摘要
为了跟踪空间目标,构建了基于局部粒子滤波器的多传感器融合方法估计空间目标状态。粒子滤波重要采样过程中,设计了基于融合估计的重要密度函数减少粒子贫化问题,并设计基于Mc DE(Memetic compact Differential Evolution)重采样策略,通过对粒子的变异与选择等进化操作来解决粒子退化问题。理论推导与仿真结果皆证明方法的有效性。
A multi-sensor fusion method based on an improved particle filter（PF） was presented to estimate the states of space targets for tracking. The improved PF was designed by adopting a fusion estimation based proposal distribution and by using the memetic compact differential evolution（Mc DE） as re-sampling to generate new particle sets. A theorem was introduced and a simulation were studied to prove the validity of the proposed fusion scheme.
出处
《海军航空工程学院学报》
2016年第2期127-132,共6页
Journal of Naval Aeronautical and Astronautical University
基金
国家自然科学基金资助项目(91438117)
关键词
粒子滤波
多传感器融合
空间目标跟踪
particle filter
multi-sensor fusion
space target tracking
