Identification of testing scenario from AIS data for intelligent ship navigation

Abstract

Ship intelligent navigation system is currently a research hotspot in the shipping industry, but the related test methods and special technical standards are still in their infancy. In the context of the massive application of artificial intelligence (AI), machine learning and other technologies in the field of intelligent navigation, the testing of intelligent navigation systems faces huge challenges. The scenario-based smart ship navigation test method is currently one of the most promising potential solutions. This paper proposes an idea and method for detecting and extracting ship intelligent navigation testing scenarios from real Automatic Identification System (AIS) data. First, it analyzes the constituent elements of the testing scenario; then, according to the collision avoidance rules and maritime practical experience, the identification model of the ship encounter situation is to establish by using ship collision model based on Distance of close point of approaching (DCPA) and Time to close point of approaching (TCPA); finally, the AIS data of the Laotieshan water area is used for the experiment. The results show that the method is feasible, and AIS can provide rich source data for intelligent navigation testing scenarios, which are helpful for subsequent intelligent ship testing and safety assessment research.