|
|
Group of Safety and Logistics of Maritime Transportation > 5.4 Maritime Safety and Human Factors > Publications
Rong, H., Teixeira, A.P. and Guedes Soares, C. (2022), “Maritime traffic probabilistic prediction based on ship motion pattern extraction”, Reliability Engineering and System Safety, Vol. 217, 108061.
This paper proposes a novel maritime traffic prediction method based on ship motion pattern extraction, considering ship destination prediction and ship trajectory prediction within a specific route. To extract ship motion patterns from historical Automatic Identification System (AIS) data, traffic departure-arrival areas are determined based on the Order Points to Identify the Clustering Structure (OPTICS) algorithm and ship trajectories following the same itinerary are clustered. A maritime traffic network abstraction consisting of nodes that represent waypoint areas and navigational legs is constructed to represent the maritime traffic at a larger scale. Multinomial Logistic Regression (MLR) and Gaussian Process (GP) regression models are developed and applied for predicting probabilistically the ships’ destinations and their trajectories along the ship route, respectively. Based on these models, the uncertainty on the ship’s future position can be estimated given its current state, characterised by the ship position, ship velocity, rate of turn and ship type. The proposed method is capable of long-term ship position prediction and provides information on the maritime traffic 10, 30 and 60 minutes ahead when the method is applied to all ships navigating in a study area. A density map is then produced by integrating the information on the predicted ships’ future positions and related uncertainties, which provides insight on the traffic density and hotspot distribution at a larger scale. The presented method may assist maritime authorities to improve the efficiency of maritime traffic surveillance and to develop strategies to improve navigation safety.
Request a copy of this article
For information about all CENTEC publications you can download: Download the Complete List of CENTEC Publications
|