Modelling of Ship’s Trajectory Planning in Collision Situations by Hybrid Genetic Algorithm

Ship collision-avoidance trajectory planning aims at searching for a theoretical safe-critical trajectory in accordance with COLREGs and good seamanship. In this paper, a novel optimal trajectory planning based on hybrid genetic algorithm is presented for ship collision avoidance in the open sea. The proposed formulation is established based on the theory of the Multiple Genetic Algorithm (MPGA) and Nonlinear Programming, which not only overcomes the inherent deficiency of the Genetic Algorithm (GA) for premature convergence, but also guarantees the practicality and consistency of the optimal trajectory. Meanwhile, the encounter type as well as the obligation of collision avoidance is determined according to COLREGs, which is then considered as the restricted condition for the operation of population initialization. Finally, this trajectory planning model is evaluated with a set of test cases simulating various traffic scenarios to demonstrate the feasibility and superiority of the optimal trajectory.