Numerical modelling of collision avoidance for marine autonomous vessels

• Andrey V. Artemyev

  Andrey V. Artemyev. Maritime State University named after G.I. Nevelskoy. Vladivostok. Russia

• Vladimir A. Petrov

  Vladimir A. Petrov. Maritime State University named after G.I. Nevelskoy Vladivostok. Russia

• Grinyak V.M.

  Victor M. Grinyak. Vladivostok State University Vladivostok. Russia

The paper is devoted to the problem of ensuring maritime traffic safety. The problem of safe passage of ships in water areas with heavy traffic is considered. A geometric interpretation of this problem is proposed when the relative motion of a pair of ships is considered. In this case, the locus of points of the speed vector of the controlled vessel is determined, corresponding to dangerous and safe speeds and courses. The paper emphasizes the special nature of the movement of autonomous surface vessels, which imposes a number of restrictions on the use of known methods of safe divergence and makes the transition from modeling the movement of pairs "ship-ship" to modeling the movement of the type "ship-group of ships" relevant. It is indicated that in the case of a group movement, the actions of an autonomous vessel must be predictable and transparent to the crew of convenional vessels. Therefore, when solving the problem of safe divergence, crews of both autonomous
craft and conventional ships need to know the intentions of the oncoming vessel. It is the knowledge of how the oncoming ship intends to maneuver that allows you to correctly choose your own move-ent and avoid getting into difficult situations (“navigation traps”). The paper proposes a new ap- proach to choosing the parameters of a possible vessel maneuver that takes into account the possible intentions of other ships in the movement. This makes it possible to accurately interpret the existing passing rules for maritime autonomous surface vessels. Possible tools for the exchange of information on planned routes between ships and the choice of an appropriate maneuver are discussed. Such an exchange of routes makes it possible to assess the intentions of ships in advance and make a decision on maneuvering that is acceptable to the entire group of ships that navigate in high density waters. The work is accompanied by examples of calculating the maneuvering parameters of a group of ships, which demonstrate the applicability of the described method. The software used to simulate the
movement of groups of ships and the numerical study of the problem under consideration are described. Recommendations are given for setting up full-scale experiments.

Keywords: maritime safety, unmanned navigation, е-navigation, a-navigation, risk assessment, collision avoidance.