INTRO: An EU-funded research programme is developing cost-effective train location and signalling for secondary lines, using satellite positioning technology that may have future benefits in ERTMS applications
BYLINE: Pierre Mertens, Jean-Pierre Franckart
and Antonin Starck *
BYLINE: * Pierre Mertens is Locoprol Project Co-ordinator at Alstom Transport Information Systems, Jean-Pierre Franckart is Technical Director, and Antonin Starck is Locoprol Project Manager
BY THE END of this year trials are due to be completed with Locoprol low-cost train control technology for low-density routes. Most of the modules have been designed, and prototypes are under evaluation. The next step, scheduled for the spring of 2004, will be practical demonstrations on the Gembloux - Jemeppe line in southeast Belgium, which was selected in October 2002 as one of two test tracks; the other will be in France.
Around three years ago, it was recognised that modern signalling and train control technologies under development for high speed lines and major trunk routes were too powerful, and thus too expensive, for cost-effective application on lines with a relatively low level of traffic. Even the European Rail Traffic Management System was not well suited to such applications.
Started in mid-2001, the three-year Locoprol research and development project is being supported by the European Commission as part of the IST 5th Framework Programme (with extra funding by ESA as part of the complementary Locoloc project). The aim is to define and test an innovative low-cost, fail-safe satellite-based train positioning system, together with related signalling equipment offering a low life-cycle cost. The programme team is well balanced between industry, railways and research centres (Table I).
Whilst Locoprol is initially targeted at secondary and rural lines in Europe, we also see considerable potential on other continents. In terms of route length, the vast majority of the world’s railways are simple, often single-track lines with medium or low traffic density and long headways between trains. The lack of cost-effective modern technology means that many of these routes still have outdated safety systems, and/or manual operation.
The partners also anticipate that many individual Locoprol modules will have other applications; for example the train positioning technology may be applied to ERTMS on-board equipment.
Meeting user requirements
The first step was to determine what the likely user would expect low-cost technology to deliver. Some guidelines were drawn from the UIC specifications for a low-cost version of ETCS, now being studied by the ETCS-LC project team. Other requirements were established by sending a questionnaire to more than 50 selected railway organisations around the world.
The principal requirements included an improvement in safety over manually-worked mechanical or electro-mechanical signalling; better productivity, and a low life-cycle cost in terms of acquisition, operation and maintenance. Others included the need for interoperability of rolling stock, better control and supervision of level crossings, and scope for service enhancements such as real-time passenger information. However, most respondents flagged up that they were operating under very tight constraints, and investment funds for secondary lines were very limited.
The next step was to determine the best elements of existing technologies which might be incorporated to reduce the development cost and timescale. The main foundation for the signalling and train control element is ETCS with its attendant principles and specifications. The biggest innovation is the use of Global Navigation Satellite System (GNSS) technology.
This offers several advantages compared to existing technologies, in that the equipment is less exposed, and less succeptible to the local environment. In addition, regular calibration is not required. To ensure fail-safe train location at minimum cost, Locoprol uses a free combination of GPS, Glonass and the future Galileo network.
The project team has developed an algorithm to calculate the confidence interval based on the different satellite ranges. This is based on a pre-loaded unidimensional route profile stored on board the train. Calculations using one pair of satellites enable the train to define a segment of track, and establish its real position between the boundaries of this segment with a known probability. Simultaneous use of three pairs of satellites provides the required level of integrity.
Railway safety requirements specify a very high level of integrity for train location. The target probability for a wrong-side failure per hour is typically between 10-11 and 10-12. An intensive analysis confirmed that the proposed Locoprol technology would meet this requirement. Possible sources of errors were identified, and characterised in terms of safety risk, probability and extent. The final solution was developed to take account of all these potential errors. A high-end receiver has been selected, and was tested in Italy during November 2002. Validation trials in Belgium are to start in September 2003.
Satellite-based positioning has frequently been proposed in conjunction with track-mounted balises to determine relative train positions on busy lines, for example to meet ERTMS/ETCS standards. Conventional positioning systems can accumulate errors over time and/or distance, and thus require frequent resetting. This means fixed beacons must be installed at regular intervals, even in remote areas. Satellite-based absolute positioning allows the number of balises to be reduced substantially, or even eliminated completely, which would save the need for the associated on-board antennae when ERTMS compatibility is not required.
When passing through pointwork, where the positioning accuracy of the satellites might not be able to distinguish the separate routes, the on-train equipment interrogates the lineside equipment by radio and updates its location accordingly.
The use of satellite train positioning is expected to save up to 60% on the capital cost, compared to track circuits or axle-counters. Maintenance costs are also cut significantly, as virtually no equipment has to be installed under the vehicles.
Low-cost signalling
In developing the Locoprol signalling and train control system, most of the design effort was focused on life-cycle cost. This led to a decision to minimise the amount of on-track equipment. With fixed block indicators instead of signals, and no train detection devices, the only remaining lineside equipment is the point machines. The level of cabling is reduced by more than 80%. The cab signalling is based on standards specified for ETCS Level 2 or 3.
