On the 1st of October 2023 the project AUTOASSESS – “Autonomous aerial inspection of GNSS-denied and confined critical Infrastructures” has officially started. AUTOASSESS is a project co-funded by the European Union, that aims to remove human surveyors from dangerous enclosed areas such as ballast tanks or cargo holds.
To achieve this objective, it will employ an autonomous robotic system that exceeds human capabilities and obtain an accurate, repeatable and quick vessel inspection.
Project ambition:
To ensure the structural safety of ships, oil and gas and other marine infrastructures, they must be surveyed to detect corrosion, cracks or deformation. Currently, this task is done by human surveyors who must climb into confined areas such as ballast tanks, or cargo holds which represent extremely dangerous GNSS-denied environments. The inspection is a physically demanding task, done in tight enclosed spaces, difficult to access, with low/no light, slippery surfaces and with low/no oxygen and toxic gases.
According to the International Marine Organization (IMO) one person is killed on average every week from accidents in such enclosed spaces. Further, during the inspection process the ship cannot operate, posing high pressure on surveyors to work quickly. Major 5-year dry dock inspections are associated with around 1 M€ per vessel, resulting in about 11B€ per year for the whole industry. The reason for these high costs is that an average inspection takes up to 15 days, during which the vessel cannot operate. If the inspection time could be reduced to 3 days, the industry would save 80% (i.e. 0.8M€ per vessel).
The goal of the EU-funded project AUTOASSESS is, therefore, to employ a robot and remove human surveyors out of harm’s way, while at the same time obtaining an accurate, repeatable, and quick vessel inspection. In recent years progress has been made in aerial systems, or drones for mapping and inspection.
Even though the inspection of enclosed marine structures in challenging conditions is still a problematic task, an aerial multi-robotic human centric system, with automated AI based scanning, mapping and Non-Destructive Testing (NDT) has the potential to remove the need for human inspection. Only by combining and integrating the latest developments in collision-tolerant UAS, multi-modal Simultaneous Localization and Mapping, path planning, autonomous drone racing, aerial manipulation, miniaturized NDT sensors and Machine Learning-based defect identification it is possible to deploy drones in these tight spaces for inspection purposes. This superhuman approach would also decrease time and costs, as inspections will take as little as 1 day, saving the industry multiple billion euros per year.
Project high-level key objectives (KO):
- KO1: Autonomous Exploration and Overall Assessment of Confined Ballast Water Tanks
- KO2: Comprehensive Inspection of Ballast Tanks & Cargo Holds including NDT Measurements
- KO3: Autonomous aerial NDT with ultrasonic testing hardware less than half the mass of currently available solutions, cm level position accuracy and NDT measurement accuracy better than 100 μ m
- KO4: ML system capable of detecting >95 % of defects across different vessels in a range of 20+ years of age.
- KO5: Transforming Robotic Scans of Ballast Tanks and Cargo Holds into Actionable Data
- KO6: Decision Support System capable of planning and initiating missions, and post processing past Missions
- KO7: Quantify repeatability and accuracy of autonomous inspections via three demonstration applications
The consortium operating the project:
DTU -Technical University of Denmark (Denmark) is the coordinator of the AUTOASSESS consortium composed by 16 members from 9 different countries: NTNU -Norwegian University of Science and Technology (Norway), Technical University of Munich (Germany), University of Twente (Netherlands), ScoutDI (Norway), Cognite (Norway), FAYARD A/S (Denmark), GLC -Glafcos Marine Ltd. (Greece), F6S (Ireland), DNV (Norway), Euronav (Belgium), Danaos Shipping Co. Ltd (Cyprus), Torvald Klaveness (Norway), University of Zurich (Switzerland), Flyability (Switzerland) and Sensima Inspection Sàrl (Switzerland).
The consortium includes many of the world leaders in the field of UAS-based inspections and brings along the latest research and development in this area. Vessel owners and surveyors are also among the partners, enabling an end-to-end survey solution which would save around 50 lives per year and provide more reliable, accurate and less costly inspection data.
AUTOASSESS will also support the participation and integration of external technology providers in the development and enhancement as well as the extension of existing use cases following the identification of particular challenges. Up to 14 startups/SMEs will be selected, via two open calls.
AUTOASSESS will run for 48 months, with an overall budget of 13M€.
