Catalogue of Services

Collection of datasets of images and sensor data using UAV

GRADIANT

Location

At user's premises

Spain

Arable farming

Horticulture

Livestock farming

Tree Crops

Viticulture

The service offers automated drone deployments to efficiently gather images and sensor data, creating comprehensive multi-spectral datasets. The drones operate based on agreed requirements, such as the study area, data type, and measurement granularity, ensuring all necessary information is captured with precision. This makes data collection easy and reduces human interference, hence reducing the risk of error and making the obtained information more valid. After flying, the data is processed and tied up to enable further analyses. The service can deliver fast, comprehensive, and well-treated datasets according to each customer's needs using methods that can be further used in artificial intelligence models. With such information, customers will be better informed during the optimisation of their operations by using advanced technologies.

Collection of datasets of images and sensor data using ground robots

GRADIANT

Location

At user's premises

Spain

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

The possibility to autonomously gather images and generic sensor data means a quantum leap in the creation of multispectral datasets in an automated and efficient way. A ground robot deployment occurs based on agreed requirements such as study area, measurement granularity, and data type so that it automatically captures all the information needed to generate a rich dataset. This not only makes data collection easy but also reduces human interference, hence reducing the chances of error, thus making the information obtained more valid. After the operation, the data is processed and tidied up to enable further analysis. In that respect, it can deliver faster and more accurate results for the customer, well-treated datasets according to each customer's needs, and at standards that can be further used in artificial intelligence models. With such information, customers will be better informed during the optimisation of their operations by using advanced technologies.

Collection of test data during physical testing

Politecnico di Milano (POLIMI)

Università degli Studi di Milano (UMIL)

Location

Italy

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

Through this service, we offer dedicated technical systems (hardware and software) and expertise for the collection of relevant data during testing.

The scope of data collected can range from data generated by the AgrifoodTEF technical infrastructure (e.g., through the installation of IoT sensors) to the collection of data generated by the system under test to the generation of ground truth for evaluating the system. If necessary, service S00111 may be leveraged to take care of the interconnection between AgrifoodTEF data collection systems and the customer’s systems.

In addition to the raw collected data, we will also provide customers with complete documentation describing the logged features and conditions of the testing environment at the time of testing, as well as any parameter values, variation ranges and specifics required for reproducibility purposes.

Where necessary, data collection and documentation activities can require the involvement of field experts (e.g., engineers, agronomists).

Note: annotation and labelling of collected data, if needed, can be provided via service S00115.

Collection of test data during digital testing

Politecnico di Milano (POLIMI)

Università degli Studi di Milano (UMIL)

Location

Italy

Remote

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

One of the key activities during digital testing is the collection of data concerning the progression and the final outcome of the tests. Such data enable the evaluation of system performance by the customer or – if needed – by agrifoodTEF (via Service S00184). This service manages the collection of data relevant to performance evaluation produced during the tests by both the system under test and the computational environment where the tests take place.

Examples of collected data comprise information produced within a virtual environment to simulate sensor data collection in a physical environment; statistics about AI model performance in the test and deployment phase (e.g., occupied memory, number of trainable parameters, training/optimisation loss, etc.); specific labels and annotations to use as ground truth for evaluating the system; and system output when subjected to a range of test conditions.  

The minimum set of data to be collected is defined by the evaluation metrics that the user chose (either on their own or with agrifoodTEF support, via Service S00178) to process them; generally, a larger set of data wrt the minimum is selected by agrifoodTEF together with the customer to provide a richer view of the system’s performance and to enable the application of other metrics in the future, if needed.

As an output of the service, in addition to the raw data, we also provide the customer with documentation describing logged features and conditions of the testing environment at the time of testing, as well as any parameter values, variation ranges and specifics required for reproducibility purposes.

Compliance assessment for robotic machines

RISE - Research Institutes of Sweden

Location

At user's premises

Sweden

Arable farming

Food processing

Horticulture

Livestock farming

Tree Crops

Ensure your company develops robotic machinery that complies with all relevant regulations, standards, and directives with the support of expert guidance. Beyond meeting existing requirements, expert assistance is provided in areas where the rules are unclear or undefined. For situations with ambiguous or missing guidelines, practical solutions are offered based on extensive experience in functional safety.

Autonomous robotic machinery often relies on complex sensor technologies, such as radar, LIDAR, and cameras, to detect nearby individuals. Demonstrating how these advanced sensors can be integrated into safety functions is challenging, as it involves considering not only traditional functional safety aspects but also emerging factors like AI and Safety of the Intended Functionality (SOTIF).

Support includes performing comprehensive risk assessments, identifying relevant safety functions, and establishing high-level safety requirements for each. Depending on the project's focus, further guidance can be provided to ensure the selected safety concept aligns with applicable functional safety standards.

