Article: ARTC's ATMS (in cab signalling)

[somebody]

http://www.railexpress.com.au/archive/2011/june/artcs-bold-plan-to-double-capacity-of-intrastate-freight-network

ARTC's bold plan to double capacity of intrastate freight network
by Rail Express — last modified May 31, 2011 03:59 PM
— filed under: Weekly Top Stories, Rail

Australian Rail Track Corporation's (ARTC) development of a unique, next-generation signalling system could revolutionise the way freight is moved throughout Australia.
   
 
ARTC's bold plan to double capacity of intrastate freight network

ATMS will provide significant capacity improvements on ARTC's 5 state, 10000km network

By Jennifer Perry

In what has been described by many in the industry as a "courageous step", ARTC is effectively creating its own signalling technology, the Advanced Train Management System (ATMS), using the "best of breed" of the various types of signalling and communications capabilities currently in operation and bringing these together in a new, integrated system.

ARTC is creating its own signalling system as at this point in time there is no signalling system in existence that could suit ARTC's operational needs.

Its intrastate network is vast and remote and signalling systems such as ETCS that are currently implemented across Europe and about to be rolled-out in Australia by Sydney passenger operator RailCorp are more suited to passenger metro networks.

ARTC's aim in developing the ATMS  is to be able to get more freight on the same amount of infrastructure in a cost effective way without any compromise to safety.

ARTC chief executive John Fullerton told Rail Express that ATMS would increase capacity of the network by reducing the spacing between trains and therefore allowing more trains to be run on the same track.

"Should the proven ATMS be implemented on ARTC's main line network, capacity will be substantially increased by enabling more trains to operate on the existing rail infrastructure. This has the potential to more than double the capacity of the network using existing rail corridors," Fullerton said.

The ATMS would enable the removal of physical "on track" constraints.Currently, ARTC's intrastate network operates using trackside signals for trains, many of which can be 20km to 50 km apart, with only one train within a signal area at any one time.

The proven ATMS system would replace on track signals and manage trains using a sophisticated computer program and a GPS-based train tracking and control device. The device has trackside and GPS transmitters on the front and rear of trains, up to 2km long, which send data about a train's exact location through the 3G network and satellites.

The system could secure the distance between trains with an added safety margin for braking and protection from trains entering into another train's space.

The new system is also able to verify and cross verify a train's position, speed and braking distance and being able to check these parameters against any other trains operating on the network.

ARTC's current signalling system gives them about 40% occupancy, with ATMS expected to increase this to around 65%.

ARTC is currently trialling a prototype of the ATMS on 105km of the interstate rail network between Crystal Brook and Port Augusta in South Australia as part of a 39-month proof of concept phase.

"The proof of concept trial commenced in June 2008, which gives an idea of the time we are investing in fully adapting the system to the Australian environment," Fullerton said.

ARTC has already completed about two-thirds of the trial and is currently nearing the end of the final stage.

ARTC will release the first stage findings of the ATMS trial in the coming weeks.

[Zoiks]

The ATMS seems to be a borderline Electronic Block/Moving Block system, probably closer to the former.

What makes it so different to the other systems out there?

[somebody]

I think all other CTC systems depend on detecting trains on the tracks by track circuiting, or axle counting.  This one depends on the train reporting in its location.  It's sort of an electronic, automatic version of train orders AIUI.

[Zoiks]

So is it superior to ETCS?
Wouldnt it be better to adopt the ETCS standard nationwide?

[somebody]

ETCS level 1 requires track side infrastructure.  IIRC, Train Orders apply from Kalgoorlie to Parkes (and Orange), so no track side infrastructure, and the ATMS is supposed to not require track side infrastructure either.

Think of the headaches which would happen when a track side radio beacon breaks down on the Nullabour!

[Zoiks]

So it is ETCS3 without trackside beacons instead relying on gps and no moving block?

[somebody]

So it is ETCS3 without trackside beacons instead relying on gps and no moving block?

I guess that's a fair description.  I'd be surprised if the ETCS3 would use GPS only.

[colinw]

Horses for courses.  What ARTC is doing is fantastic for long distance freight operations, not so great for urban rail.

You cannot eliminate balises even at ETCS level 3, as absolute position references are required as a diverse check for the operation of the GPS.  No diverse check = no SIL 4 (Safety Integrity Level 4) rating = no safety certificate.

ATMS seems to be geared more for main line freight purposes, and I suspect would be unsuitable for dense urban rail systems.

ETCS at the higher levels is the opposite.  Great for dense urban environments, but too darn expensive for long distance freight lines.

The French are playing with an ETCS derivative for low traffic branch lines that uses GPS to create "virtual balises" and runs with no track & trackside infrastructure except at the approach to stations.

[somebody]

I think the ATMS might be largely only suitable for single track railways.  I don't think GPS is accurate enough to distinguish between two parallel rail tracks at the required level of accuracy so alternate arrangements may apply at crossing loops.  Just my theory.  I'm fairly sure ARTC's solution would be unsuitable for urban areas.  Tunnels for a start are a bit of a problem for GPSes!

[colinw]

The usual solution for combining GPS with multi track is a system of "link transponders" placed on or beyond crossovers, so that the system can tell when it takes a diverging route.  Basically you position transponders so that after a turnout if you see transponder "A" you know you went one way, and if you see "B" you know you went the other way.

[somebody]

The usual solution for combining GPS with multi track is a system of "link transponders" placed on or beyond crossovers, so that the system can tell when it takes a diverging route.  Basically you position transponders so that after a turnout if you see transponder "A" you know you went one way, and if you see "B" you know you went the other way.

Fair enough.  Requires some track side infrastructure, but not near the level of track circuits, or even axle counters.

[colinw]

A passive transponder is relatively low cost.  It only when you have active transponders that interact with an interlocking to change the data content based on signal states that things get expensive.

I have seen research papers from the UK that are looking at using the RFID tags embedded in most concrete sleepers to provide continuous high accuracy position tracking. The RFID tags are in there for asset tracking during manufacturing & distribution, it just happens that they can be read by equipment mounted on a train even at quite high speeds. There have been tests in the UK at Network Rail's test track in Derby that shows that the RFID asset tags in some concrete sleepers can be read at speeds above 200 km/h.

Suffice it to say that some very interesting technologies in train position determination are on the horizon, and if you combine those with active reporting of train position you can squeeze a lot more capacity out of the railway.