Photo: William Barton/iStock/Getty Images Plus/Getty Images
Of the hundreds of papers researchers presented at the Institute of Navigation’s annual ION GNSS+ conference, which took place virtually Sept. 21–25, the following five focused on advanced technologies in high-precision GNSS. Papers are available at www.ion.org/publications/browse.cfm.
Railway health with GPS + Galileo
Railway infrastructure and vehicle maintenance expenditures are estimated to cost more than €20 billion per year at the European level. This indicates the demand for developing a low-cost system capable of providing prognostic information about the health status of the railway at the points of the interaction between the vehicle and the infrastructure (wheelset, pantograph, rail and catenary). To achieve these capabilities, SIA (System for vehicle-infrastructure Interaction Assets health status monitoring) is being developed by a consortium from five different European countries. Within the SIA, events are captured by a network of sensors, which are time stamped and then accurately geo-referenced by the positioning sub-system of SIA. The positioning sub-system is based on European GNSS (EGNSS) positioning algorithms tailored for the railway environment and comprises onboard as well as back-office processing.
GNSS-based positioning in the railway environment is very challenging. Hence, Galileo with its advanced signal structure is utilized in SIA (in addition to GPS) to improve availability as well as accuracy.
The onboard positioning algorithm has been developed based on a novel GNSS-inertial measurement unit (IMU) hybridized approach. The new approach can overcome frequent measurement gaps within the GNSS observations and maintain the accuracy level required by the SIA. An overview of the back-office positioning in SIA complements the presentation of the onboard processing.
Citation. Moradi, Ramin, Zheng, Yuheng, Hutchinson, Michael, Roth, Michael, Jahan, Kanwal, Goya, Jon, Alvarado, Unai, “Positioning for Train-infrastructure Asset Health Status Monitoring within the SIA-project,” Proceedings of ION GNSS+ 2020, pp. 2948–2959. https://doi.org/10.33012/2020.17636
Snapshot positioning
Snapshot positioning — from a very brief interval of the received satellite signal — is becoming popular for various applications. This paper studies the feasibility of achieving real-time kinematic (RTK) positioning using snapshot data, a method termed Snapshot RTK (SRTK). A major difference of this positioning method is the generation of code and carrier-phase GNSS observables, a procedure the authors explain. To explore the feasibility of achieving RTK under different scenarios, the rate of integer ambiguity resolution (IAR) is assessed by using snapshot measurements generated with different integration times and signal bandwidths under zero-baseline configuration. Under these assumptions, the key factor that influences the RTK fix rate is the code measurement noise. Double difference code measurement errors are evaluated and plotted with the resulting IAR fix rates to find the relationship between them. The performance of using multi-constellation and multi-frequency signals is tested as well. The fix rate can reach 100% when multiple constellations are used. The achieved positioning accuracy is shown to be less than 5 mm in horizontal domain when IAR is achieved successfully.
Citation. Liu, Xiao, Ribot, Miguel Ángel, Gusi-Amigó, Adrià, Closas, Pau, Garcia, Adrià Rovira, Subirana, Jaume Sanz, “RTK Feasibility Analysis for GNSS Snapshot Positioning,” Proceedings of ION GNSS+ 2020, September 2020, pp. 2911–2921. https://doi.org/10.33012/2020.17768
Cooperative positioning
Advances in low-latency communications networks combined with the paradigm of Intelligent Transportation Systems (ITS) have opened opportunities to develop network-based collaborative positioning and navigation. Recent research has fostered the concept of networked GNSS receivers supporting the sharing of raw measurements with other receivers connected to the network. Such measurements (for instance, pseudorange and Doppler) can be processed through Differential GNSS techniques to retrieve inter-receiver distances that can be integrated to improve positioning performance.
This paper investigates an improved Bayesian estimation for a sensorless, tight integration of Differential GNSS-based collaborative measurements through a modified particle filter. A particle filter natively supports the non-Gaussian noise distribution characterizing GNSS-based inter-receiver distances, so the proposed particle filter was designed, implemented and optimized according to the architecture of a proprietary INS-free GNSS software receiver and tested with realistic RF signals, thus showing remarkable improvement in positioning accuracy.
Citation. Minetto, Alex, Gurrieri, Alessandro, Dovis, Fabio, “DGNSS-based Cooperative Positioning using Statistics-Adaptive Particle Filter,” Proceedings of ION GNSS+ 2020, pp. 2652–2666. https://doi.org/10.33012/2020.17530
Virtual base station
RTK (Real Time Kinematic) is a positioning approach that provides centimeter level accuracy by using a reference station. When the rover and the base station are in proximity (short baseline), all common mode errors are eliminated by the double difference, allowing carrier phase ambiguity resolution. But in medium and long baseline cases, ionospheric and tropospheric delays are not completely eliminated, which affects positioning accuracy. This has limited the application of RTK, especially in certain regions where the closest base station is more than 50 km away.
Algorithms like RTK long baseline and VBS (virtual base station) have emerged as an alternative. The virtual base station (VBS) algorithm processes surrounding bases to generate a virtual one within a short distance of the moving rover. By doing so, atmospheric errors will continue to be eliminated in the double-difference model, and, presumably, RTK processing will be assured all across continents.
In this paper, a performance assessment of the algorithm is conducted under various conditions, including high ionospheric activity, high baseline, harsh multipath environments and, finally, in a long trajectory. The results show that the developed VBS algorithm ensures centimeter-level accuracy even under the harshest conditions.
Citation. Saidani, M., Sarri, P., Guinamard, A., Maya, D. Gallego, “Virtual Base Station Algorithm and Performance Assessment,” Proceedings of ION GNSS+ 2020, pp. 2696–2709. https://doi.org/10.33012/2020.17533
Open-world virtual reality
The Open-World Virtual Reality (OWVR) concept combines precise GNSS positioning and a smartphone-grade inertial sensor to provide globally-referenced centimeter-and-degree accurate tracking of a virtual reality headset. Unlike existing augmented and virtual reality systems, which perform camera-based inside-out headset tracking relative to a local reference frame (for instance, an ad-hoc frame fixed to a living room), OWVR’s globally referenced tracking enables a VR experience in which the user’s outdoor exploration is robust to extremes in lighting conditions and local visual texture. This paper introduces the OWVR concept and presents a prototype system with two candidate sensor-fusion architectures, one loosely and one tightly coupled. Comparative performance is evaluated in terms of tracking accuracy and availability of an integer-aperture-test-validated fixed tracking solution. For scenarios with degraded GNSS availability, which will be typical for outdoor VR, the tightly coupled architecture is shown to offer a critical tracking robustness advantage.
Citation. Humphreys, Todd E., Kor, Ronnie Xian Thong, Iannucci, Peter A., Yoder, James E., “Open-World Virtual Reality Headset Tracking,” Proceedings of ION GNSS+, pp. 2931–2947. https://doi.org/10.33012/2020.17635