Vessel route planning based on traffic historical data clustering

• Grinyak V.M.

  V. M. Grinyak. Vladivostok State University of Economics and Service. Vladivostok. Russia

This work is about navigation safety of marine traffic at sea areas. One of the elements of traffic management in areas of heavy navigation is the system for establishing the ship routes. It is a set of restrictions imposed by a certain ship traffic pattern, traffic rules adopted in a particular water area. Ships that violate the rules of movement in the water area, from the point of view of classical concepts, may not present a danger now. However, vessels can lead to an intractable dangerous navigation situation after a while. The navigational safety of vessels can only be ensured if they follow a certain traffic pattern in conditions of heavy traffic. The paper considers the problem of planning a route for a vessel to cross water areas with heavy traffic. It should be borne in mind that the trajectory of the vessel should be consistent with established navigational practices and collective navigation experience. Isolation of established patterns of movement of a specific sea area from retrospective information about its traffic by clustering the parameters of vessel movement is a promising way to identify such an experience. The task is considered relevant due to the promising development of unmanned marine vehicles. Ship's passage routes planning passage should be carried out considering the specified restrictions when moving through the water areas with established routes. Isolation of patterns of movement of a specific marine area from retrospective information about its traffic is a possible way of identifying
these restrictions. Model representations of such a problem can be formulated based on the idea of clustering the parameters of ship traffic. The model of the route planning problem is based on finding the shortest path on a weighted graph. There are several ways to construct
such a graph: a regular mesh of vertices and edges, a layered mesh of vertices and edges, a random mesh of vertices and edges, vertices and edges based on historical data. The weight of the ribs is proposed to be set as a function of the “desirability” of a particular course of the vessel
for each point of the water area, considering the identified movement patterns. The water area is divided into sections and for each of them clustering of rates and velocities is performed. Possible clustering methods are discussed in the paper, and a choice is made in favor of subtractive clustering, which does not require preliminary specification of the number of clusters. Services of the Automatic Identification System can serve as a source of data on water area traffic. The paper shows the possibility of using AIS data available on specialized Internet resources. These data reflect well the summary features of the water area traffic despite their “sparseness”. The historical AIS data of sea traffic at Tokyo bay is used for identifying traffic schema and ship routes planning with the model designed under presented research.

Keywords: navigation safety, route planning, ship track system, clustering, AIS.