基于VB-UKF的SINS/GPS自适应融合技术  

Adaptive fusion technology for SINS/GPS based on VB-UKF

作　　者：郝燕玲[1] 张召友 

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《华中科技大学学报（自然科学版）》2012年第1期54-57,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(60775001)

摘　　要：针对SINS/GPS组合导航中量测噪声统计特性不准确引起卡尔曼滤波精度下降的问题,提出基于变分贝叶斯自适应无迹卡尔曼滤波(VB-UKF)的非线性融合方法.分析了线性的变分贝叶斯自适应卡尔曼滤波(VB-KF)算法的原理与性能,针对其仅适用于线性系统的问题,将VB-KF与UKF结合导出了非线性的VB-UKF算法.该算法可对系统状态和时变的量测噪声方差进行同步非线性估计,且与传统的UKF算法具有统一的形式.导航仿真结果表明:VB-UKF对于突变或慢变的量测噪声方差均能实时跟踪,较常规UKF算法可有效降低噪声统计特性不准确给系统造成的不利影响,提高定位精度.As the performance in the Kalman filters will degrade when the precise noise distributions are not completely known in strapdown inertial navigation system （SINS）/global positioning system （GPS） integrated system, thus an adaptive data fusion method based on variational Bayesian unscented Kalman filter （VB-UKF） was presented. The performance of linear variational Bayesian Kalman filter （VB-KF） was analyzed. By considering that VB-KF was applied only to linear system, a combination of UKF and VB-KF was carried out forming a VB-UKF algorithm. The algorithm can estimate time-varying states and variances, and has a unified form with standard UKF. Simulation results and comparison analysis demonstrate that the algorithm can not only effectively track the suddenly or slowly changing measurement noise, but also can achieve higher accuracy than the normal UKF.

关 键 词：组合导航 变分贝叶斯 无迹滤波 自适应滤波 融和技术 

分 类 号：TN967.2[电子电信—信号与信息处理]