Learning from Algorithm Based decision Support systems (COLREG-LABS)
Research Project, 2020
– 2022
Ships with different levels of automation (i.e. ranging from fully manned to fully autonomous) will be involved in traffic situations where COLREGs are applicable (e.g. Rule 8 explicitly states that "any action taken to avoid collisions shall, if the circumstances of the case admit, be positive, made in ample time and with due regard to the observance of good seamanship").
Smart vessels and automated systems will require a reliance upon algorithm-based solutions that infer interpretations of the COLREGs. In reality, decisions made on manned ships are influenced by the experience of the operator and his/her interpretation of good seamanship. Furthermore, the geographical area, traffic pattern, complexity of the traffic situation, ship type, as examples, will influence operator decision-making and application of COLREGs.
Can experience, good seamanship, situational awareness and other non-technical skills that impact upon safe and efficient navigation be implemented into algorithms? Can traffic situations that have both human operators and smart/autonomous vessels be safely resolved?
Participants
Scott Mackinnon (contact)
Chalmers, Mechanics and Maritime Sciences (M2), Maritime Studies
Reto Weber
Chalmers, Mechanics and Maritime Sciences (M2), Maritime Studies
Collaborations
Fraunhofer Center for Maritime Logistics and Services
Hamburg, Germany
RISE Research Institutes of Sweden
Göteborg, Sweden
SEAMADE/Kamahura Teknik AB
Göteborg, Sweden
Funding
Swedish Transport Administration
Funding Chalmers participation during 2020–2022
Lighthouse
Funding Chalmers participation during 2020–2022
Related Areas of Advance and Infrastructure
Sustainable development
Driving Forces
Transport
Areas of Advance
Basic sciences
Roots
Chalmers Maritime Simulators
Infrastructure