基于非线性滤波算法的磁偶极子跟踪

Magnetic Dipole Tracking Based on Nonlinear Filtering Algorithm

为了实现具有高度非线性特点的磁偶极子跟踪,将磁偶极子的位置、速度和磁矩等参数的估计归结为动态系统的贝叶斯估计问题,提出了使用递归方法估计其状态参数。在此基础上应用高斯混合采样粒子滤波(GMSPPF)算法实现了磁偶极子跟踪,并通过实测试验检验了算法的性能。结果表明,与粒子滤波(PF)和Sigma点粒子滤波(SPPF)算法相比,GMSPPF算法具有更好的性能和较低的计算量。

To realize the magnetic dipole tracking with high nonlinearity characteristic, the estimation of magnetic di- pole＇s position, magnetic moment, and velocity is formulated as a Bayesian estimation problem for dynamic systems. A recursive approach is proposed to evaluate the state parameter of the target. Based on the proposed method, the Gaus- sian-mixture sigma-point particle filter（GMSPPF） is adopted to realize the magnetic dipole tracking. The performance of the proposed method is verified through experiment. The results indicate that the proposed method can achieve higher tracking performance, and GMSPPF performs better in both estimation and computational efficiency than the particle filtering and sigma-point particle filtering algorithms.