Trackside train detection and interlocking functions are replaced by Locoprol, using ’electronic tokens’, similar to the Radio Electronic Token Block pioneered by British Rail in the early 1980s. As shown in Fig 1, the track is divided into short sections, designated ’resources’, which broadly correspond to individual track circuit areas in conventional signalling installations. Each resource has a single ’token’, and the control system works on the principle of exclusive allocation of these resources to different trains or other users as required.
Before using a resource (track section, or point machine), a train has to be in possession of the related token, giving it exclusive permission to perform an action. An itinerary will be allocated by the control centre and updated periodically, at a rate appropriate to the traffic density or the complexity of the infrastructure. This is in effect a sequenced list of section identities.
At the appropriate time, the tokens are requested and allocated to the train, allowing the construction of a movement authority that defines the area in which the train can move without risk. This is passed to the driver via the cab signalling display, and also fed into the automatic train protection equipment. The principle of exclusive resource allocation ensures that another train will not try to enter the same section of line, and also guarantees that a train will never pass its authority limit.
Positive train detection requires a regular or continuous assessment of train integrity. This can be achieved either by having an integrity device on board the train, or by assessing train integrity at dedicated track locations, possibly by a formal procedure.
To permit the operation of an unequipped train or to detect vehicles stabled out of use, the signalman can allocate the token(s) corresponding to the location of stationary vehicles or the route set for an unequipped train through his ’office’ interface. A ’worker protection terminal’ will enable on-track staff to impose temporary speed limits or request a token to take complete possession of a section of track, thereby improving their safety.
Another user requirement was for simple procedures in the event of system degradation. Management of degraded situations can either be highly assisted by the signalling equipment, with a corresponding impact on the cost of the system, or by using manual intervention and relying on robust procedures. When the impact on traffic, and therefore on productivity, is limited, manual procedures would normally be used, but it is important to avoid the cost of having extra signalling staff at intermediate stations.
To meet the call for interoperability, Locoprol has been designed to meet ERTMS and ETCS standards. The ETCS standards are primarily concerned with the functionality of automatic train protection, and we believe Locoprol can be considered as a complement, applicable to a dedicated market. The system is highly compliant with the present standards, and will allow ERTMS/ETCS interoperability, where necessary. The additional cost is likely to be the need to assess train integrity.
One key difference is the communications technology. We recognised that the cost of installing and maintaining radio equipment is a major factor in design of any communications-based signalling system. Whilst ERTMS and ETCS require the use of GSM-R, Locoprol has been designed to allow users to make their own choice of communications medium, be it GSM-R, conventional radio or even mobile phone channels. Thus users can take into account any existing infrastructure they may have, and the availability of public services.
In addition to these main concepts, Locoprol offers the scope for introducing virtual blocks to enhance traffic on existing infrastructure. Level crossing controls can be configured to provide a constant warning time, optimising the interaction between road and rail traffic.
CAPTION: Fig 1. The Locoprol concept is designed to minimise the amount of lineside equipment whilst ensuring fail-safe operation
CAPTION: New signalling equipment being tested on SNCB’s Rivage - Gouvy line. Locoprol will be demonstrated on the Gembloux - Jemeppe route in 2004
TABLE: Table I. The Locoprol consortium
Alstom Belgium SA Belgium
Inrets France
Honeywell Regelsysteme GmbH Germany
Alstom Transport Spa Telecom Italy
Trasys SA Belgium
Alstom Transport SA France
Beratung und Planung im Verkehrswesen Germany
Septentrio NV Belgium
RFF/SNCF France
SNCB Belgium
Northern Jiaotong University China
Ertico Belgium
Une signalisation économique pour lignes à faible trafic
Un programme de recherche financé par l’Union européenne met au point un système de suivi des circulations et une signalisation à bas coût pour les lignes secondaires, compatible avec les spécifications de l’ERTMS. Le suivi des trains utilise la technologie de positionnement par satellite GNSS, et les autorisations de départ sont données sous la forme de bâtons-pilotes virtuels. La signalisation en cabine aide à la réduction des équipements le long des voies. Le consortium Locoprol de chemins de fer européens, fournisseurs et consultants, espère commencer les essais pratiques l’an prochain, sur la ligne Gembloux - Jemeppe, dans le sud-est de la Belgique
Kostengünstiges Sicherungssystem für schwach befahrene Strecken
Ein durch EU-Mittel unterstütztes Forschungsprogramm erarbeitet kostengünstige Systeme zur Zugsicherung und Standortbestimmung für Nebenstrecken, zu den ERTMS-Spezifikationen kompatibel sind. Die Bestimmung des Standorts des Zugs erfolgt über GNSS Satelliten-Positionsbestimmungs-Technologien, und die Fahrberechtigungen werden über virtuelle Erlaubniszeichen erteilt. Die Führerstandssignalisierung erlaubt es, ortsfeste Sicherungseinrichtungen zu minimieren. Das Locoprol-Konsortium, welches aus europäischen Bahnen, Lieferanten und Beratern besteht, hofft nächstes Jahr mit Praxistests auf der Strecke zwischen Gembloux und Jemeppe im Südosten von Belgien durchführen zu k
