利用地理信息的广义对地跟踪研究

地理信息能够减少由于地面目标由于机动而带来的建模不确定性,因而有助于提高对地跟踪性能。通过利用地理信息自适应精简目标运动模型集,提出了一种广义对地跟踪框架,以避免模型间的过度竞争,减少了计算负荷。基于此框架,进一步给出了一种广义的地图调整方差法(EMTV),并给出了自适应模型集精简及模型参数设计的策略。仿真结果表明,EMTV算法与传统的卡尔曼滤波及地图调整方差法(MTV)相比,有较好的性能。

Research on general ground target tracking based on terrain information

By introducing terrain information, the ground target tracking performance may be further improved because the constraints of the terrain information are helpful to decrease the target modeling uncertainties resulting from unexpected maneuvers. A general ground target tracking scheme is proposed that adaptively determines a reduced model set, so that the "excessive competition" of multiple models is avoided and the computation burden is reduced. Furthermore, the extended map tuned variance (EMTV) method is presented based on the above scheme. In the EMTV, the strategy of adaptively obtaining the reduced model set and adjusting model-conditional variance is also presented. The results of simulation show that EMTV is the best compared with the traditional Kalman filter and the map-tuned variance (MTV) method.