Integrity Performance for Precise Positioning in Automotive

Laura Norman, Eduardo Infante, Lance de Groot

Abstract: Emerging safety critical applications in the automotive industry such as Advanced Driver Assistance Systems (ADAS) and Autonomous Driving (AD) have created a need for absolute localization with both high accuracy and integrity. Providing GNSS integrity for automotive use cases presents several challenges beyond the traditional techniques for integrity in GNSS previously developed by the aviation industry. First, automotive applications demand higher accuracy than aviation, down to lane level in some cases. This requires the use of carrier phase-based positioning methods, in contrast to the pseudorange positioning used in aviation. Secondly, the local environment for automotive applications is different than in aviation, particularly with regards to multipath. Lastly, automotive applications are more cost sensitive and will use mass market chipsets and antennas, which can result in increased measurement noise and poorer multipath rejection than hardware used in other GNSS applications. In this paper, we present results for a RAIM algorithm implemented with Hexagon Positioning Intelligence’s (PI) Precise Point Positioning (PPP) filter using TerraStar X corrections, fused with PI’s Inertial Navigation System (INS) technology. Data is collected in both static and kinematic conditions using a multi-frequency mass market chipset, a candidate automotive grade antenna, and an automotive grade IMU. Kinematic data includes open sky, highway, and suburban environments, representing the expected use cases for level 2 and 3 conditional automation systems. Results from the automotive grade system are compared against a post-processed reference trajectory using measurements from a NovAtel survey grade receiver and antenna paired with a tactical grade IMU to determine the position error to centimetre level. Protection levels derived from the RAIM algorithm are assessed both in terms of integrity against the actual position error and availability against various alert limits, through time series and Stanford diagrams. We show that protection levels reaching 1-2 metres can be achieved with automotive grade hardware in open sky environments, with a targeted integrity risk of 10-7 /h. In addition, fault injection is performed to test the robustness and performance of the algorithm against various types and magnitudes of faults. Transient, step, and ramp faults on both pseudorange and carrier phase measurements are introduced, and the response of the algorithm is assessed in terms of the ability of the protection levels to continue to bound the position error. The benefits of integration with an inertial sensor are discussed. Including inertial measurements improves availability by reducing the re-convergence time of the PPP filter following a GNSS outage such as a tunnel or overpass. This is an important consideration for automotive applications where such obstructions are expected and frequent in many use cases. In summary, we show that protection levels can be obtained from a GNSS-only PPP solution at a magnitude which permits absolute localization to the level of a road, with a high degree of integrity. We show that the solution is robust both under real conditions as well as injected faults. Finally, we show that inertial integration can improve the availability of a high integrity GNSS solution when it is subject to frequent GNSS outages.
Published in: Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019)
September 16 - 20, 2019
Hyatt Regency Miami
Miami, Florida
Pages: 1653 - 1663
Cite this article: Norman, Laura, Infante, Eduardo, de Groot, Lance, "Integrity Performance for Precise Positioning in Automotive," Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019), Miami, Florida, September 2019, pp. 1653-1663.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In