh approach to the evolving blackgrass landscape
Read more: How to manage black grass when returning land to cropping from SFI options
Read more arable news.
