摘要
GPS导航系统噪声具有非先验性,而传统的卡尔曼滤波器要求假设动态模型和观测模型的噪声统计特性已知,因此传统卡尔曼滤波导航定位的方法定位精度不高。本文提出采用前向神经网络辅助调节卡尔曼滤波器,这样既考虑了现实环境的动态变化对系统模型造成的随机干扰影响,又融合了神经网络的自学习性和自适应性,使其具有自适应能力以应付动态环境的扰动,其中自适应利用神经网络的BP算法得以实现。仿真研究表明:该算法优于普通的卡尔曼滤波器。
Since GPS navigation system with uncertain noise and the conventional Kalman filter assumes that the statistical properties of the noise in dynamic model and observation system are exactly known, traditional methods which adopt Kalman filtering navigation orientation have low orientation precision. This paper presents a method, which uses the adaptive capability, which obtained by neural network-aided Kalman filtering scheme, to deal with the disturbance in dynamic situation. The method not only considers the random interference with the dynamic change, but also utilizes neural network＇s capability of self-taught and self-adapted. And the embodiment of adaptive capability is take advantage of the back-propagation algorithm of neural network. The simulation results show that the algorithm is better than that of traditional Kalman filtering.
出处
《电子测量技术》
2007年第6期1-3,48,共4页
Electronic Measurement Technology
基金
国家自然科学基础资助项目(49901013)
