一种基于神经网络的卡尔曼滤波改进方法

蒋恩松; 李孟超; 孙刘杰

doi:10.3724/SP.J.1146.2006.00188

一种基于神经网络的卡尔曼滤波改进方法

doi: 10.3724/SP.J.1146.2006.00188

计量
- 文章访问数: 3587
- HTML全文浏览量: 110
- PDF下载量: 2007
- 被引次数: 0
出版历程
- 收稿日期: 2006-02-23
- 修回日期: 2006-08-14
- 刊出日期: 2007-09-19

An Improved Method of Kalman Filter Based on Neural Network

摘要

摘要: 卡尔曼(Kalman)滤波是一种基于最小方差估计的递推式滤波方法，它要求信号的状态模型是已知的，这就限制了它在实际中的应用。利用神经网络的良好的非线性映射能力对实际系统进行系统辨识，可以获得符合精度要求的系统状态方程，很大程度改进了卡尔曼滤波的效果。相对于一些经典的卡尔曼滤波改进算法，这种方法具有应用范围广和数学建模简单易行的优点。将神经网络与卡尔曼滤波相结合的方法用于图像复原实验，结果表明，该方法具有可行性和有效性。
- 图像复原;卡尔曼滤波;神经网络
Abstract: Kalman filter is a recursive filtering method based on minimum variance estimation, but it assumes that the signals state model is exactly known, which restricts its application in practice. Nevertheless the systems state equation can be obtained through the identification of systems by using neural networks good abilities of non-linear mapping. In contrast to some classic improved algorithm of Kalman filtering, this method has the advantages of wide application range, simple and feasible mathematical modeling. In this paper, the method which integrates neural network and Kalman filter is implemented for image restoration. The experiment result shows that the provided method is effective and available.

HTML全文

参考文献(1)

Grewal M S and Andrews A P. Kalman Filtering: Theory and Practice[M]. New Jersey: Prentice Hall, 1993: 1-10.[2]Parkum J E, Poulsen N K, and Holst J. Recursive forgetting algorithms[J].International Journal of Control.1992, 55(1):109-128[3]Safak E. Adaptive modeling, identification, and control of dynamic structural systems, I: theory[J].Journal of Engineering Mechanics, ASCE.1989, 115(11):2386-2403[4]Koh C G and See L M. Identification and uncertainty estimation of structural parameters[J].Journal of Engineering Mechanics, ASCE.1994, 120(6):1219-1237[5]Lin J W, Betti R, and Longman R W, et al.. On-line identification of non-linear structural systems using a variable trace approach[J].Earthquake Engineering and Structural Dynamics.2001, 30(9):1279-1303[6]徐天河，杨元喜. 改进的Sage自适应滤波方法[J]. 测绘科学，2000, 25(3): 22-24. Xu Tian-he and Yang Yuan-xi. The improved method of sage adaptive filtering[J]. Science of Surveying and Mapping, 2000, 25(3): 22-24.[7]邓自立，王建国. 非线性系统自适应推广的Kalman滤波[J].自动化学报，1987, 13(5): 375-379. Deng Zi-li and Wang Jian-guo. Adaptive extended Kalman filtering for nonlinear systems[J]. Acta Automatica Sinica, 1987, 13(5): 375-379.[8]Ahmed N U and Radaideh S M. Modified extended Kalman filtering [J].IEEE Trans. on Automatic Control.1994, 39(6):1322-1326[9]周荻，胡振坤，胡恒章. 自适应推广Kalman滤波应用于导弹的被动制导问题[J]. 宇航学报，1997, 18(4): 31-36. Zhou Di, Hu Zhen-kun, and Hu Heng-zhang. Adaptive extended Kalman filtering for passive missile guidance problem[J]. Journal of Astronautics,1997, 18(4): 31-36.[10]顾成奎，王正欧. 非平稳环境下提高神经网络辨识能力的新方法[J]. 系统工程学报 2003, 18(4): 300-301. Gu Cheng-kui and Wang Zhe-nou. New method to improve identification ability of neural networks in non-stationary environment[J]. Journal of Systems Engineering, 2003, 18(4): 300-301.[11]于亦凡，周祥龙. 模糊图像的Kalman滤波恢复方法[J]. 青岛大学学报(工程技术版), 2001, 16(3): 67-69. Yu Yi-fan and Zhou Xiang-long. Image Restoration by Kalman Filter[J]. Journal of Qingdao University, 2001. 16(3): 67-69.[12]Parles Alexander C. An accelerated learning algorithm for multiplayer perception networks[J].IEEE Trans.on Neural Networks.1994, 5(3):493-495

施引文献

资源附件(0)

访问统计