基于新息序列修正的输入估计算法

针对传统输入估计算法跟踪机动目标时检测概率和估计精度不高的缺点,提出一种改进的输入估计跟踪算法。算法利用对检测窗内新息序列进行修正以及对卡尔曼滤波状态变量进行补偿的方法,消除了已有机动检测估计对后继机动输入估计的影响,提高了对机动的检测概率和估计精度。理论分析和仿真结果表明,改进后的算法相比原有算法和IMM3算法,不仅跟踪性能有显著提高,而且对加速度缓变的机动也有较好的跟踪能力。

Input estimation algorithm based on modification of innovation sequence for maneuvering target tracking

Aiming at the imprecision of maneuvering target tracking by using traditional input estimation algorithms,a modified maneuvering target tracking algorithm of IE is proposed.It improves the precision of detection and estimation of successive jump input through modification of innovation sequence in detection window and compensation of state variable.The approaches significantly improve the detection probability as well as estimation precision.Simulations verify that modified algorithm performs better than the original IE and IMM3 algorithm in tracking of continuous maneuvering.It can track maneuvering targets with variable acceleration as well.