Search

The Rail and Road transportation systems are evolving to progressively introduce automation with the ultimate goal of autonomous driving. The expectations are to improve efficiency and to guarantee higher level of safety. As transport networks will evolve towards high mobility, vehicles will become connected and self-positioned opening new challenges to comply with stringent safety requirements. Although vehicles are increasingly equipped with multiple advanced onboard positioning sensors, GNSS corrections allow to minimize the confidence error and to standardize the on board units. In the ERTMS (European Railways Train Management System) GNSS positioning must comply with the Tolerable Hazard Rate of 10E-9/hour at system level. Similar requirements are targeted by the connected cars when the management of vehicles is performed by a centralized authority as in the ERTMS system. Both ERTMS and Connected Car applications rely on the knowledge of the vehicle’s position with similar integrity levels. Considering also that rail and roads are frequently close to each other’s sharing the same operational environment, a multimodal augmentation network has been designed in the HELMET project to satisfy the new market needs. The aim of this paper is to assess the technical feasibility and the economic sustainability of this multi-modal augmentation network designed for the rail and road applications. The prerequisites are to reuse existing and future EGNSS infrastructures - without creating a direct dependence on themselves - to guarantee the interoperability and to allow a stepped deployment in line with the plans of the rail and road infrastructure managers. The paper includes the results of a recently concluded field test and the requirements of RFI – the Italian railways infrastructure manager who is managing a working group with rail industries to contributing to a standardizable and interoperable GNSS-based ERTMS system. Follow up activities to consolidate results for an operational system are on-going.

Satellite navigation plays a critical role in new railway signaling and train control systems. It will allow higher levels of automation at a reduce cost in installation and maintenance as compared to current infrastructure-based technologies. However, in order to be integrated in railway systems, GNSS must guarantee the stringent railway safety requirements. GNSS safety has been extensively quantified for the aviation community, however, in the railway environment the probability of local GNSS hazards like interferences is larger due to its operation in a non-restricted space. Railway GNSS receivers must be therefore equipped with mechanisms to protect the position determination from being corrupted by undesired interferences (jamming or spoofing) without being aware of it. In this paper and based on Commercial-Off-The-Shelf (COTS) hardware we implemented interference detection functions suitable for railway localization. We detail the algorithm design as well as the necessary calibration procedure to determine the nominal signal model. The performance of the interference detection is analyzed with an experimental setup on a commercial train during the measurement campaign organized for the EU project ERSAT-GGC. Figures about the detection capability and detected events along the selected line are finally provided.

Before deployment and commissioning of new positioning products/applications, manufacturers need to perform long and often expensive field trials in targeted operational environments in order to assess and certify future behaviour. The community trust on GNSS (Global Navigation Satellite System) is so high that more and more safety critical applications based on satellite navigation are under development. With the increase of criticality level of GNSS applications, a relevant means to predict the performances and reliability of future applications is needed. Thus in this paper we introduce a new architecture based on Machine Learning that combines classical observables for local hazard detection, with the outcome of advanced RAIM in order to determine whether a given point of a railway is suitable for a safe and reliable use of GNSS for train positioning.

The adoption of the GNSS for the European Rail Train Management System – ERTMS is one of the objectives included into the ERTMS evolution plan to improve its competitiveness on the market. To this aim, in 2015 the European Global Navigation Satellite System Agency (GSA) approved the “ERTMS on SATELLITE –Enabling Application Validation (ERSAT–EAV)”, project with the object to verify the suitability of GNSS to generate Virtual Balises as an alternative/complement means to the physical balises prescribed by the ERTMS. The architecture of ERSAT EAV – based on joint use of multiple constellations and public augmentation networks - would be a contribution to foster cost-effective ERTMS solutions, primarily for the local and regional lines in Europe for which the economical sustainability is a priority. The paper describes the hazard’s mitigation techniques affecting the on board GNSS receiver when operating in harsh environments, and the scenarios for a smooth introduction of GNSS into the ERTMS system. Field test results carried out on a Test Bed implemented on a 50km line in Sardinia are presented and discussed.