Key Takeaways:
- Garford Farm Machinery introduces Robocrop AI, a multi-modal vision guidance system for mechanical hoeing and weed control.
- Robocrop AI leverages artificial intelligence, combining colour, infrared, and depth data for precise crop and weed identification.
- The system features advanced ‘green on green’ plant recognition, enabling higher weed control efficacy and resilience to light and weed pressures.
- Robocrop AI will be available in the 2025 season, initially for Garford’s InRow electric rotor weeders, with further expansions planned.
Introducing Robocrop AI: The Next Step in Precision Farming
Garford Farm Machinery unveiled the next generation of its Robocrop system. The new Robocrop AI vision guidance system was launched at the World Agri-Tech Innovation Summit in London, offering significant advancements in precision farming through artificial intelligence.
Developed by Garford’s team of mechatronic, software, and data engineers, Robocrop AI aims to deliver higher uptime, enhanced capacity, and reduced operational costs for growers. This new system brings improved performance in varying light conditions and weed pressure, making it a critical tool for modern agriculture.
Green on Green Plant Recognition
One of the most innovative features of Robocrop AI is its ‘green on green’ plant recognition, a significant upgrade from previous ‘green on brown’ systems that relied solely on colour data. Robocrop AI now integrates multiple data streams, including colour, infrared, and depth, to distinguish crop plants from weeds or non-crop plants with high precision.
“Green on green plant recognition is a step change from the green on brown system that uses colour only. This new system uses multiple different data streams from the camera and is able to identify crop plants and distinguish those from weeds or non-crop plants,” explained Jonathan Henry, managing director of Garford Farm Machinery.
Multi-Modal Vision Guidance for Enhanced Accuracy
A major differentiating factor of Robocrop AI is its multi-modal camera system, which uses both infrared and colour cameras along with depth imaging to accurately detect stem locations, even when multiple leaves overlap. The system combines data from these inputs using data fusion techniques to calculate vegetative index and deliver highly accurate crop detection.
“In practice, this means we can hoe closer to the plant to achieve higher weed control efficacy, ultimately bringing value to the user,” Mr. Henry added.
Adaptability to Light and Weed Conditions
Robocrop AI is designed to perform reliably in challenging field conditions, providing resilience to varying light levels and cloud cover, as well as intense weed pressure. The AI neural network, trained to recognize the crop plant and stem location using colour, leaf shape, and profile, ensures accurate operation even in difficult environmental conditions.
“Using multiple data streams means that, if for example, the operator is going out late and the weed pressure has intensified, you can still identify the crop plants and hoe with high precision,” Mr. Henry noted.
Key Features and Availability for the 2025 Season
Robocrop AI will be available for the 2025 season, initially integrated into Garford’s InRow electric rotor weeders. The first release will include single-section steering systems with single and double camera configurations. Future updates will expand the system’s capabilities, adding support for multiple sections.
The system comes with a 10.1-inch terminal featuring a Garford-developed interface, complete with live stream images and sideshift positioning to give operators better control and visibility during use.