针对目前常用的按标量、对角阵、矩阵加权融合方法过程较复杂,计算负担较重的问题,结合三维目标跟踪模型提出一种高斯马尔可夫融合方法来融合卡尔曼滤波估计值,不需要计算局部稳态滤波误差互协方差,只需要知道传感器的观测噪声方差就可以了。计算方法和操作相对简单,省略了很多繁琐步骤也能达到不错的滤波融合效果,通过仿真证明了此方法的可行性和高效性。
For the problem of complicated current fusion methods based on weighing by scalar, weighing by diagonal matrices and weighing by matrices, a new fusion method Gauss-Markov algorithm which fuses Kalman filtering estimations needing only the sensor' s observing yawp variance without the need of computing local steady state filtering error cross covariance is proposed. The computing and operation of this method is easier, it can still get good result without many cumbersome procedures. The simulation proves that the method is feasible and efficiency.