摘要
对采用余度配置的MEMS-IMU/GPS组合导航系统进行了研究。分析了微小型组合导航系统的特点和误差模型,针对惯性/GPS伪距组合导航模式下,卡尔曼滤波器需要对量测方程线性化的缺点,提出了基于改进平淡粒子滤波的滤波算法。该算法采用权值控制参数决定粒子是否进入平淡卡尔曼滤波器,有效降低了滤波计算量,并和UPF算法精度相当。研究表明,改进平淡粒子滤波算法对系统性能有明显提高,在GPS信号受到遮挡、暂时不可用的情况下,具有较好的抑制误差作用,适合余度微惯性/GPS组合导航系统的应用。
Redundant MEMS-IMU/GPS integrated navigation system is studied. Characteristics and the error model of the micro integrated navigation system are analyzed. Aimed at SINS/GPS pseudo range integrated navigation mode, the Kalman filter needs to make measurement equation linearization. The unscented particle filter algorithm for the redundant MEMS-IMU/GPS integrated navigation system is presented. This algorithm adopts the weight control parameter to decide whether the particle enters unscented Kalman filter, thus decreasing filtering computing amount and having the same precision of UPF. Simulation shows that unscented particle filter improves the system performance. When GPS signal is sheltered, the algorithm restrains the navigation error and is suitable for redundant MEMS-IMU/ GPS integrated navigation system.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2007年第5期570-575,共6页
Journal of Nanjing University of Aeronautics & Astronautics
基金
国家自然科学基金(60472125)资助项目
关键词
MEMS
余度配置
组合导航
改进平淡粒子滤波
平淡卡尔曼滤波
MEMS
redundant configuration
integrated navigation
improved unscented particle filter
unscented Kalman filter（UKF）
