非合作机动目标平滑跟踪研究

Non-Cooperative Target Tracking Based on Smoothers

本文针对非合作机动目标跟踪问题,引入平滑思想作为跟踪估计方法,构造了一种新的跟踪方法;首先,分析了经典卡尔曼平滑器(KS),结合交互式多模型(IMM)方法,进一步构造并分析了交互式多模型平滑(IMMS)方法;同时,针对目标机动特性以及雷达目标跟踪量测产生的强非线性等问题,构造了包括对非合作目标的随机概率选择模型的场景,结合交互式多模平滑方法,对机动目标跟踪进行仿真验证,实验结果验证了新方法的有效性。

This paper presents the problem of real--time tracking measurements of maneuvering target, discusses the optimal kalman smoother, introduced smooths methods, and compared the methods of Kalman filter and Kalman smoother, Interacting Multiple Model Smoother （IMMF） based on the Multiple Model idea. Meanwhile, designed the non--cooperative target random probability selection model scene, meanwhile, combined extended Kalman and unscented Kalman methods with Interacting Multiple Model Smoother methods , simula- ted strongly non--linear multi--radar maneuvering target tracking, experiment results show the effectiveness.