A GNSS/INS-integrated system for an arbitrarily mounted land vehicle navigation device

被引:26
作者
Mu, Mengxue [1 ,3 ]
Zhao, Long [1 ,2 ,3 ]
机构
[1] Beihang Univ, Sch Automat Sci & Elect Engn, Beijing 100191, Peoples R China
[2] Beihang Univ, Sci & Technol Aircraft Control Lab, Beijing 100191, Peoples R China
[3] Beihang Univ, Digital Nav Ctr, Beijing 100191, Peoples R China
基金
中国国家自然科学基金; 国家重点研发计划; 北京市自然科学基金;
关键词
Low-cost MEMS IMU; Arbitrarily mounted device; Misalignment angle; Motion mode recognition; Non-holonomic constraints; MISALIGNMENT ESTIMATION; ALIGNMENT;
D O I
10.1007/s10291-019-0901-8
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
For the traditional implementation of inertial navigation system, aligning the inertial sensor axes with the vehicle body frame is a necessary process. While the development of micro-electromechanical system brings considerable cost and size advantages, the undesirable alignment process is still a challenge for widespread civil use of portable inertial devices. Aimed to avoid complicated manual mounting of an inertial device used in land vehicle navigation systems, an algorithm is proposed to automatically estimate the misalignment angles between the sensor platform and the vehicle body frame, which enables the device to be mounted in an arbitrary orientation. The foundation of this method is formed by two facts. The first is that the accelerometer can exactly estimate its own posture when it is stationary on the platform. The second is the yaw calculated from the horizontally aligned device has a constant error when the horizontal component of the angular velocity is zero. A robust motion mode recognition technique, which compares the statistical characteristics of the measurements with an empirical threshold, is applied to detect whether the vehicle is parking, turning, or moving straight. Validation experiments show that the error of the coarse estimation algorithm is within 2 degrees when the heading misalignment is less than 45 degrees. This guarantees that the arbitrarily mounted device achieves the equivalent performance as the well-aligned one, whenever the global navigation satellite system (GNSS) signal is available. In addition, the positioning error of the misaligned device during short GNSS signal blockage is within 7 m with the application of auxiliary velocity updates.
引用
收藏
页数:13
相关论文
共 25 条
[1]  
Abowd GD, 1999, LECT NOTES COMPUT SC, V1707, P304
[2]  
Ali A, 2013, I NAVIG SAT DIV INT, P1626
[3]   Context-Aided Sensor Fusion for Enhanced Urban Navigation [J].
David Marti, Enrique ;
Martin, David ;
Garcia, Jesus ;
de la Escalera, Arturo ;
Manuel Molina, Jose ;
Maria Armingol, Jose .
SENSORS, 2012, 12 (12) :16802-16837
[4]   The aiding of a low-cost strapdown inertial measurement unit using vehicle model constraints for land vehicle applications [J].
Dissanayake, G ;
Sukkarieh, S ;
Nebot, E ;
Durrant-Whyte, H .
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, 2001, 17 (05) :731-747
[5]   Nonlinear filtering methods application in INS alignment [J].
Dmitriyev, SP ;
Stepanov, OA ;
Shepel, SV .
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, 1997, 33 (01) :260-271
[6]  
Elhoushi M, 2014, IEEE POSITION LOCAT, P158, DOI 10.1109/PLANS.2014.6851370
[7]   Detection of posture and motion by accelerometry: a validation study in ambulatory monitoring [J].
Foerster, F ;
Smeja, M ;
Fahrenberg, J .
COMPUTERS IN HUMAN BEHAVIOR, 1999, 15 (05) :571-583
[8]   An integrated land vehicle navigation system based on context awareness [J].
Gao, Nan ;
Zhao, Long .
GPS SOLUTIONS, 2016, 20 (03) :509-524
[9]   Multi-sensor context-awareness in mobile devices and smart artifacts [J].
Gellersen, HW ;
Schmidt, A ;
Beigl, M .
MOBILE NETWORKS & APPLICATIONS, 2002, 7 (05) :341-351
[10]  
Groves PD, 2014, IEEE POSITION LOCAT, P773, DOI 10.1109/PLANS.2014.6851443