Cost-effective, Safe and AI-based Autonomy Software for Ships

Around the globe, the need for efficient, safe and sustainable waterborne mobility is growing. There is also an urgent need to decarbonize the maritime industry, which comes with a substantial cost and can only be realized if coupled with leaner operations. This need is reinforced by a global shortage of trained maritime crew, inhibiting further growth of waterborne mobility. An increasing number of initiatives point to autonomy as a key enabler for this transition. Based on these trends, we expect to see the emergence of a range of electric and autonomous vessel concepts to serve societal and business needs. Ensuring safety when introducing new autonomous transport modes is an important prerequisite for its adoption into society. The automotive industry has worked towards autonomy for over a decade with investments exceeding 100 billion dollars, resulting in a range of cost-efficient and highly performant autonomy hardware products, as well as a huge body of knowledge in the form of published research and international standards. The types of sensors, algorithms, and safety methodologies used in automotive and waterborne autonomy are very similar. Leveraging the developments from automotive can therefore drastically reduce the cost and time needed to realize scalable autonomous waterborne transport. This project will focus on three key areas: 1) Safe and cost-efficient integration of automotive autonomy products into autonomous vessels. 2) Agile design and development processes suitable for safety-critical and AI-based software. 3) Methods for establishing and maintaining a safety case, which is a structured argument, backed by evidence, to demonstrate the safety of a system. The research will involve review of the scientific literature and international standards, collaboration with relevant automotive equipment manufacturers (OEMs) and internationally recognized automotive experts, and a pilot implementation on a Zeabuz autonomy product.

Around the globe, the need for efficient, safe and sustainable waterborne mobility is growing. There is also an urgent need to decarbonize the maritime industry, but this comes with a substantial cost and can only become a reality if coupled with leaner operations. This need is reinforced by a global shortage of trained maritime crew, inhibiting further growth of waterborne mobility. An increasing number of initiatives and references point to autonomy as a key enabler for this transition. Summarizing these trends, it appears inevitable that we will see the emergence of a range of electric and autonomous vessel concepts to serve societal and business needs, and with this the need for scalable, trustworthy, and safe maritime autonomy. The automotive industry has worked towards autonomy for over a decade with combined investments exceeding 100 billion dollars, resulting in a range of cost-efficient and highly performant autonomy hardware products, as well as a huge body of knowledge in the form of published research and international standards. The types of sensors, AI algorithms, and safety tools that are needed for enabling waterborne autonomy are very similar to what has been developed for automotive. Leveraging the developments from automotive can therefore drastically reduce the cost and time needed to realize scalable autonomous waterborne transport. This project will focus on three key areas where automotive developments can be leveraged: 1) Safe and cost-efficient integration of automotive autonomy products into autonomous vessels. 2) Agile design and development processes suitable for safety-critical and AI-based software. 3) Methods for establishing and maintaining a safety case. The research will involve systematic review of the scientific literature and international standards, collaboration and dialog with relevant automotive OEMs, collaboration with internationally recognized automotive experts, and pilot implementation on a Zeabuz autonomy product.