P. Jędrzejowicz et al. (Eds.): ICCCI 2011, Part II, LNCS 6923, pp. 221–230, 2011. © Springer-Verlag Berlin Heidelberg 2011 Evolutionary Sets of Safe Ship Trajectories: Problem-Dedicated Operators Rafal Szlapczyński 1 and Joanna Szlapczyńska 2 1 Gdańsk University of Technology, Narutowicza 11/12, 80-233 Gdańsk, Poland 2 Gdynia Maritime University, Morska 81-87, 81-225 Gdańsk, Poland rafal@pg.gda.pl, asiasz@am.gdynia.pl Abstract. The paper presents the optimization process of the evolutionary sets of safe ship trajectories method, with a focus on its problem-dedicated operators. The method utilizes a customized evolutionary algorithm to solve a constrained optimization problem. This problem is defined as finding a set of cooperating trajectories (a set is an evolutionary individual) of all the ships involved in the encounter situation. The resulting trajectories are safe (meeting optimization constraints) while minimizing the average way loss ratio. When developing a new version of the method, the authors decided to introduce a number of changes. This upgrade enforced redesigning of the optimization process, especially the problem-dedicated collision-avoidance operators. Keywords: Evolutionary computing, multi-ship encounter situation. 1 Description of the Problem The main approaches to the problem of planning optimal ship trajectories in encounter situations are based on either differential games [1] or on evolutionary programming [2]. The authors have proposed a new approach, which combines some of the advantages of both methods: the low computational time, supporting all domain models and handling stationary obstacles, typical for evolutionary method [3, 4, 5]), with taking into account the changes of motion parameters (changing strategies of the players involved in a differential game). Instead of finding the optimal own trajectory, an optimal set of safe trajectories of all the ships involved is searched for. The search is performed in real time and assumes most probable behavior of all ships. The method is called evolutionary sets of safe trajectories [6]. It assumes that we are given the following data: − stationary constraints (such as landmasses and other obstacles), − positions, courses and speeds of all the ships involved, − ship domains (a ship domain is an area around the ship that should be free from other ships and obstacles during the voyage), − times necessary for accepting and executing the proposed manoeuvres. The goal is to find a set of trajectories, which minimizes the average way loss spent on maneuvering, while fulfilling the following conditions: