非线性无序测量无迹滤波在地面跟踪目标指示器的应用

对于多传感器融合系统,在处理中心往往由于通信数据链的失误,出现滞后数据(即无序量测),这对数据不利于融合的正常进行.本文对此提出一种最优和次优的无序量测无迹卡尔曼(UKF)滤波器,基于UT变换,把多步滞后量测转化为等价一步滞后量测,用无序量测直接滤波更新.通过最优、次优UKF滤波器和扩展KF滤波器在不同情况下对地面跟踪目标指示器(GMTI)组合跟踪的无序量测处理作比较,可看出本文所述滤波器提高了非线性条件下的无序量测滤波精度.

Application of nonlinear out-of-sequence measurement Unscented Kalman filtering in GMTI tracking

For multi-sensor fusion system,the center processor could not always receive the measurements.Because of communications mistakes,the measurement would be delayed,which was named as out-of-sequence measurement,and this measurement was not conducive to normal fusion.An optimal and sub-optimal Unscented Kalman filter for filtering out-of-sequence measurement was proposed.The estimation was directly updated with out-of-sequence measurement,and multiple steps lag measurement was transformed to one step lag equivalent measurement based on unscented transformation.Then the optimal and sub-optimal Unscented Kalman filter and Extended Kalman filter were used for out-of-sequence measurement in the combined GMTI sensors target tracking.The comparison of the simulation results shows that the presented algorithms enhance the estimation accuracy for nonlinear out-of-sequence measurement.