유조선의 선수각 제어를 위한 RCGA 기반 선형 능동 외란 제거 제어기 설계

Abstract

In the maritime industry, autopilot system, which control a ship's rudder angle exposed to various internal and external disturbances during navigation, are of significant importance for the advancement of Maritime Autonomous Surface Ships (MASS) grounded in digital technology and artificial intelligence. To enhance autopilot system, a Linear Active Disturbance Rejection Control (LADRC) design was proposed. Here, the rate of change in a ship's heading due to yaw movement was selected as a crucial factor, and a 3-degree-of-freedom nonlinear tanker model of a ship's motion equations, including the rudder, was used. Disturbances applied to the system and internal uncertainties were considered as total disturbances, and a controller design for more robust heading control was proposed by estimating these in real time. The gains of the proposed controller were tuned using the Real-Coded Genetic Algorithm (RCGA), minimizing the evaluation function using the Integral Absolute Error (IAE). Simulations were conducted considering the squat effect affecting navigation during ship operation, environmental disturbances, measurement noise, and uncertainties and errors in the model. Consequently, it was confirmed that the proposed controller effectively maintains and follows the set ship’s heading, even when considering modeling errors of the nonlinear model, uncertainties, noise from measurement sensors, and disturbance factors such as waves. Additionally, the economic aspect of the proposed controller was validated by considering the energy consumption through the evaluation index.