Conformity assessment and compliance tests

Lukasiewicz Poznanski Instytut Technologiczny (L-PIT)

Location

At user's premises

Poland

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

As part of this service, we test and measure, among other things, mechanical, physical, acoustic, radio, electromagnetic compatibility and electrical parameters for the purpose of assessing the overall user safety of agricultural, horticultural, forestry and food machinery, equipment and components. Verification is based on the requirements of standards and directives declared by the manufacturer. We carry out an initial assessment based on the design documentation provided or/and on measurements taken on a prototype based on harmonised and non-harmonised standards with relevant EU and sectoral legislation. This is to ensure that for instance essential requirements coming from for instance New Legislative Framework directives and other EU law are fulfilled to better protect both consumers and professionals from unsafe products to be placed on the European internal market. One of the aims is to help manufacturers in legal placing agrifood products on EU single market and in CE-marking process. The results obtained can be used in the further process of product labelling, declaration of conformity to affix the CE mark to the device within the scope of EMC, LVD, RED, MD, MR on other NLF directives and regulations.

Container image generator that facilitates efficient software testing and deployment

GRADIANT

Location

Remote

Spain

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

Container images have become the de facto standard for software delivery in a variety of industries, including agrifood. Packaging an application and all its dependencies in a container image ensures consistency across different deployment environments, allowing these applications to be tested more quickly and efficiently. Once the container image is built, it can be run by different container management tools, for example, Docker or Podman.

However, creating secure and optimised container images is often complex. This service simplifies the process by automatically generating container images from the application source code. It detects dependencies and environmental requirements, removing the need for the customer to perform and/or to possess technical expertise in container technologies.
The process is fully automated: the user provides the source code reference with secure access, and the service handles the rest. This eliminates the need to manage image creation files and reduces potential human error during the image creation phase, allowing developers to focus on innovation and functionality.

Crop Dataset Generation

Instituut voor Landbouw-, Visserij- en Voedingsonderzoek (ILVO)

Location

At user's premises

Belgium

Remote

Arable farming

Greenhouse

Horticulture

Tree Crops

Viticulture

In the Crop Dataset Generation service, we collect large-scale datasets on a frequent basis using drones or autonomous robots. AI and data-driven solutions require extensive data, shifting the focus from limited feature engineering to acquiring large, balanced datasets of images and labels for training convolutional deep-learning networks. We offer flexible data collection options tailored to customer needs. We can use our systems and fields to gather data on crops under various conditions, such as different lighting, angles, growth stages, and soil types. Alternatively, we can use customer-provided devices to collect data for testing and validation, such as evaluating sensor performance in detecting heat in crops or testing robot navigation in crop fields.

Curation and provision of annotated datasets

Lukasiewicz Poznanski Instytut Technologiczny (L-PIT)

Poznan Supercomputing and Networking Center (PSNC)

Location

Poland

Remote

Arable farming

Food processing

Greenhouse

Horticulture

Livestock farming

Tree Crops

Viticulture

The service provides access to annotated data collected from devices tested as part of other project services. It includes both data annotation services (manual and automatic) and the provision of annotated databases or data packages. In addition, we offer disc space for temporary archiving of the collected data along with its annotation in cases where this is not possible at the client's site. As a result of the service, procedures for both manual and automatic annotation will be identified that are appropriate to the type of measurement data and the way it is collected and assembled. The methods of labelling and assigning specific data classes collected by the agriFoodTEF technical infrastructure will be proposed. Possibilities for documenting and storing data in datasets will be indicated. In addition, procedures for extracting selected features from stored research data can be offered.

Data analysis and quality evaluation for agricultural equipment and AI algorithms

University of Cordoba (UCO)

Location

Remote

Spain

Arable farming

Horticulture

Tree Crops

Viticulture

This service provides independent evaluation of agricultural datasets to determine their suitability for use in the testing, development, or validation of AI- and robotics-based systems. The focus is on verifying the quality, structure, and statistical consistency of the data to ensure it meets the requirements for use in intelligent technologies operating in agricultural environments. Our evaluation process includes the application of descriptive statistical techniques to assess data completeness, identify anomalies, and quantify variability across key parameters such as crop yields, irrigation records, and fertilisation schedules. We assess measures of central tendency, dispersion, and distribution to evaluate the stability and reliability of the datasets. This helps identify issues like missing values, outliers, or inconsistencies that could compromise the performance or fairness of automated systems trained or tested on this data. By systematically evaluating data integrity and structure, we help researchers, developers, and integrators ensure their AI algorithms or robotic platforms are tested with datasets that reflect real-world conditions. This contributes to more effective experimentation, better system generalisation, and ultimately, more trustworthy agricultural technologies.

Data Analysis for AI Performance Evaluation/Assessment

Institut Francais de la Vigne et du Vin (IFV)

Location

France

Viticulture

When you’re developing an AI-based solution, you need to assess its accuracy and its reliability. Whether it is for detection or prevision purposes, we can compare the outcomes of AI with ground truth and measure the accuracy of the detection or prevision of an AI-based model. If the model needs improvements, we can provide new datasets; see S00102.