Ship Collision Avoidance Path based on Hermit Crab Optimizer with Multiple-Objective Evolutionary Algorithm

In recent years, ship collisions causing significant risks to environmental sustainability, people safety and maritime safety. The main reasons for ship collisions are human errors, increase in number of vessels, and complex waterways which are affecting not only to environment but also to marine lives. The existing methods failed in identifying the collision in various factors Therefore, this research proposed Hermit Crab Optimization with Multiple Objective Evolutionary Algorithm (HCO-MOEA) method for optimal ship collision avoidance path. The proposed method employed on Automatic Identification System (AIS) dataset which provided textual information.The obtained data from the dataset had preprocessed by min-max scaling normalization for scaling the data into similar ranges. These preprocessed data features are selected by employed HCO for searching the optimal path. Furthermore, these selected relevant features are utilized for detecting the risk of ship collision effectively using MOEA. The proposed HCO-MOEA model performed effective results such as RSME (0.98), Enterprise value (EV) (0.25), MAPE (0.81) and $\mathrm{R}^{\wedge} 2(0.84)$. when compared with other existing models like Multi Logistic Regressor (MLR).