卡尔曼滤波器设计及应用精要.docxVIP

PAGE \* MERGEFORMAT12 卡尔曼滤波器设计及应用 摘要：卡尔曼滤波器 (Kalman filtering)是一种递归状态估计，即已知上一时刻的状态估计值以及当前状态的观测值就可以计算出当前状态的估计值，因此不需要记录观测或者估计的历史信息。它提供了一种高效可计算的方法来估计过程的状态，并且使的估计的最小均方差最小。它是一种利用线性系统状态方程，通过系统输入输出的观测数据，对系统的状态进行最优估计的算法。由于观测数据中包括系统中的噪声和干扰的影响，所以最优估计也可看作是滤波过程。通过对卡尔曼滤波器进行系统的论证和分析，建立以陀螺仪和加速度计测量数据为误差的数学模型,采用卡尔曼滤波的方法,实现数据融合,有效地提高了姿态检测系统的检测精度，并通过以K60单片机为核心的惯性测量单元进行试验验证。结果表明，该算法能够有效避免动态噪声对加速度??感器数据的影响，提高直立角度测量的精度。 关键词: 最优估计 数据融合 卡尔曼滤波 直立角度测量 The Kalman Filter Design and Applied Research Abstract: Kalman filter is a kind of recursive state estimation, which is known on the moment of the state estimation and the observation of the current state value can be calculated on the estimated value of the current state, so there is no need to record observations or estimation of historical information. It provides a highly efficient method of calculation to estimate the status of the process, and the estimation of the minimum variance minimum. It is a linear state equation of the system, through the data of system input and output, the optimal estimation algorithm of system state. Due to the impact of noise and interference in the system including the observational data, so the optimal estimation can also be seen as a filtering process. Through the demonstration and analysis system of Kalman filter, mathematical model is established based on gyro and accelerometer error, using the method of Kalman filtering, data fusion, effectively improve the detection accuracy of attitude detection system, and test the inertial measurement unit based on K60 microcontroller as the core. The results show that, the algorithm can effectively avoid the influence of dynamic noise of acceleration sensor, improve the measurement accuracy of upright posture. Key words: optimal estimation; data fusion; Kalman filter; upright attitude measurement 绪论 信号是传递和运输信息的时间或空间的函数，信号在传输和检测过程中不可避免的会受到外来干扰和内部噪声的影响，为获取所需要的信号，排除干扰，就要对信号进行滤波。滤波就是从测量到的或接受到的各种带有干扰的信号中取出有用的信号的方法或技术。 最小均方误差估计是一种小常用的比较简单的经典准则。对于平稳时间序列的最小均方误差的第一个明