DLR e.V., Institute for the Protection of Maritime Infrastructures

Assuring safe and compliant design.

The dimensions of safe operation of a MUM are technical reliability, legal admissibility and social acceptance. The DLR Institute for the Protection of Maritime Infrastructures will provide research contributions to these aspects and demonstrate their success together with the research partners during the project demonstration.

For the operation of a MUM, the capabilities under water are in the foreground, but the safe participation in maritime traffic at the surface is also an important aspect of the later use. To this end, a MUM must meet the requirements set by the IMO and flag states for the operation of remotely operated or autonomous vehicles without a crew on board with regard to collision prevention. Essential for this are sensors for object and environment detection and the evaluation of the respective sensor data. The development of such a sensor system based on gated viewing technology and the development of AI-based evaluation methods by DLR serve to ensure that the vehicle can reliably avoid obstacles and behave in accordance with collision prevention regulations. Through close collaboration with thyssenkrupp Marine Systems, these technologies will be integrated into the vehicle’s control system.

The goal of further research by DLR is to determine, through legal and social science research, which technical design decisions will be influenced by (future) regulation and user requirements, so that these requirements can be taken into account at an early stage in the development of the MUM. This research takes place in close coordination with the associated partner Bundesamt für Seeschifffahrt und Hydrographie (BSH) and relevant industry partners.

Underwater research by DLR will also result in a capability of the MUM that is significant for near-market demonstration of capabilities. Autonomous grasping of an object by an ROV deployed by the MUM requires reliable object identification and trajectory calculation. This will be enabled by means of optical object recognition and AI-based data analysis and will be integrated into the sub-project of the University of Rostock.