The diversity of road applications and intelligent transportation systems (ITS) makes the definition of positioning integrity a real challenge because the requirements are changing from one application to another. Even within liability or safety critical applications, the role of positioning may vary if the application layer requires a position at specific location (e.g. emergency call) or a series of positions along a vehicle’s trajectory (e.g. transport of dangerous goods). -- This paper is focusing on the positioning assessment of vehicle trajectories collected by different navigation sensors (GNSS and redundant MEMS Inertial Measurement Units (R-IMU)) in real test scenarios. Single GNSS, integrated GNSS and R-IMU are compared to a high quality ground truth solution based on a high-end navigation system. The low cost equipment is based on a single frequency GNSS receiver combined with four IMUs (triad of accelerometers and gyroscopes) of the same type. The architecture and the algorithms for the sensors integration have been developed at EPFL and are used on several mobile platforms (land vehicles, ultra-light planes, micro-drones). A series of measurements of different kind and thus dynamics have been conducted by EPFL and NTUA during a scientific mission form the COST Action TU1302 on Satellite Positioning Performance Assessment for Road Transport (SaPPART). Several test scenarios were performed in different traffic and environmental conditions in order to face with challenging GNSS signal reception. Road sections have been selected ranging from open sky condition (e.g. rural roads) down to poor GNSS reception (e.g. urban road network). The evaluation of the positioning quality is done by comparing the position-output from several solutions: single GNSS, D-GNSS, integrated GNSS/R-IMU. The comparison to a reference trajectory of precisely time-stamped positions allows to calculate and to plot along-track as well as cross-track differences. This visualisation of the results make sense for many road applications like road user charging (RUC), pay as you drive and some advanced driver assistance systems (ADAS). Finally, this quality assessment of vehicle positioning in real conditions will be a valuable material for future simulations of navigation systems in severe conditions. This contribution is fully adequate to the goals of the COST Action SaPPART, especially for the definition of the performance assessment methodology.
[1]
Adrian Wägli,et al.
Trajectory determination and analysis in sports by satellite and inertial navigation
,
2009
.
[2]
J. Skaloud,et al.
Redundant MEMS-IMU integrated with GPS for performance assessment in sports
,
2008,
2008 IEEE/ION Position, Location and Navigation Symposium.
[3]
Nima Alam,et al.
Collaborative navigation with ground vehicles and personal navigators
,
2012,
2012 International Conference on Indoor Positioning and Indoor Navigation (IPIN).
[4]
Ling Yang,et al.
An enhanced MEMS-INS/GNSS integrated system with fault detection and exclusion capability for land vehicle navigation in urban areas
,
2013,
GPS Solutions.
[5]
Paul Zarchan,et al.
Fundamentals of Kalman Filtering: A Practical Approach
,
2001
.
[6]
Gérard Lachapelle,et al.
Relative Positioning for Vehicle-to-Vehicle Communications-Enabled Vehicle Safety Applications
,
2011
.
[7]
Yannick Stebler.
Modeling and Processing Approaches for Integrated Inertial Navigation
,
2013
.
[8]
Vassilis Gikas,et al.
Positioning Accuracy of Vehicle Trajectories for Road Applications
,
2015
.
[9]
Paul Bennett,et al.
Vehicle positioning for C-ITS in Australia (background document)
,
2013
.
[10]
Torgeir Vaa,et al.
Intelligent transport systems and effects on road traffic accidents: State of the art
,
2007
.
[11]
Matthew G. Karlaftis,et al.
ITS Solutions and Accident Risks: Prospective and Limitations
,
2008
.