基于改进核极限学习机和集成学习理论的目标机动轨迹预测

为了提高目标轨迹预测的精度以及预测模型的泛化能力,提出基于改进蝙蝠算法优化的核极限学习机(Kernel Extreme Learning Machine,KELM)和集成学习理论目标机动轨迹预测模型。构建KELM模型,并采用改进的蝙蝠算法对KELM的参数进行优化；以优化后的KELM神经网络为弱预测器,结合集成学习算法生成强预测器,通过训练不断优化强预测的结构和参数,得到一种基于集成学习理论的目标机动轨迹预测模型；基于不同规模的样本,将所得预测模型与逆传播神经网络、支持向量机和极限学习机等模型进行对比分析。仿真结果表明:所提目标机动轨迹预测模型具有较好的预测精度和泛化能力。

Maneuver trajectory prediction of target based on improved KELM and ensemble learning theory

In order to improve the forecasting accuracy and generalization ability, a target maneuver trajectory forecasting approach based on ensemble learning theory and KELM (kernel extreme learning machine) optimized by the modified bat-inspired algorithm was proposed. A KELM model optimized by improved bat-inspired algorithm was constructed. Combined with the ensemble learning theory, the improved KELM neural network was regarded as weak predictor to generate strong predictor, the structure and parameters of the strong predictor were continuously optimized through training, and a target maneuver trajectory prediction model based on the ensemble learning theory was obtained. Based on samples of different sizes, the prediction performance of the model proposed in this paper was compared with BP (back propagation) neural network, support vector machine and extreme learning machine. The simulation results show that the generalization ability and prediction accuracy of the prediction model proposed is good.