| 基于KLD的蝙蝠算法优化自适应粒子滤波 |
| |
| 引用本文： | 滕飞,薛磊,李修和.基于KLD的蝙蝠算法优化自适应粒子滤波[J].控制与决策,2019,34(3):561-566. |
| |
| 作者姓名： | 滕飞 薛磊 李修和 |
| |
| 作者单位： | 国防科技大学电子对抗学院,合肥,230037;国防科技大学电子对抗学院,合肥,230037;国防科技大学电子对抗学院,合肥,230037 |
| |
| 基金项目： | 武器装备预研重点基金项目(9140A33020112JB39085). |
| |
| 摘 要： | 针对粒子滤波存在计算效率低和因粒子贫化导致的计算精度下降问题,基于KLD(Kullback Leibler distance)采样和蝙蝠算法,提出一种可动态调整粒子规模的自适应粒子滤波算法.首先,在重要性采样中利用KLD采样动态调整粒子规模;然后,使用蝙蝠算法定向优化粒子集,并在迭代更新中使蝙蝠算法和KLD采样相互作用,从而达到同时提升计算精度和计算效率的目的.实验结果验证了所提出算法的可行性和有效性.
|
| 关 键 词： | 粒子滤波 蝙蝠算法 KLD采样 粒子多样性 状态估计 非线性非高斯 |
Adaptive particle filter with bat optimization based on KLD sampling |
| |
| TENG Fei,XUE Lei and LI Xiu-he.Adaptive particle filter with bat optimization based on KLD sampling[J].Control and Decision,2019,34(3):561-566. |
| |
| Authors: | TENG Fei XUE Lei and LI Xiu-he |
| |
| Affiliation: | Electronic Countermeasures Institute,National University of Defense Technology,Hefei230037,China,Electronic Countermeasures Institute,National University of Defense Technology,Hefei230037,China and Electronic Countermeasures Institute,National University of Defense Technology,Hefei230037,China |
| |
| Abstract: | In order to solve the problem of particle impoverishment and low computational efficiency of a particle filter, an adaptive particle filter which can dynamic adjust the particle set size based on Kullback Leibler distance(KLD) is presented in this paper. Firstly, KLD sampling is used to dynamiclly adjust the particle set size in importance sampling. Then, the bat algorithm is used to optimaize the particle set purposefully, and it interacts with KLD sampling in iterative updates, so as to achieve the purpose of enhancing calculation accuracy and efficiency. The simulation results show the feasibility and effectiveness of the proposed algorithm. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《控制与决策》浏览原始摘要信息 |
|
点击此处可从《控制与决策》下载全文 |
|
