基于GA-BP神经网络的船舶航迹预测

Prediction Model of Ship Trajectory Based on GA-BP

为了更精确地预测船舶航行轨迹,针对BP神经网络在船舶航迹预测过程中容易陷入局部最优、预测误差较大等问题,提出采用GA算法优化BP神经网络进行船舶航迹的预测.通过构建GA-BP神经网络,对BP神经网络中的初始权值和阈值进行优化,综合考虑经纬度、航速、航向等行为特征,实现对船舶航迹的预测.以太平口水道区域内某船舶真实航行的AIS信息作为测试案例进行验证,并将预测结果与BP神经网络预测结果进行对比.结果表明,基于GABP神经网络的航迹预测模型的预测误差不超过(410^-4)°,与BP神经网络相比,MAPE和RMSE更低,具有较高的预测精度.

In order to predict the ship's trajectory more accurately,aiming at the problems that BP neural network is easy to fall into local optimum and has large prediction error in the process of ship track prediction,this paper proposes to use GA algorithm to optimize BP neural network for ship track prediction.By constructing GA-BP neural network,the initial weights and thresholds of BP neural network are optimized,the network is trained to realize the prediction of ship track.The AIS information of a ship sailing in taipingkou waterway area is taken as a test case to verify,and the prediction results are compared with those of BP neural network.The results show that the prediction error of the model based on GA-BP neural network is less than 410^-4°,compared with BP neural network,MAPE and RMSE are lower,and the prediction accuracy is higher.