低成本MIMU/GPS组合导航系统设计与实验

高精度惯性传感器的昂贵价格在一定程度上限制了惯性导航系统(INS)的进一步推广应用,然而低成本微型惯性测量单元(MIMU)的出现,为INS的推广应用了带来了新的契机。文中提出了一种适合于工程应用的低成本MIMU/GPS组合导航系统的结构与方案,并给出了组合导航系统位置速度交替组合模式的数学模型。考虑到未知工作环境中各类噪声统计参数的未确定性易使经典的卡尔曼滤波器发散,因此文中采用了闭环自适应卡尔曼滤波器对子系统的数据进行融合处理;设计了低成本的组合导航系统原理样机,并进行了跑车实验。跑车实验及事后数据处理分析结果表明:该系统具有较高的长时间导航精度以及良好的可靠性和鲁棒性。

Design and Test of Low-Cost MIMU/GPS Integrated Navigation System

Aim. In order to bypass the limitation imposed by the high price of precision inertial sensor,we propose using low-cost MIMU(Micro Inertial Measurement Unit) to design low-cost MIMU/GPS integrated navigation system suitable for practical applications.In the full paper,we explain our design in detail;in this abstract,we just add some pertinent remarks to listing the three topics of explanation:(1) the structure and scheme of low-cost MIMU/GPS integrated navigation system,(2) the mathematical model of low-cost MIMU/GPS integrated navigation system,and(3) closed-loop adaptive Kalman filter;in topic 2,eqs.(1) through(6) in the full paper are taken from the open literature;in topic 3,we point out that,owing to the fact that statistical parameters of various noises are uncertain,traditional Kalman filter usually diverges;so a special adaptive Kalman filter must be designed;also in topic 3,eqs.(7) through(17) are all taken from the open literature but some of them are modified for design.Finally,a prototype of low-cost MIMU/GPS integrated navigation system was built.The data processing results of some car tests,given in Figs.5 and 6 in the full paper, show preliminarily the efficiency of our low-cost MIMU/GPS integrated navigation system;it provides a positioning accuracy of 10 m and velocity accuracy of 0.5 m/s when GPS works normally.When we purposefully locked GPS signals for 60 s, the data processing results of some car tests,given in Figs.7 and 8 in the full paper,worsened but the positioning accuracy of 20 m and velocity accuracy of 2 m/s are still within allowable limits;once the system captures GPS signals again,it restructures quickly and works normally under integrated navigation state.