粒子滤波在MEMS陀螺仪初始对准中的应用

摘要

针对微机电系统MEMS(Micro-Electro-Mechanical-System)陀螺仪的随机误差,引入了粒子滤波处理MEMS IMU的输出数据.借助于机动目标的Singer模型建立了系统状态方程,论文讨论了粒子滤波算法在MEMS IMU滤波处理的应用,详细描述了算法的推导过程.应用经典卡尔曼滤波和粒子滤波分别处理MEMS陀螺仪初始对准时输出的数据,滤波结果发现,两种算法滤波后随机误差得到有效减小,而粒子滤波有一定优势.%According to the random errors of the MEMS( Micro-Electro-Mechanical-Systems) gyro, particle filtering algorithm is introduced. By means of Singer model of Moving Target, the system state equations are presented. The application of the particle filtering algorithm in the output data of MEMS gyro is discussed and the algorithm is deduced in detail. In this paper, Kalman filtering and particle filtering are used in filtering to the initial alignment output data of MEMS IMU respectively. By using two kinds of filtering method, the random errors are reduced, and the particle filtering is better.