CAJ | 학술논문

为了解决再入阶段的弹道目标跟踪问题，提出一种修正的强跟踪滤波算法，即强跟踪有限差分滤波算法（STFDEKF）。修正的强跟踪算法做出来以下改进：使用有限差分方法来近似多项式，不需要求解非线性函数的导数；其次，新算法中使用强跟踪的次优渐消因子对先验协方差矩阵进行修正。通过蒙特卡罗仿真实验对修正的新算法进行了仿真验证，结果表明，与现有的非线性滤波算法相比，除了计算复杂性稍大外，新算法在跟踪精度及滤波可靠性上均比扩展卡尔曼滤波器和有限差分扩展卡尔曼滤波器优越。
위료해결재입계단적탄도목표근종문제，제출일충수정적강근종려파산법，즉강근종유한차분려파산법（STFDEKF）。수정적강근종산법주출래이하개진：사용유한차분방법래근사다항식，불수요구해비선성함수적도수；기차，신산법중사용강근종적차우점소인자대선험협방차구진진행수정。통과몽특잡라방진실험대수정적신산법진행료방진험증，결과표명，여현유적비선성려파산법상비，제료계산복잡성초대외，신산법재근종정도급려파가고성상균비확전잡이만려파기화유한차분확전잡이만려파기우월。
In order to resolve the ballistic target tracking problem in the reentry phase, we propose a modified strong tracking filtering algorithm, strong tracking finite-difference extended kalman filter (STFDEKF). The modified strong tracking algorithm make some modify: This algorithm use finite-difference method to approximate the derivative of the nonlinear function, and don't need to compute the derivative. Moreover, the new algorithm use sub-optimal fading factor to modify priori covariance matrix. The Monte Carlo simulation experiment compares the new algorithm with EKF and FDEKF. The result is the new algorithm is priori to the EKF and FDEKF in tracking accuracy and filtering credibility.