In maritime transport, VHF radio is the number one communication channel, where officers of seagoing vessels coordinate their behavior in individual traffic situations with each other and with coastal radio stations. Even if there are traffic rules at sea, the different speed, size and maneuvrability of ships regularly lead to a situation- specific need for information. The conditions prevailing at sea and on board make communication difficult. Ambient noise or bad weather conditions in the form of rain or storm can negatively affect the intelligibility and transmission of radio messages. For this reason, the German Maritime Search and Rescue Association DGzRS, together with the Fraunhofer CML and Fraunhofer IAIS as well as Rhothetha Elektronik GmbH, launched the ARTUS research project (short for „Auto- mated Transcription of Maritime VHF Radio Communication for Search and Rescue (SAR) Missi- on Coordination“) last year. The aim of ARTUS is the automatic transcription of VHF radio messages and the simultaneous localization and identification of ships in order to make the coordination of rescue operations more efficient.
The first solutions have already been implemented: a draft of the localization algorithm is being tested and the error rate in the transcription of maritime radio messages can be improvedinitially under laboratory conditions. In situations in which loud back- ground noise overlays speech or a radio operator has a distinctive dialect, the technology is confronted with special challenges. These are countered by the use of artificial intelligence in the context of speech recognition training.
In the further course of the project, the results of the automatic radio message transcription and localization will be prepared for different users and visually dis- played on their own user inter- faces. The most important information for the SAR cruiser‘s crew is the position and bearing of the casualty. The coordinator of an operation in the MRCC (Mariti- me Rescue Coordination Center) needs during a rescue operation above all a general overview of the available radio message con- tent and ships in the region. And instructors who carry out rescue missions in the ship handling simulator to train SAR forces need their own user interface to process and display relevant information and situations as part of the debriefing.
The project partners have sufficient time to implement the complex technical solutions: ARTUS is funded by the BMBF under the Federal Government‘s „Research for Civil Security“ programme until the end of February 2022.