Disruptive innovation consultancy firm Cambridge Consultants has announced a new programme integrating AI in smart farming projects.
The ‘AI at the edge’ initiative, which aims to implement low-cost machine learning technology solutions to conduct complex tasks, has debuted with the launch of precision crop sprayer Fafaza as a demonstration of potential applications for AI in smart farming. The Fafaza system uses granular data analysis in real time to single out weeds in large fields and deploy targeted doses of herbicide, from the back of a moving tractor. It operates using standard off-the-shelf components, with an AI computing platform costing less than $100 (€88.68) and a camera capable of capturing images at 20 frames per second.
Projects utilising AI in smart farming have previously been able to recognise and identify individual plants and weeds in fields; however, this has typically been successful using specialist, high performance platforms operating on a delay, so the next step in development – as exemplified by Fafaza – has been to create a resilient, cost-effective AI system capable of performing weed identification and neutralisation in real time and in motion. Cambridge Consultants expects that this new system will act as a demonstration use case for ‘AI at the edge’, where relatively uncomplicated devices are deployed to perform complex tasks using AI without the need for reliance on cloud computing and storage.
Niall Mottram, Head of Industrial and Energy at Cambridge Consultants, said: “I’m really excited by the innovations that moving AI processing to the edge will bring in precision agriculture. Connectivity is generally poor in rural environments and the ability to run complex machine learning based applications on a tractor implement opens up many new opportunities to collect, process and act on data in real time. Fafaza demonstrates how AI at the edge approaches can become practical, cost effective and deployable today.”