|
|
论文:2015,Vol:33,Issue(2):320-325 |
|
|
引用本文： |
|
|
高社生, 杨一, 高兵兵. 基于渐消滤波的神经网络优化算法及其应用[J]. 西北工业大学学报 |
|
|
Gao Shesheng, Yang Yi, Gao Bingbing. Neural Network Learning Algorithm Based on Fading Kalman Filtering and Its Application[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 基于渐消滤波的神经网络优化算法及其应用 |
|
| 高社生, 杨一, 高兵兵 |
|
| 西北工业大学自动化学院, 陕西西安 710072 |
| 摘要: |
| 针对BP神经网络在训练过程中易陷入局部极小值的问题,提出一种新的基于渐消滤波的神经网络学习算法。该算法采用渐消卡尔曼滤波对神经网络的权值进行更新,有效避免了梯度下降算法产生的局部极小问题;与卡尔曼滤波相比,在渐消滤波中充分考虑了最新量测值的影响,能更合理地利用新的有效信息,从而提高了学习算法的收敛速度。学习后的网络不仅具有普通神经网络的自主学习能力,而且具有渐消滤波的最优估计性能。将提出的神经网络算法应用于SINS/BDS组合导航系统进行仿真验证。结果表明,提出的算法在逼近精度方面优于BP算法和卡尔曼滤波算法,可以有效提高神经网络的泛化能力。 |
| 关键词:
神经网络
渐消滤波
学习算法
组合导航
|
|
| Neural Network Learning Algorithm Based on Fading Kalman Filtering and Its Application |
|
| Gao Shesheng, Yang Yi, Gao Bingbing |
|
| Department of Automatic Control, Northwester Polytechnical University, Xi'an 710072, China |
| Abstract: |
| Aiming at the local minimum problem in the training process of BP neural network, this paper proposes a novel neural network learning algorithm based on the fading Kalman filtering. This algorithm avoids the problem of local minimum value because it provides the fading Kalman filtering for weights optimizing in neural network training. Compared with Kalman filtering, the fading Kalman filtering gives full consideration to the effects of the newest measurements and uses the effective information more reasonably. Consequently, the convergence speed of the algorithm is improved. The network after training not only has the autonomous learning ability of general neural network, but also has optimal estimation performance of the fading Kalman filtering. The proposed algorithm is applied to SINS/BDS integrated navigation system. Simulation results and their analysis demonstrate preliminarily that the approximation accuracy of the proposed learning algorithm is better than those of BP algorithm and Kalman filtering algorithm. The proposed algorithm, we believe, can improve the generalization capability of neural network effectively. |
| Key words:
algorithms
back propagation algorithms
convergence of numerical methods
covariance matrix
efficiency
error analysis
errors
navigation
neural networks
fading factor
fading Kalman filtering
learning algorithm
integrated navigation
SINS/BDS(Strapdown Inertial Navigation System/BeiDou System)
|
|
| 收稿日期: 2014-09-18
修回日期:
|
| DOI: |
| 基金项目: 国家自然科学基金(61174193)资助 |
| 通讯作者:
Email： |
| 作者简介: 高社生(1956-),西北工业大学教授、博士生导师,主要从事导航、制导与控制的研究。
|
|
| 相关功能 |
|
|
|
|
| 作者相关文章 |
|
| 高社生 在本刊中的所有文章 |
| 杨一 在本刊中的所有文章 |
| 高兵兵 在本刊中的所有文章 |
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Aboelmagd N, Ahmed E, Mohamed B. GPS/INS Integration Utilizing Dynamic Neural Networks for Vehicular Navigation[J]. Information Fusion, 2011, 12(1):48-57
[2] Alsmadi M K, Noah S A. Back Propagation Algorithm:The Best Algorithm among the Multi-Layer Perceptron Algorithm[J]. International Journal of Computer Science and Network Security, 2009, 9(4):378-383
[3] Hamid M, Nitin K T. Cascading Artificial Neural Networks Optimized by Genetic Algorithm and Integrated with Global Navigation Satellite System to Offer Accurate Ubiquitous in Urban Environment[J]. Computers, Environment and Urban Systems, 2013, 37(1):35-44
[4] Azami H, Sanei S, Mohammadi K. Improved the Neural Network Training for Face Recognition Using Adaptive Learning Rate, Resilient Back Propagation and Conjugate Gradient Algorithm[J]. Journal of Computer Applications, 2011, 34(2):22-26
[5] Daniel J S. Training Radial Basis Neural Networks with the Extended Kalman Filter[J]. Neurocomputing, 2002, 48(1/2/3/4):455-475
[6] 马野,王孝通,戴耀.基于UKF的神经网络自适应全局信息融合方法[J].电子学报,2005,33(10):1914-1916 Ma Ye, Wang Xiaotong, Dai Yao. Study on Nonlinear Optimal Estimation for Neural Networks Data Fusion[J]. Acta Electronica Sinica, 2005, 33(10):1914-1916(in Chinese)
[7] Wen Y. Nonlinear System Identification Using Discrete-Time Recurrent Neural Networks with Stable Learning Algorithm[J]. Information Science, 2004, 158:131-147
[8] 高社生,薛丽,魏文辉.渐消自适应Unscented粒子滤波及其在组合导航中的应用[J].西北工业大学学报,2012,30(1):27-31 Gao Shesheng, Xue Li, Wei Wenhui. Fading Adaptive UPF(Unscented Particle Filtering) Algorithm and Its Application to Integrated Navigation[J]. Journal of Northwestern Polytechnical University, 2012, 30(1):27-31(in Chinese)
[9] 夏启军,孙优贤,周春晖.渐消卡尔曼滤波器的最佳自适应算法及其应用[J].自动化学报,1990, 16(3):210-216 Xia Qijun, Sun Youxian, Zhou Chunhui. An Optimal Adaptive Algorithm for Fading Kalman Filter and Its Application[J]. Acta Automatic Sinica, 1990, 16(3):210-216(in Chinese)
[10] Yang Y, Xu T. An Adaptive Kalman Filter Based on Sage Windowing Weights and Variance Component[J]. The Journal of Navigation, 2003, 56(2):231-240
[11] 杨一,高社生,吴佳鹏,等.模型预测前向神经网络算法及其在组合导航中的应用[J].中国惯性技术学报, 2014, 22(2):221-226 Yang Yi, Gao Shesheng, Wu Jiapeng, et al. Feed-Forward Neural Network Algorithm Based on Model Predictive Filtering and Its Application in Integrated Navigation[J]. Journal of Chinese Inertial Technology, 2014, 22(2):221-226(in Chinese) |
|
|
|
相关文献: |
|
|
1．高社生, 薛丽, 魏文辉.渐消自适应Unscented粒子滤波及其在组合导航中的应用[J]. 西北工业大学学报, 2012,30(1): 27-31 |
|
|
|
|
|
|
