基于UKF的GPS／MIMU组合导航滤波算法研究  

Research on GPS/MIMU Integration Navigation Based on UKF

作　　者：李栋[1] 贾润[1] 刘剑锋[1] 

机构地区：[1]中国航天科工集团第三研究院8357所,天津300141

出　　处：《导航》2008年第1期18-21,共4页

摘　　要：高精度的导航定位是无人机研究中所面临的主要挑战之一。采用GPS／MIMU组合导航是目前主流的导航模式。当前实现GPS／MIMU组合导航的技术主要是卡尔曼滤波。文中将UKF（Unscented Kalman Filter）滤波技术应用于组合导航，建立系统的状态与观测量的非线性模型，在给定初始值和噪声方差阵的情况下进行了仿真。数值仿真表明所设计的卡尔曼滤波器可行，该方法能很好地改善滤波效果，提高了估计精度，同时减小了计算量。High precision navigation is one of the primary Challenges in Unmanned Aerial Vehicle （UAV） research. Nowadays, GPS/MIMU integrated navigation is the popular mode in UAV navigation. One of the most common methods of integrated navigation are known as the Kalman filter. The UKF （unscented Kalman filter） is applied to the integrated navigation. The status model and observation model of the system are built and simulation is finished with the given initial values and variance matrix of the noises. Simulation results show that the algorithm can quickly converge even for large initial attitude errors, and is feasible and efficient.

关 键 词：GPS/INS 组合导航 UKF 非线性滤波 

分 类 号：TN967.2[电子电信—信号与信息处理]