An Integrity Concept for GNSS/INS/Odometry Rail Localisation

In the scope of the CLUG2 project, a LOCalisation On-Board or ‘LOC-OB’ unit is being developed to provide along-track, track selectivity and start-of-mission positioning functions to the ERTMS (European Rail Traffic Management System). In this paper, we treat primarily the along-track positioning where the section of track is known but the absolute position along the track is unknown. The CLUG2 architecture employs a high-precision high-safety (10 -9 ) digital map along with a Dual-Frequency Multi-Constellation (DFMC) Satellite-Based Augmentation System (SBAS) Global Navigation Satellite System (GNSS) receiver, Inertial Measurement Unit(s) (IMU) and odometry sensors. An Extended Kalman Filter (EKF) is employed to correct the train’s front-end position and speed. An integrity risk allocation is then performed based on the assumptions described relating to the failure rates of each sensor. A monitoring scheme is defined in which robustly and conservatively accounts for all fault modes including local failures. Since DFMC SBAS is employed, satellite payload faults are handled by SBAS monitoring and bounding, whilst local faults, due to extreme multipath or Non-Line-of-Sight (NLOS) are addressed at the on-board level. A scheme combining innovations based FDE and divergence-free code-minus-carrier (CMC) statistics is proposed. Fault bounding is then used to characterise the monitors, through analysis of both analytical worst case faults in the case of the innovations monitors and the combination of heuristic arguments and numerical testing in the case of the CMC monitors. Protection levels are then obtained after the definition and use of new time-dependent slope parameters, a generalisation of the slope parameter used in RAIM and ARAIM algorithms. Finally, sensitivity analyses are performed in order to determine how performance, that is the size of the protection levels, varies as a function of the assumed error model and monitoring tuning parameters. Performance is found to be highly dependent upon the error temporal correlation, especially for values which might be expected when the train is at rest, whilst protection levels of the order of tens of meters are obtained for low temporal correlation when the vehicle is moving.