|
|
Group of Marine Dynamics and Hydrodynamics > 2.3 Ship Manoeuvring and Control > Publications
Meng, Y., Zhang, XK., Zhang, XF., Duan, YT., Xu, H.T. and Guedes Soares, C. (2025), Multi-step online prediction of 4 degrees of freedom motion under various wind scales based on pre-trained strategy and ensemble learning framework, Ocean Engineering, Vol. 339, 122096.
Accurate multi-step prediction enhances navigational safety and optimises the performance of autonomous control and collision avoidance systems. Given the limitations of single data-driven models and single-step prediction schemes, this paper proposes a multi-step online motion prediction scheme based on the weighted multi-kernel relevance vector machine, pre-trained strategy, and ensemble learning framework. To evaluate the effectiveness and generalisation performance of the proposed scheme, four degrees of freedom motion data are collected under various wind conditions. A subset of motion data is used as training samples to pre-train three sub-models, whose parameters are tuned using an improved grey wolf optimiser algorithm. To integrate the prediction results of the sub-models, this paper develops an ensemble learning framework and proposes an adaptive weight updating rule for the sub-model using the mean absolute error as the evaluation standard. Based on the proposed framework, this paper conducts single-step, 5-step, and 10-step predictions of ship motion under various wind conditions. The comparison with a single sub-model confirms the applicability and efficiency of the proposed scheme. The final experimental results show that the proposed multi-step prediction scheme maintains good generalisation and time efficiency while ensuring low model complexity. The mean squared errors of the single-step and multi-step prediction results are below 8 X 10-3 and 0.28. This scheme can provide reliable prior information for the sea trial and safe navigation of ships and for the adaptive controllers of unmanned surface vehicles.
Request a copy of this article
For information about all CENTEC publications you can download: Download the Complete List of CENTEC Publications
|