海洋移动目标多模型运动预测方法

摘要

This paper analyze the characters of maritime moving target, develops a grey forecasting method of interpolating before prediction,and improves track change prediction (TCP) and potential area prediction (PAP). By integrating uniform motion prediction (UAMP), TCP, TP and PAP,a multi-model prediction method is presented. Based on statistical comparison between observation and prediction during slipping time window,This paper evaluates recent performance of each prediction method and decides the appropriate one of candidate prediction methods. Simulation shows that multi-model prediction method reduces prediction risk compared with single one prediction method and improves prediction precision.%分析海洋移动目标的运动特征,提出了预测前插值的灰色预测方法,改进了航迹变更预测和潜在区域预测模型.通过集成匀速运动预测、航迹变更预测、基于航迹的预测和潜在区域预测,提出多模型运动预测方法及其模型参数配置依据,根据滑动时间窗口中的观测数据与预测值的统计比较,评价不同预测方法的近期预测效果,决策下阶段适宜选择的预测方法.仿真实验表明多模型运动预测比使用单一运动预测方法降低预测风险,提高了预测精度.