Search

Delivering high-accuracy GNSS using a low-grade receiver is seen as a holy grail to many positioning and navigation researchers and engineers. Proposed in this paper is a methodology that overcomes many challenges posed by the noisy measurements of an Android smartphone, targeting the decimeter milestone. Known as The Whole Works, the algorithm was developed by The Whole Shebang as a solution to the Google Decimeter Accuracy Competition. The Competition provided a set of data collected from a vehicle driven around the San Francisco Bay area over multiple days using three different smartphones. The team placed 4th in the contest, out of 810 international teams, using an approach that maximised benefit of the GNSS and IMU sensors available on the mobile. The approach adopted a UKF as the core processor, employing also Huber-like readjustments, reverse smoothing, static vehicle detection and a series of pre- and post-processing steps to ensure only quality measurements and solutions were delivered by the algorithm. All typical environments were considered in testing the solution, including motorways, residential and urban areas. Significant benefits are derived by an appropriate inclusion of each step, where decimeter accuracies are retrieved on the smartphone device.

GNSS is widely known to be vulnerable to an evolving range of threats (natural and man-made, unintentional and deliberate). This drives the demand for an alternative system to protect those critical applications that depend on GNSS from a safety, security and/or liability perspective. A potential alternative system is the R(anging)-mode concept, which is a positioning system that involves the transmission of time synchronised ranging signals using communications channels of existing maritime radio infrastructure.VDES is the planned successor to AIS, intended to address the current issue of over-crowding of AIS. It supports shipto-ship and ship-to-coast communications as well as a satellite component. R-mode has the potential to be applied to both the terrestrial and satellite components of VDES, which may be referred to as VDE-TER R-Mode and VDE-SAT R-Mode respectively. This study presents a system and service concept for a VDE-SAT R-Mode system capable of providing PNT services through a combination of space-based and terrestrial infrastructures, providing an analysis of the user needs, performance of the VDES ranging signal, scenario analysis of various constellation architectures and considerations to the ground and sensor segment.

FLAMINGO is developing a service to facilitate for high accuracy positioning in smartphones and IOT devices. Currently, smartphones can only achieve accuracies of several meters at best, and IOT systems desire low power, connectivity, and a highly accurate location. FLAMINGO will deliver a service that targets sub-meter accuracies in both client- and server-side positioning scenarios. Through this, a new suite of location-based applications will be enabled, that were before infeasible, as well as enhancing existing location products. Substantial server infrastructure is developed to remove user GNSS knowledge needs, where an Application Program Interface (API) is provided to simplify integration. Modified PPP and RTK processing procedures are used to provide the high accuracy positioning.

In this paper, a sub-meter real-time positioning service for smartphones is presented. Known as FLAMINGO, the service has been in operation since early 2020. Relying only on GNSS currently, FLAMINGO can deliver 50 cm or better accuracies to consumer devices such as smartphones or IOT. The Android smartphone is considered in this paper, harnessing the availability of Google Android Raw GNSS Measurements, which have been made available to Huawei, Samsung, Xiaomi, and other Android handsets since 2016. Since the release of these measurements, the GNSS community has made significant progress to improving the performance of GNSS in the smartphone. FLAMINGO harnesses the best-of techniques from PPP and RTK, modifying, and tweaking the configuration and pre-processing to improve the raw GNSS measurements from an Android device. These measurements are prone to high errors due to the poor quality of the mobile chipset receiver. Given these drawbacks, sub-meter positioning is achieved. This is demonstrated through static baseline tests, and suburban and urban kinematic trials. These tests have been repeated to ensure the user always experiences an accurate and reliable service. It is important to consider that all tests have been performed live, where a test app was developed to replicate the user experience of FLAMINGO.

