Path-following control of autonomous tugs based on Gaussian process regression and arithmetic optimization algorithm

• Published in: 2024 3rd Conference on Fully Actuated System Theory and Applications (FASTA) pp. 1234 - 1241
• Main Authors: Li, Shijie, Li, Yang, Hu, Xinjue, Liu, Jialun, Wei, Muheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 10-05-2024
• Subjects: arithmetic optimization algorithm; Autonomous surface ships; Computational modeling; Gaussian process regression; Gaussian processes; path-following control; Prediction algorithms; Predictive models; Simulation; Task analysis; Transforms
• DOI: 10.1109/FASTA61401.2024.10595296

Intelligent tugs with high autonomy are becoming a worldwide research focus, particularly for the needs of automated port operations. The path-following control of tug is an important part of intelligent tugs research. This paper proposes a path-following control method that combines Gaussian Process Regression (GPR) and Arithmetic Optimization Algorithm (AOA) for autonomous tugs. A GPR model is constructed to identify the correlation between the heading angle and the rotation angle of the azimuth thrusters, which acts as the prediction model in the path-following controller. The AOA is utilized to determine the optimal control inputs for each sampling interval, adopting the sine-cosine search strategy instead of addition and subtraction operations and introduces an inertia factor to improve the computational efficiency of AOA. A Line-of-Sight (LOS) algorithm is used as the guidance law to transform reference waypoints into reference heading angles, and the path-following controller is designed based on the GPR model and AOA. Simulation results show that the proposed method performs well in the path-following task without having prior knowledge regarding the hydrodynamic coefficients and ship parameters.