摘要
针对非线性观测的目标跟踪问题 ,本文在极坐标观测方程条件下 ,基于去偏转换测量卡尔曼滤波算法 (CMKF D) ,计算观测值转换到直角坐标系后的观测误差均值和方差。以此为基础提出了一种由前两点观测值建立初始估计的方法。仿真结果表明 ,使用该方法建立初始估计 ,能使滤波器迅速收敛 ,提高初始阶段的跟踪精度。
Based on the algorithm of the Converted Measurement Kalman Filter with Debiasing (CMKF-D), the error mean and error coviarance can be computed with statistical method after the polar measurements are transformed to Cartessian. In this paper, a new method on how to establish the initial estimation of target tracking by the first and second measurements is put forward under the condition that the measuring equation is in polar coordinates. Compared with the old one, the new method shows a good result no matter whether the measuring error is small or great.
出处
《雷达与对抗》
2002年第2期17-21,72,共6页
Radar & ECM
关键词
目标跟踪
初始估计
去偏转换
测量
卡尔曼滤波
target tracking
initial estimation
Converted Measurement Kalman Filter with Debiasing
