基于遗传算法-v支持向量回归的船舶轨迹预测

Ship trajectory prediction based on genetic algorithm-v support vector regression

为了提高船舶轨迹预测精度,避免船舶海上航行事故的发生,本文采用遗传算法对v-支持向量回归进行参数寻优,以此来分别构建关于经纬度的船舶轨迹预测模型。选取水上移动业务标识码为356772000的货船在2022年6月的船舶自动识别系统数据作为研究对象。将该模型的预测结果分别与粒子群优化算法和网格搜索算法优化的v-支持向量回归模型、遗传算法-支持向量回归模型进行比较。实验结果表明:遗传算法v-支持向量回归模型关于航迹经、纬度预测结果的均方误差、平均绝对百分比误差和平均绝对误差相比于其他模型最低,关于经度分别为4.29×10^(-7)(°)、4.50×10^(-4)和5.47×10^(-7)(°)2,关于纬度的分别为1.82×10^(-6)(°)、4.02×10^(-3)和1.08×10^(-3)(°)2。基于遗传算法-v支持向量回归模型的预测效果最好,预测误差波动最小。本文将遗传算法与v-支持向量回归相结合,为船舶轨迹预测模型的优化提供参考,也为海上智能交通提供思路。

In order to improve the accuracy of ship trajectory prediction and avoid the occurrence of ship maritime navigation accident,genetic algorithm(GA)is used to optimize the parameters of v-support vector regression(v SVR)to construct ship trajectory prediction models on longitude and latitude,respectively.Automatic identification system(AIS)data from a cargo ship with maritime mobile service identity(MMSI)as the 356772000 was selected as the subject of the study.The prediction results of the model were compared with the v SVR models optimized by the particle swarm optimization(PSO),the grid search algorithm(GS),and the GA-SVR model,respectively.Experimental results show that mean square error(MSE),mean absolute percentage error(MAPE)and mean absolute error(MAE)of the GA-v SVR model on the prediction results of the trajectory longitude and latitude are the lowest compared with other models,with longitudes of 4.29×10^(-7)(°),4.50×10^(-4),and 5.47×10^(-7)(°)2,and latitudes of 1.82×10^(-6)(°),4.02×10^(-3),and 1.08×10^(-3)(°)2.The GA-v SVR model has the best prediction effect and the least fluctuation of prediction error.In this paper,the GA is combined with the v SVR to provide a reference for the optimization of the ship trajectory prediction model,which also provides new ideas for marine intelligent transportation.