基于Student's t分布的自适应重采样粒子滤波算法

针对粒子滤波在跟踪非线性状态突变系统的隐状态时,因粒子贫化导致估计精度下降的问题,提出一种基于Student''s t分布的自适应重采样粒子滤波算法.首先,将Student''s t分布作为采样尺度转移方程,再自适应地将粒子依据权值大小分为两个子集;然后,对子集执行自适应交叉和变异操作,得到新生粒子集,从而自适应地提升粒子多样性,达到提升估计精度的目的.实验结果验证了所提出算法的可行性和有效性.

Self-adaptive resampling particle filter based on student's t distribution

For the estimation accuracy problem that a particle filter used in hidden state tracking in nonlinear state mutation system suffers from particle impoverishment, a self-adaptive resampling particle filter based on student''s t distribution is proposed.Firstly, the algorithm employs the student''s t distribution as the transfer function of sampling scale. Then, the particle set is divided into two subsets according to the weight. Finally, self-adaptive crossover and mutation operations are performed on the two subsets to obtain the newborn sets. This algorithm can improve the estimation accuracy by self-adaptively improving the particle diversity. The simulation results show the feasibility and effectiveness of the proposed algorithm.