Artificial intelligence for ship speed management during navigation in the Arctic

Abstract

Difficult ice conditions are one of the major threats to ships, shipboard crew, and cargoes carried in Arctic waters. Ship-speed choice is influenced by many factors including the navigational conditions (ice, weather, bathymetry, etc.), the purpose of the voyage, personal preferences of the shipmaster (and/or the ice pilot) as well as onboard regulations (e.g., ice passport, polar water operational manual). Faults and incidents during navigation in ice (i.e., excessive ship speed) almost always result in ship damage. This can be fixed with the help of electronic environments that are sensitive and responsive to the presence of ice. This work describes AI-infused systems that can capture details of ice conditions around the ship and place them in the context of safe ship speed and ice conditions reporting. The focus is on online sensors-to-service solutions that run in the background and rely on shipboard sensors. Apart from just being AI-infused information systems, some solutions could also be applied to provide an early alert to the user if the user is approaching potentially dangerous ice object(s) at elevated speeds. The presented systems are not tested during real ship operations, but we provide illustrative examples for exploration of the solution space based on Arctic data. Ice navigation is about masters taking decisions and keeping ships and their crew safe. AI is to enhance human performance and not to replace human operators.