磁偶极子跟踪的渐进贝叶斯滤波方法

发布时间：2018-04-03 19:16

本文选题：磁偶极子　切入点：跟踪　出处：《自动化学报》2017年05期

【摘要】：提出一种新的非线性滤波器应用于磁偶极子目标跟踪问题.建立了跟踪问题的状态空间模型,在此基础上,从高斯矩近似误差的角度分析了现有卡尔曼滤波更新在磁偶极子跟踪中的问题.对此,将贝叶斯更新过程等效为求解连续时间上的渐进贝叶斯问题,在线性高斯条件下推导了其解析解,表明渐进贝叶斯更新可等效为定常系统的Kalman-Bucy滤波器;进一步采用一阶Taylor展开得到非线性近似解表达式,导出一种渐进贝叶斯滤波器,从理论上分析了与现有方法的异同.仿真与实测磁目标跟踪试验结果表明,渐进贝叶斯滤波器具有良好的精度和收敛性,能够有效抑制磁目标跟踪中由于大初始误差导致的性能下降和滤波发散,且计算效率与扩展卡尔曼滤波器相当,适于实际应用.
[Abstract]:A new nonlinear filter is proposed for target tracking of magnetic dipole.Based on the state space model of tracking problem, the current Kalman filter updating problem in magnetic dipole tracking is analyzed from the angle of Gao Si moment approximation error.In this paper, the Bayesian updating process is equivalent to solving the asymptotic Bayesian problem in continuous time, and the analytical solution is derived under the condition of linear Gao Si. It is shown that the asymptotic Bayesian update can be equivalent to the Kalman-Bucy filter of the time-invariant system.Furthermore, the first order Taylor expansion is used to obtain the expression of the nonlinear approximate solution, and a kind of asymptotic Bayesian filter is derived. The similarities and differences with the existing methods are analyzed theoretically.The experimental results of simulation and measurement show that the asymptotic Bayesian filter has good accuracy and convergence, and can effectively suppress the performance degradation and filtering divergence caused by large initial error in magnetic target tracking.The computational efficiency is equivalent to that of the extended Kalman filter, which is suitable for practical application.
【作者单位】：海军工程大学兵器工程系;
【基金】：国家自然科学基金(51509252)资助~~
【分类号】：TN713

【相似文献】