Project description
Tech-driven farming for a sustainable future
The agricultural sector is grappling with the challenge of increasing production while minimising the adverse effects of this on society, climate and biodiversity. There is a growing need for climate-friendly sustainable agriculture. Automated milking systems (AMS) have proven beneficial by reducing manual labour, cutting costs and enhancing milk quality. The EU-funded dAIry 4.0 will revolutionise AMS by integrating AI, data and robotics solutions. By leveraging multimodal learning techniques, self-supervised data augmentation and novel explainable AI as well as actively involving farmers in the process, the project seeks to optimise AMS production while minimising its impact on the environment and animal welfare. It promises to be a game changer for both the farming sector and the food industry.
Objective
The agricultural sector has a big challenge: producing more with fewer raw materials and less adverse effects on society, production animals, climate and biodiversity. Optimal use of resource is even more important now, due to the imminent food crisis. Climate-friendly sustainable agriculture, with care for natural resources, is essential for our food production and quality of life, today and for future generations.
Automated Milking Systems (AMS) were developed in the late 20th century under the perspective of reducing manual labour & costs and improving quality of life for the farmers. Not only have these machines improved in harvesting milk efficiently, but they also have the added ability to collect a greater amount of data about production, milk composition, cows health and behaviour. This could allow producers to make more informed management decisions, while in parallel reducing emissions and increasing animal welfare.
Nevertheless, currently available AMS have important limitations in terms of optimising their operation.
dAIry 4.0 addresses these challenges, integrating and optimising AI, data and robotics solutions to demonstrate how this combination will optimise AMS production aspects and minimise adverse effects on society, climate and biodiversity. The approach will be demonstrated through real-world use cases of interest both for the farming sector and the food industry. In terms of AI tools to be used, the project will focus on the following novelties:
- Developing multimodal learning techniques to efficiently utilize multiple types of information for animal health & overall animal status monitoring
- Developing self-supervised and novel data augmentation techniques to reduce the amount of labelled training data needed
- Exploring novel explainable AI techniques to increase transparency of the system and eventually facilitate acceptance by the users
- Including the farmer in the loop to build the cognitive abilities for the system
Fields of science
- agricultural sciencesagriculture, forestry, and fisheriesagriculturesustainable agriculture
- agricultural sciencesanimal and dairy sciencedairy
- natural sciencesbiological sciencesecologyecosystems
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringrobotics
Programme(s)
Funding Scheme
HORIZON-IA - HORIZON Innovation ActionsCoordinator
1060 Lefkosia
Cyprus
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.