- Background
- Use case ecosystem
- Key facts
- Uniqueness
- Architecture
- Beyond the State of the Art
- Open source/proprietary
- Target users
- Benefits
- Potential societal or environmental impacts
Background
Beck et al. Services S.R.L. (Beck) is a digital service provider that supports companies on their journey to digitisation, helping them get a grip on data analysis and setting up cloud-based or hybrid solutions.
Gregoire is a French company specialised in harvesting machines and sprayers manufacturing, based in Cognac. The company produces a wide range of vineyard maintenance tools and holds the leading position in the viticulture sector worldwide.
Use case ecosystem
This use case combines the efforts of:
- Beck et al.: The company develops competitive edge over market players and remove cloud-only limitations.
- Gregoire: The company provides the sensors, install the materials and test potential solutions for the use case.
- Universidad Politecnica de Valencia: The university develops the cloud framework for efficient provisioning and configuration of AI resources.
- Cloud & Heat: The company provides the infrastructure and computing resources (including GPUs) for the test and validation of the use case.
- Politecnico di Milano: The university provides its expertise in designing the sensor suite to be installed on the agricultural machines and in defining and optimising the AI models.
Key facts
Industry 4.0 exploits digital technologies (internet of things, big data, artificial intelligence) to optimise its processes. This sector has been growing rapidly over the last few years by using more accurate analysis and knowledge exchange enabled by these technologies.
The agricultural sector has also started to embrace this digital revolution. Two key innovations are linked to the adoption of GPS technologies and sensors that enable precision agriculture (PA). According to Euractiv, almost 80% of new farming equipment has embedded PA components, with 4500 manufacturers, 450 machine types, 135,000 employees using digital farm systems worldwide. This digitisation will unleash €26 billion from Agriculture 4.0 in coming years.
The agricultural sector is continuously expanding its research on digitalisation to further improve processes and production systems with artificial intelligence playing a key enabling role in driving change.
Uniqueness
The AI-SPRINT use case on Farming 4.0 will increase know-how and help the partners directly involved as end-users to establish themselves as market leader in vineyard technology.
The Farming 4.0 use case will use to optimise phytosanitary treatments. It will use edge and intelligent sensors to optimise phytosanitary treatments, helping to preserve and protect the environment with the support of AI-SPRINT technology experts.
In particular, this use case will:
- Develop novel models for phytosanitary product optimisation, which have never been used before in farming.
- Design AI models to collect data from the sensors deployed on grape harvesting machines.
- Develop and deploy solutions on a heterogeneous IT environment, benefitting also from support to multi-cloud and open-source standards.
New AI-based process will ultimately help reduce the quantity of phytosanitary products needed for treating the vineyards while ensuring effectiveness.
Architecture
In terms of design and runtime, the underlying architecture is very similar to the AI-SPRINT reference architecture but without Federated Learning and encryption of the containers.
The deployable infrastructure consists of a Kubernetes cluster for training the AI-model and edge devices. The device will rely on OSCAR for FaaS functionality. At this early stage it is assumed that the edge device will run several containers orchestrated by Kubernetes and deployed using the Infrastructure Management.
The edge containers interacting with sensors or with tractor hardware (spraying system) will use containers running ROS. MinIO will be used to provide local storage services and synchronisation with the cloud.
Advances beyond the State of the Art
Adaptive treatment in vineyards hardly exists in today’s productive environments. The quantity of chemicals used is currently adapted based on the speed of tractors. Specific aspects such as foliage volume, canopy shape etc. are not taken into account., although they have a known impact on the amount of substances used.
Being able to run AI components on the cloud in a managed but cloud-provider agnostic manner is a really important innovation. AI-SPRINT aìcompoents and architecture enable the use of services provided by large cloud-providers while reducing the risks of lock-in.
Open source/proprietary
Tools used are open source.Sensors and connectivity between Edge device and tractors are proprietary.
Who is this for?
- Farmers
- Association of wine producers
- Manufacturers of tractors and agricultural equipment
Benefits
Precision farming in viticulture allows optimised usage of phytosanitary products which brings financial benefits and - more importantly - has environmental benefits such as pollution reduction, increased good insects for the farm ecosystem , prevention of new species resistant to treatment, fewer chemical products, fewer products harmful to biodiversity.
Potential societal or environmental impacts
The Farming 4.0 use case will lower pollution levels by at least 35% by reducing the amount of chemical products used during agricultural processes. The use case will contribute also to biodiversity preservation.
The Farming 4.0 use case will pave the way for a resource-efficient and ultimately regenerative circular economy, protecting jobs and increasing competitiveness for a sustainable future. The use case will thereby help drive the implementation of Europe’s Farm to Fork strategy and ensure sustainable consumption and production patterns in line with the United Nation's Sustainable Development Goal 12.