基于粗大误差检测和补偿的改进型EKF动态目标跟踪算法

Gross Error Detection and Compensation Based Modified EKF Algorithm for Dynamic Target Tracking

卡尔曼滤波作为当前动态目标跟踪中的常用滤波算法,研究其动态跟踪的准确性对于军事制导,交通导航等领域具有重大意义;针对动态系统目标跟踪观测过程中存在的坏值、静差和漂移3种粗大误差,基于传统扩展卡尔曼滤波(Extended Kalman Filter,EKF)算法框架,引入了一种粗大误差检测和补偿方法,实现了对动态系统观测值中粗大误差的准确辨识和优化补偿,使得扩展卡尔曼滤波能够结合粗大误差检测和补偿方法有效地排除观测值中的粗大误差,滤波后的状态估计值更加准确地逼近真实值;经过仿真实验和对比,提出的改进型EKF算法能有效地排除粗大误差观测值对状态预测过程的影响,并且实现了对动态系统目标的准确跟踪,这大大提高了动态目标跟踪的精确度.

Kalman filtering is a commonly used filtering algorithm in current dynamic target tracking.It is of great significance to study its dynamic tracking accuracy for military guidance,traffic navigation and other fields.Three kinds of gross errors including outliers, biases and drifts,are considered in the observations of dynamic target tracking system.Based on the traditional Extended Kalman Filter(EKF)framework,a methodology of gross error detection and compensation realizes the accurate identification and compensation of the gross errors in the observations.So that the modified EKF can effectively eliminate the gross errors in the observations by combining the system equations and the results of state estimation is closer to the true values.After simulations and comparisons,the modified EKF algorithm achieves accurate tracking of dynamic targets by effectively eliminating the influence under the condition of different types of gross errors in the observations.It can greatly improve the accuracy of dynamic target tracking.