【摘要】 针对标准粒子滤波算法粒子退化和贫化问题,提出了一种基于高斯-牛顿迭代思想的容积卡尔曼粒子滤波算法.该算法利用当前量测信息,使用容积数值积分原则通过以一组确定的点集和相应的权值直接计算非线性随机函数的均值和方差,避免了求导运算,并通过Gauss-Newton迭代方法对容积卡尔曼滤波(CKF)的非线性最小二乘问题进行求解,减小了线性化误差,以此来产生粒子滤波算法的重要性密度函数,使得迭代CKF产生的重要性密度函数更接近于真实后验概率分布,从而改进了滤波性能.仿真结果表明,与粒子滤波和CPF滤波相比,迭代CKF粒子滤波具有更高的估计精度.更多还原

【Abstract】 In order to improve particle impoverishment and sample size degeneracy of the standard particle filter(PF) algorithm,a novel iterated cubature Kalman particle filter based on the GaussNew ton iteration method is proposed.The new algorithm directly uses the cubature rule-based numerical integration method to calculate the mean and covariance of the nonlinear random function by a set of the certain particles and their w eights.Meanw hile,the Gauss-New ton iteration method is used to solve the least square problem of the cubature Kalman filter(CKF),thereby reducing linearization error and generating the importance density function of the algorithm.The importance density function of the new algorithm that incorporates the latest observations is more close to the system state posterior probability density.So the filter performance is improved.Simulation results show that compared w ith the PF and the CPF,the iterated Cubature Kalman particle filter method has a higher estimation accuracy.更多还原