REMARO – Reliable AI for Marine Robotics

23 September 2020

EIVA is part of the new European Horizon 2020 research project REMARO (Reliable AI for Marine Robotics). The project focuses on underwater robotics and some of the activities currently encompassed by EIVA today: deep learning and VSLAM (Visual Simultaneous Localisation and Mapping), in relation to subsea operations.

'We have witnessed immense technological change over recent decades. Always seeking to embrace the possibilities of cutting-edge applied science, we have a strong track record of investing wherever necessary in pursuit of refining our solutions, developing new ones, and pushing the boundaries of maritime technologies. Most recently, EIVA has commenced the use of machine learning in its software products to maintain the competitive edge and with the purpose of making subsea and offshore tasks better, safer, faster and cheaper. EIVA introduced deep learning-based products into the market (NaviSuite Deep Learning) in 2017 for automatic anomaly / damage detection during subsea inspections,’ said Software Manager Ole Kristensen, EIVA.

A 3D mesh generated from survey data with EIVA’s VSLAM

What is the purpose of the REMARO project?

Robots can operate in places where human presence can be both costly and dangerous, such as food production at the bottom of the ocean, subsea inspections and maintenance of vessels and structures. However, robots operating under the water have to tackle many challenges: currents, vegetation and marine life, reduced visibility, restricted communication – and they often have to manage all of this on their own when beneath the waves.

The development of reliable AI systems is key to unmanned subsea activities. This is the challenge addressed by the REMARO project. 15 PhD students at different universities across Europe will collaborate to design brains for robots that will be self-learning and autonomous, and at the same time operate safely both for themselves and for the environment. Experts in AI and reliability analysis will cooperate with companies developing underwater robots. Furthermore, REMARO addresses reliability and how well AI implementations function and will be an important component in pushing forward the underwater robotics industry in Europe.

'A major challenge for European industry today is the reliability of AI-based systems, but we lack an education on this topic in Europe today. The REMARO project will train 15 experts on reliable AI through an innovative training program at the doctoral level, in which universities and industry from European countries with significant marine sectors will collaborate closely,’ said Professor and Project Coordinator Andrzej Wasowski, IT-University Copenhagen and REMARO project.

The application and benefits of deep learning in NaviSuite software

EIVA’s contribution to the future of underwater robotics

EIVA will assist training the PhD students in the application and deployment of formal methods for robotic systems over the 48 months of the project starting from Spring 2021. Aarhus University are among the 15 partners who will also establish PhD students during this period.

We at EIVA are enthusiastic to be a part of the REMARO project, yet it is still in early stages. There will be more info to come in the near future with details about our participation. In the meantime, sign up to the EIVA newsletter for ‘automated’ updates of the project’s development.

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 956200.

Pipeline inspections with NaviSuite Deep Learning

Contact EIVA

REMARO – Reliable AI for Marine Robotics

What is the purpose of the REMARO project?

EIVA’s contribution to the future of underwater robotics

Related news

A new type of hydrographic survey solution developed for UK MoD has passed FAT

Subsea 3D digitisation from camera: VSLAM vs photogrammetry technologies

A new representative sailing out NaviSuite software for offshore survey and positioning