Catalogue of Services

Testing and evaluation of mobility algorithms with aerial robot

National Institute for Research in Digital Science and Technology  - INRIA

Location

France

At user's premises

Arable farming

Food processing

Greenhouse

Horticulture

Tree Crops

Viticulture

The sophia infrastructure will offer the possibility to test and evaluate the mobility algorithms embedded on a ground robot. M

obility Algorithms concerns the classical robotics functionalities of Mapping, Localization, SLAM, and Navigation.

The aerial robot is equipped with an array of sensors, including Camera, LiDAR, IMU, and RTK-GPS (for the ground truth evaluation). The service consists of three main steps:

1. First of all, the algorithm is evaluated using representative datasets. 
2. After that, the algorithm is integrated in a ROS2 architecture and evaluated with the local agrifoodTEF test infrastructure (different areas are possible). The performance of different attributes of the algorithm is evaluated with quantitative and qualitative metrics. A possibility of benchmarking will be proposed as a complementary option in order to position the performance of the proposed algorithm regarding the current state of the art. 
3. Finally, the last step will be to perform the field testing in real condition and in a particular end-user or customer site using the mobile living lab (it consists of a mobile laboratory going to the field connected with the real robot for monitoring and evaluation purposes).

Testing and evaluation of mobility algorithms with aerial robots

National Institute for Research in Digital Science and Technology  - INRIA

Location

At user's premises

France

Arable farming

Food processing

Greenhouse

Horticulture

Tree Crops

Viticulture

The SOPHIA infrastructure will offer the possibility to test and evaluate the mobility algorithms embedded on an aerial robot. Mobility algorithms concern the classical robotics functionalities of mapping, localisation, SLAM, and navigation. The aerial robot is equipped with an array of sensors, including a camera, LiDAR, IMU, and RTK-GPS (for ground truth evaluation). The service consists of three main steps. To begin with, the algorithm is evaluated using representative datasets. After that, the algorithm is integrated into a ROS2 architecture and evaluated with the local agrifoodTEF test infrastructure (various areas are possible). The performance of different attributes of the algorithm is evaluated using quantitative and qualitative metrics. Benchmarking could be proposed as a complementary option to position the performance of the proposed algorithm in relation to the current state of the art. The final step involves field testing under real conditions at a specific end-user or customer site using the mobile living lab (which consists of a mobile laboratory deployed in the field and connected to the real robot for monitoring and evaluation purposes).

Testing and evaluation of mobility algorithms with ground robot

National Institute for Research in Digital Science and Technology  - INRIA

Location

At user's premises

France

Arable farming

Food processing

Greenhouse

Horticulture

Tree Crops

Viticulture

The sophia infrastructure offers the possibility to test and evaluate the mobility algorithms embedded on a ground robot. 

Mobility Algorithms concerns the classical robotics functionalities of Mapping, Localization, SLAM, and Navigation. The ground robot is equipped with an array of sensors, including Camera, LiDAR, IMU, and RTK-GPS (for the ground truth evaluation). 

The service consists of three main steps:

1. First of all, the algorithm is evaluated using representative datasets. 
2. After that, the algorithm is integrated in a ROS2 architecture and evaluated with the local agrifoodTEF test infrastructure (different areas are possible). The performance of different attributes of the algorithm is evaluated with quantitative and qualitative metrics. A possibility of benchmarking will be proposed as a complementary option in order to position the performance of the proposed algorithm regarding the current state of the art. 
3. Finally, the last step will be to perform the field testing in real condition and in a particular end-user or customer site using the mobile living lab (it consists of a mobile laboratory going to the field connected with the real robot for monitoring and evaluation purposes). 

Testing and evaluation of variable rate distribution equipment 

University of Cordoba

Location

Spain

Arable farming

This service focuses on the evaluation and testing of variable input distribution technologies, such as seeders and fertiliser spreaders, at the Rabanales Experimental Farm. Through field tests, the service measures variability across plots and monitors crop yield with yield monitors. Satellite data complements these assessments, enabling the creation of zoning and crop prescription maps. These resources help optimise the operation of precision agriculture equipment and improve input efficiency under various ISO standards.

Testing and Validating Crop Variability with Remote Sensing and Predictive Models

University of Cordoba

Location

Spain

Arable farming

Tree Crops

Viticulture

This service measures spatial variability in crops and plots through nearby and remote sensors, including electromagnetic probes, yield monitors, drones, and satellites. The data collected allows for the generation of zoning information and the validation of predictive models and digital agriculture systems, such as variable input distribution tools. By integrating environmental data, the service provides essential insights to improve decision support systems and optimize resource allocation in precision agriculture.

