Artificial intelligence (AI) or machine learning (ML) algorithms in collaboration with an assembly of environment perception sensors provide an autonomous vessel with precise location and situational awareness. However, the following two gaps still need to be addressed in this research domain. Firstly, to enhance the traditional sensor assembly and related AI techniques and study their performance under wintertime sea and weather conditions. Especially, the prevalent sea-ice conditions need to be detected via on-board RGB and infra-red cameras and validated using LiDARs under all-weather and visibility conditions as part of the autonomous situational awareness process. Secondly, the resilience of the sensor assembly needs to be investigated in the presence of intentional and unintentional disruptions to the vessel’s positioning system due to GNSS signal jamming. This includes a study of the nature, possible causes, and impact of GNSS interference on the overall sensor assembly. This project employs the cruise vessel MS Tallink Megastar, which sails daily between Helsinki and Tallinn as the validation platform. The possibility of wintertime sea-ice along the Helsinki-Tallinn route makes this an ideal test corridor. Furthermore, occasional cases of GNSS disruptions have been encountered during vessel berthing operations in both these ports. This project will investigate if a multi-constellation (especially GPS and Galileo) and multi-frequency (L1/E1 and L5/E5) combination provides resilience to the overall sensor assembly under such scenarios.
