Blackgrass is one of the most economically disastrous weeds in western Europe. Control of blackgrass has relied on the intensive use of herbicides. However, 80% of blackgrass-infested fields surveyed across England were resistant to all three major herbicide groups. Severe blackgrass infestations reduce yields by up to 12% (c. £300 per/ha) costing UK farmers £300M-£500M and lost yield of c. 860,000T. Blackgrass is creating significant jeopardy for wheat farmers, with serious negative environmental impacts.

This project lays the foundation for next-generation robotic vehicles powered by renewable energy and tooled to control blackgrass. It drives productivity by increasing yield through weed eradication. It contributes to Net Zero Agriculture by reducing input waste (N/pesticides) and removing fossil fuels from key farming operations.

In 2 previous IUK feasibility studies (IUK\_105137; IUK\_78600), ARWAC and University of Lincoln (UoL) developed and patented 2 new robotic weeding vehicles; a small 3KW platform (V4) and a larger 32KW vehicle (V5). Both vehicles are autonomous steerage hoes powered by novel machine learning technology for weed recognition, row finding and vehicle path planning. Drones are used to develop weed maps of crops; this mapping data optimises autonomous route planning, whilst cameras on the robots facilitate high-precision hoeing between rows.

The objectives of this project are to push the technology from laboratory stage to full testing in multiple farm environments, supported by our network of c20 arable farms, where farmer input that has been secured to support the development and user design input required to demonstration in an operational environments. On-farm demonstration and co-creation with the our network of involved farmers (20 farms; 40,000 acres of cereals) across multiple fields provides compelling proof-of-concept, value-chain analysis and forward-looking design improvements. This enables ARWAC to build trust to secure necessary onward investment that scales the technology and generates revenue. Subject to farm trial data, our plan is a hoeing-as-a-service model, where ARWAC provide "hives" of robots to farmer groups. The initial use case is blackgrass hoeing, but the platforms and autonomy software can be adapted for multiple secondary markets.