Wave data prediction with optimized machine learning and deep learning techniques

Abstract Maritime Autonomous Surface Ships are in the development stage and they play an important role upcoming future. Present generation ships semi-autonomous controlled by ship crew. The performance of is predicted using data collected from with help machine learning deep methods. Path planning for autonomous necessary estimating best possible route minimum travel time it depends on weather. However, even during navigation, there will be changes weather should order to reroute ship. information such as wave height, period, seawater temperature, humidity, atmospheric pressure, etc., external sensors, stations, buoys, satellites. This paper investigates ensemble approaches seasonality approach prediction. historical meteorological six stations near Puerto Rico offshore Hawaii offshore. We explore techniques collected. divided into training testing apply models predict test data. hyperparameter optimization performed find parameters before fitting train data, this essential results. Multivariate analysis all methods errors computed models.