In recent months, smartphones and IoT devices have become the leading products of mass-market GNSS. However, they are still limited in their positioning capability and, even with the release of Android raw measurements in late 2016, the use of PPP and RTK processing is still yet to be optimised. Smartphones are not ideal GNSS receivers, their measurements suffer from interference and multipath effects due to the multi-purpose nature of the antenna, shared with other systems such as Wi-Fi and Bluetooth. The Xiaomi Mi 8 represented a milestone within the GNSS mass-market, offering improved accuracies with the multi-frequency GNSS Broadcom BCM47755 chipset. The additional L5/E5 signals improve the resilience, but not immunity, to multipath effects in urban environments. This work is completed through the FLAMINGO initiative (Fulfilling enhanced Location Accuracy in the Mass-market through Initial GalileO services) which endeavours to facilitate mass-market devices to achieve sub-metre accuracies by using raw GNSS measurements and advances in GNSS hardware, systems and infrastructure. The PPP and RTK accuracies achievable in urban areas are evaluated through a field test within a varied semi-open sky and sub-urban environment. Objects have been identified with a FLAMINGO target accuracy, allowing evaluation of the positioning accuracy achievable in variious test scenarios. The results obtained highlight the sensitivity of the smartphone to multipath effects, especially in the sub-urban scenario. Furthermore, the use of L5/E5 observations increases the final accuracy, but the use of an ionosphere-free combination does not always provide the best results due to increased noise.

This paper presents an analysis into the computational results for modelling of a two degree-of-freedom nonlinear vibrating structure. Fast Fourier Transformations, Short Time Fourier Transformations, Hilbert Transformations and wavelets are used to model this system. These techniques aim to locate and quantify the nonlinear behaviour of the system. Coulomb friction was detected with a number of these techniques, however other nonlinearities could not be detected. From the analysis conducted, improved computational methods are necessary for the detection of nonlinearities.

This paper discusses FLAMINGO, an initiative that will provide a high accuracy positioning service to be used by mass market applications. The status and future for the initiative are discussed, the required accuracies and other location parameters are described, and the target applications are identified. Finally, the currently achieved accuracies from today’s Smartphones are assessed and presented. FLAMINGO (Fulfilling enhanced Location Accuracy in the Mass-market through Initial GalileO services), part funded through the European GNSS Agency, is a collaborative venture comprising NSL (as lead organization), Telespazio France, University of Nottingham, Rokubun, Thales Alenia Space France, VVA, BQ, ECLEXYS and Blue Dot Solutions. The initiative is developing the infrastructure, solutions and services to enable the use of accurate and precise GNSS within the mass-market, thereby operating predominantly in an urban environment. Whilst mass-market receivers are yet to achieve accuracies below one metre for standard positioning, the introduction of Android raw GNSS measurements and the Broadcom dual frequency chipset (BCM47755), has presented the devices such an opportunity. FLAMINGO will enable and demonstrate the future of high accuracy positioning and navigation information on mass-market devices such as smartphones and Internet of Things (IoT) devices by producing a service delivering accuracies of 50cm (at 95%) and better, employing multi-constellation, PPP and RTK mechanisms, power consumption optimisation techniques. Whereas the Galileo High Accuracy Service targets 10cm precision within professional markets, FLAMINGO targets 30-50cm precision in the mass-market consumer markets. By targeting accuracies of a few decimetres, a range of improved and new applications in diverse market sectors are introduced. These sectors include, but are not limited to, mapping and GIS, autonomous vehicles, AR environments, mobilelocation based gaming and people tracking. To obtain such high accuracies with mass market devices, FLAMINGO must overcome several challenges which are technical, operational and environmental. This includes the hardware capabilities of most mass-market devices, where components such as antennas and processors are prioritised for other purposes. We demonstrate that, despite these challenges, FLAMINGO has the potential to meet the accuracy required. Tests with the current Smartphones that provide access to multi-constellation raw measurements (the dual frequency Xiaomi Mi 8 and single frequency Samsung S8 and Huawei P10) demonstrate significant improvements to the PVT solution when processing using both RTK and PPP techniques.