Testing and Validation of AI-based Agri-tech Technologies with Drones

Danish Technological Institute

Location

Denmark

Arable farming

Greenhouse

Horticulture

Livestock farming

Tree Crops

The service focuses on testing and validating AI-based agri-tech technologies using drones. Drones equipped with advanced RGB and spectral cameras provide detailed area overviews and precise measurements of plant and tree growth and stress levels. This can include biomass determination, growth model development, yield potential assessment, plant cover evaluation, individual plant counting, damage assessments, and disease stress symptom identification. Our expertise lies in leveraging high-precision multi- and hyperspectral cameras in field trials to obtain "ground truth" data, essential for model validation. We transform drone data into actionable insights through AI analysis, agronomic expertise, web programming, and mathematical modelling.

Testing and validation of an AI-based NDVI detection algorithm in a pivot irrigation system

University of Cordoba

Location

Spain

Arable farming

Horticulture

Tree Crops

The evaluation process will encompass a comprehensive analysis of the entire parcel, concentrating on areas where the irrigation system distributes water. Zones with an NDVI below a specific threshold, identified via satellite imagery, will be highlighted. To obtain more granular data, drone flights equipped with LiDAR technology will be conducted, revealing areas of vegetation stress that may not be adequately detected by satellite imagery. The results from both systems will then be cross-referenced to assess irrigation efficiency accurately

Testing and validation of precision livestock farming solutions for grazing animals

University of Cordoba

Location

Spain

Livestock farming

This service supports the testing and validation of precision livestock management tools designed for grazing animals such as horses, cattle, and sheep. Conducted on collaborating research farms, the service evaluates hardware, firmware, and software components to confirm they meet manufacturer specifications. It includes direct observation of animal behaviour and performance, along with comparative analyses against similar tools. Clients receive actionable recommendations to address any detected limitations, enhancing the tool’s functionality and market viability in precision livestock farming.

Testing and Validation of Soil Sensors

University of Cordoba

Location

Spain

Arable farming

Horticulture

Tree Crops

The validation service would evaluate soil sensor precision by comparing real-time sensor readings with soil levels from lab-analysed samples collected nearby. Soil samples would be taken under various conditions to ensure a robust comparison, enabling a clear assessment of the sensors’ accuracy across different environments.

Testing and validation of the performance of predictive models based on nondestructive spectral sensor data (NIR and hyperspectral data).

University of Cordoba

Location

Spain

Food processing

Horticulture

Tree Crops

Viticulture

This service validates the performance of multivariate predictive models that utilise nondestructive spectral sensor data, such as NIR and hyperspectral imaging. Through statistical and chemometric analysis, the service assesses model accuracy and provides recommendations for optimization. Clients benefit from access to a comprehensive software suite dedicated to chemometrics and statistical studies, as well as options for database transfer to adapt models across devices. The service also includes access to a large agri-food sample bank for model evaluation and refinement.

Testing and validation using pilot and lab-scale test beds for agricultural AI solutions

Danish Technological Institute

Location

Denmark

Arable farming

Greenhouse

Horticulture

Tree Crops

Viticulture

This service involves deploying AI models in authentic agricultural environments to validate their predictions against real-world data. The process includes field deployment where AI models are integrated with drones or other data collection systems. Ground truth data is collected using manual measurements or industry-standard equipment, serving as a benchmark for accuracy evaluation. Models are compared against this data to assess performance and robustness across various conditions. If required, a feedback loop can be established to refine models based on discrepancies found, ensuring continuous improvement and reliable predictions for agricultural decision-making.

Testing and validation using pilot and lab scale test beds for agricultural AI solutions

Danish Technological Institute

Location

Denmark

Arable farming

Greenhouse

Horticulture

Livestock farming

Our pilot and lab scale testbeds provide a controlled environment for customers to execute physical tests of their agricultural technology, machines, or algorithms. These testbeds can be configured to simulate real-world agricultural environments, allowing for precise control and monitoring of variables. This setup enables AI models to be trained and refined using accurate, real-time data. The testbeds incorporate automation to mimic various agricultural tasks, offering a comprehensive platform for testing and optimising AI-driven agricultural solutions. This service bridges the gap between theoretical development and practical application in the agricultural sector.