Contact us

M.A.R.E.

Machine learning TOapplied to the Rresearch on Emarine ecosystems via AUV

Machine Learning Applied to Marine Ecosystem Research via AUVs
Go to the project
Research focus
The MARE (Machine learning Applied to Marine Ecosystem Research using AUVs) project aims to fill current gaps in the observation of biological, ecological, chemical, and geological parameters in the marine environment. The strategic objective is to develop a new integrated system for long-term environmental monitoring using autonomous underwater vehicles (AUVs) and minimizing human intervention at sea.
Solution found
The MARE solution is a multi-modular system integrating several key components: an AUV (Autonomous Underwater Vehicle), a Docking Station and a Lander (now a single semi-permanent underwater structure), a Hub Buoy on the surface, and a Ground Control Station (GCS) on land. The AUV will collect environmental data and transfer it to the Lander, to which it will connect via automatic docking assisted by Artificial Intelligence. The inductive charging system and data transfer between the AUV and Lander are innovative for this application. The Buoy, connected via cable to the Lander, will ensure power supply via solar panels and data exchange with the GCS, where the data will be collected and sent to a decision support system based on a Data Analytics Engine developed by Relatech for statistical analysis and machine learning algorithms, equipped with data visualization capabilities via intuitive dashboards.
Advanced technological solution
MARE uses cutting-edge technologies:
Artificial Intelligence (AI) Algorithms

Based on libraries like OpenCV, they recognize geometric shapes (such as the LED strips on the Docking Station) to guide the AUV towards autonomous and unsupervised docking. The AUV is controlled by a Camera Processing Unit (CMU) and a Vital Unit (VU), which interpret visual data to coordinate maneuvers.

Advanced Lander/Docking Station Design

The structure is a "canopy" made of perforated stainless steel sheets, inclined to facilitate the entry of the AUV by exploiting its natural buoyancy. Custom components are made using additive manufacturing (3D printing) in materials such as PLA, to optimize the design, reduce waste, and ensure lightweight and environmentally sustainable construction.

Data communication infrastructure

LoRaWAN technology was selected for wireless transmission between the buoy and the ground station due to its ability to cover long distances with minimal power consumption, ideal for marine IoT applications. The data communications infrastructure for the MARE project was implemented by Relatech.

Data Analytics Engine (DAE)

Developed by Relatech, a data analytics engine for statistical analysis and machine learning algorithms, equipped with the ability to visualize the collected data through intuitive dashboards.

Financing
The MARE project (Machine learning applied to research on marine ecosystems using AUVs) is funded under the "cascade calls" of the RAISE (Robotics and AI for Socio-economic Empowerment) project. This initiative is part of the National Recovery and Resilience Plan (PNRR) – Mission 4, Component 2, "From research to business", Investment 1.5, and is supported by the European Union through the NextGenerationEU program. The MARE project specifically addresses the challenges of Spoke 3, which focuses on "Protection and care of the environment.".

Partnership
The MARE project is the result of the synergistic collaboration of five partners with multidisciplinary expertise, including Relatech. Other project partners include Edgelab (Lead), Automation srl, Superfici srl, and BioAge srl. The partnership also includes scientific collaboration with research organizations such as the National Institute of Geophysics and Volcanology (INGV) and the University of Padua, which provide infrastructure and specialized expertise.
Do you have an Innovation project?
Share it with us