Maritime transport challenges include increasing transport volumes and more stringent environmental requirements. A shortage of skilled manpower is also a threat in the future. These challenges can be overcome by technology: this is what autonomous ships are about.
Safety and efficiency are important in maritime navigation. Efficiency, however, comes only in the 4th place within the e-Navigation strategy of International Maritime Organization (IMO). The primary listed driving factor is safety: E-navigation is the harmonized collection, integration, exchange, presentation and analysis of marine information on board and ashore by electronic means to enhance berth to berth navigation and related services for safety and security at sea and protection of the marine environment.
The ESA-funded research project is titled Artificial Intelligence / Machine Learning Sensor Fusion for Autonomous Vessel Navigation (Maritime AI-NAV) and the team will utilize the Tallink Grupp’s newest vessel Megastar for practical field tests on the Helsinki-Tallinn route on the Baltic Sea. ESA’s overall objective is to improve European know-how in the field of autonomous transports and to study how European space-based positioning and navigation infrastructure, such as Galileo and EGNOS, can contribute to enhancing scientific innovation and consequently business opportunities in this field.
Maritime-AI focuses on increasing the situational awareness, and in performing system integrity monitoring. Specifically, a sensor fusion of vision, sound, radar, lidar, GNSS/IMU, and AIS signals is studied. Note that these sensors are all listed in the front-running Rolls’ Royce AAWA whitepaper for autonomous ships.Project Objectives and Scope
– to integrate absolute positioning and environmental perception sensors
relevant for autonomous vessels,
– to implement AI techniques based on data from sensors for extracting features, recognizing them, and fusing this information,
– to perform learning-data recording and concept validation campaigns,
– to investigate how the results can enable intelligent decision making related
to sensor integrity monitoring, vessel situational awareness, and navigation safety.
– Sensors within the scope: GNSS, INS, visual cameras, microphones, RADAR,
LiDAR*. External databases such as AIS, COLREGS, etc. may be utilized.
– Possible AI techniques: Neural networks, machine learning, decision trees, Bayesian methods. 3 classes of AI methods: batch supervised/ machine learning, online learning, failure detection and correction.
This research project started in January 2019 and lasts for two years.
Please note that the views expressed on this website reflects the project partners opinions and can in no way be taken to reflect the official opinion of the European Space Agency.