AI and Maritime

We'll meet before the Navigate trade show and talk about Maritime related AI. Event is supported by AI-ambassador project (more info: AI-lähettiläs) and lecturer Antti Tuomisto & project researcher Tanja Vähämäki.

Event host Mirva Salokorpi from Novia University of Applied sciences will start the event with short introduction about current activities in Novia.

We have two excellent speakers:

Maritime Situational Awareness through Multi-sensor Fusion
Fahimeh Farahnakian, post-doctoral researcher, lecturer
Multi-sensor fusion technology has been widely used in many real applications to achieve complementary detailed environment description based on the information from different sensors. With the rapid development of Deep Learning (DL) networks for computer vision tasks, DL-based fusion methods can significantly improve the performance of sensor fusion. In order to understand the main development status of DL-based image fusion methods pipeline, thoroughly and deeply, in this presentation, I first provide a comprehensive overview of a variety of DL-based image fusion methods.

MASS research at the MAST institute
Johan Lilius, professor of Embedded Systems at Åbo Akademi
The MAST (Maritime Software Technology) institute is a joint research initiative of Åbo Akademi Information Technology and Novia Maritime Academy founded in 2018. In this talk I will cover our ongoing research and discuss some the central problems that we are addressing. The main demonstrator of our research is the autonomous vessel Åboat, that has been under construction from 2019. Currently the vessel is able to autonomously navigate along a set route and has been demonstrated in the Marina at Taalintehdas. I will present the sensor and software architecture of the vessel and discuss some of the issues that we have had to solve along the way. The vessel does not yet include colission avoidance algorithms, therefore we are now actively working towards developing solutions to this problem. The talk will thus close with a discussion about collision avoidance algorithms and their COLREG-compliance.

Johan Lilius: Johan Lilius is full professor of Embedded Systems at Åbo Akademi since 2001. He is an experienced academic researcher, educator and leader. He was co-PI of the research group "Formal Methods in Programming" that was awarded the status of Centre of Excellency by the Academy of Finland. He has won the Ten-Year Most Influential Paper Award at the ACM/IEEE Conference on Model Driven Engineering Languages and Systems, and a number of Best Paper awards. In 2015 he won the Gadd Prize for Research Excellence at Åbo Akademi. He has experience in leading large academic organisations. He has been director of TUCS, Turku Center for Computer Science, and Head of the Department of IT under several periods. Under his leadership he has planned and implemented the restructuring of the department and its education as part of two major reorganizations of the University. He is a member of the steering group of the Digivisio2030 project, a large national project involving all HEIs in Finland, that is creating a new ecosystem for education in Finland. He is the founder and co-PI of the MAST Institute, a joint research institute of Åbo Akademi and Novia Maritime Academy developing tools and techniques for safe autonomous navigation of vessels. He is experienced in leading PhD reserach, he has been head of the TUCS graduate programme, he has completed the supervision of 11 doctoral students, and is curretly supervising 4 students. He has co-organised and co-chaired several workshops, summer schools and conferences. He has published over 170 research papers. His research interests include Energy-efficient Software, Safety of autonomous vessels, and Machine Learning.