摘要
针对粒子滤波存在的权值退化问题,从研究粒子滤波的建议分布函数出发,提出自适应渐消扩展Kal-man粒子滤波方法。该方法使用自适应渐消扩展Kalman滤波产生建议分布,可在线调节参数,从而使得系统具有更好的自适应性和鲁棒性。与用转移先验、扩展Kalman滤波产生建议分布的粒子滤波方法相比,自适应渐消扩展Kalman粒子滤波进一步提高了粒子滤波的精度。
In respect to the degradation of the weight existing in the particle filtering, from the point of view of the proposal distribution function, a new particle filtering method named the adaptive fading extended Kalman particle filtering is put forward. This method takes advantage of the adaptive fading extended Kalman filter to generate the proposal distribution function, and can adjust the parameter on line, which has better self adaptability and robustness. Compared with the common particle filtering methods whose proposal distribution function coming from the transition prior or the extended Kalman filtering, the adaptive fading extended Kalman particle filter has improved the accuracy of the particle filtering.
出处
《大地测量与地球动力学》
CSCD
北大核心
2010年第1期99-103,共5页
Journal of Geodesy and Geodynamics
基金
国家自然科学基金(40974009
40474007)
关键词
粒子滤波
自适应渐消滤波
遗忘因子
扩展Kalman滤波
组合导航
particle filtering
adaptive fading filtering
forgetting factor
extended Kalman filter
integrated navigation