毕业设计（论文）-基于嵌入系统的Kalman滤波器的设计.doc

PAGE PAGE \* MERGEFORMATI 本科毕业设计基于嵌入式系统的Kalman滤波器的设计XXxxx指导教师 XX 讲师学院名称电子工程学院 专业名称电子信息工程论文提交日期2016年5月11日 论文答辩日期2016年5月12日PAGE \* MERGEFORMATIV摘 要20 世纪 60 年代，卡尔曼滤波方法由 R.E.Kalman 提出，此方法克服了经典 维纳滤波算法的局限。卡尔曼滤波是最优化的自回归数据处理算法，其滤波标准 是均方误差最小，从而实现递推估计。在时域内，此算法选用状态空间方法来设 计滤波器，能对多维系统和非平稳系统的随机过程进行估计。卡尔曼滤波器因其 自身的优点：递归运算、计算简单、 自适应性、 前瞻性等，能够对随机干扰下的 线性动态系统进行最优估计，因而有着广泛的应用。 卡尔曼滤波已经被广泛应用于导航、 制导、通信工程、故障诊断、语音信号处理、工业过程控制等许多领域。但在实际工程应用中，例如语音增强系统中， 往往存在着滤波发散以及不能满足实时性要求等问题。为此，本论文在综合分析导致卡尔曼滤波发散的诸多原因及其抑制方法，阐述并比较了卡尔曼滤波器几种常见的实现方式，确定利用FPGA硬件来实现卡尔曼滤波器设计。对卡尔曼滤波算法的公式进行分解处理，利用MATLAB来建立模型对包含过程噪声的信号进行滤波实验。通过模型仿真，验证了卡尔曼滤波器的有效性。关键词：卡尔曼滤波器 系统仿真 FPGA Kalman filter design based on embedded systemXXX(College of Engineering, South China Agricultural University, Guangzhou 510642, China)Abstract: In the 1960 s, kalman filtering method by R.E.K alman proposed, this method overcomes the limitation of classical wiener filtering algorithm. Kalman filtering is the optimization of regression algorithm of data processing, the filter criteria is minimum mean square error, so as to realize recursive estimate. In time domain, the algorithm selects the state space method to design filter, for multidimensional system and non-stationary random process. Kalman filter because of its own advantages: recursive computation, simple calculation, and adaptability, prospective, the linear dynamic system under random disturbance can be the optimal estimation, and thus has a wide application. Kalman filter has been widely applied in navigation, guidance, communication engineering, fault diagnosis, speech signal processing, industrial process control and many other fields. But in the practical engineering application, such as speech enhancement systems, often there is a filter divergence and can not meet the requirement of the real-time problem. Therefore, in this paper comprehensive analysis of the cause of kalman filtering divergence and its inhibition method, expounds and compares the kalman filt