Slashdot Mirror
British Rail Moving Forward with Sat-Nav/GPS
de1orean writes "The BBC is reporting that after a successful limited trial using GPS satellite navigation to improve train safety and efficiency, British Rail is committed to instituting sat-nav throughout the system. It may be in operation as early as 2008."
It's a train. It's not like these things can wander all over the place.
Maybe if they were able to get them to run on time like they do in Japan and Fascist Italy, they could tell where the trains were by just looking at the clock.
Why do fools insist on going to high tech solutions when they can't even get the low tech stuff right?
I have no personal experience with British rails, but I have read about the numerous nasty accidents they have had recently.
I do have experience with San Francisco's BART and the Tokyo subways about the same time, mid 1970s.
BART had fancy computer controlled trains which sometimes left the station without the operator in the cab. They actually stopped correctly at the next station, usually, but sometimes the trains stopped past the station, or shot off the end of the rails for the last station, and sometimes they opened doors on the wrong side of the train, right over the third rail. They were having one heck of a time even running the trains as close as 5 minutes apart.
Meanwhile, Tokyo's Ginza line, built just after the 1923 earthquake I believe, a completely manual system, had been running trains every minute or two without problems for years. That line was so funky that car lights would go off for a second or two as they crossed junctions; you could watch this light blanking travel down the train towards you.
Why do these idiots insist on spending a fortune on high tech solutions when low tech solutions have been around for a hundred years and yet they can't get it right, even with examples around the world of making them work? Is it just empire building?
Infuriate left and right
They renationalised and didn't tell anyone?
sulli
RTFJ.
CSX rail in NY has been using sat nav to guide trains and to also NOT have a live person DRIVING the train!!!
So, they're implying the conductor is dumber than a GPS unit?
Is it that hard to follow the track?
This joke did the rounds during the first US-Iraq war and shortly after a big train accident..
What is the difference between British Rail and a Scud missile?
British Rail kills people.
Evil people are out to get you.
No. However the infrastructure across the UK is ancient and so neglected since Thatcher put the knife in that it pretty much needs completely replacing.
Rather than installing a *lot* of cabling, a wireless system would obviously be cheaper.
They are doing it because its cheaper.
Evil people are out to get you.
and they say canadians are dumb, at least we dont get lost following rail tracks
Sure high tech makes for better systems in the future. This system will make it possible to know exactly where trains are, they could have monitors at all sations, or web sites, showing expected arrival times down to the second, great stuff.
... they wasted so much money and trouble on their computerized train controls and forgot all the basics like double tracking to allow trains to pass and sidings to allow broken trains to get out of the way.
But when they can't even get basic block controls down right, and guarantee switches are in the right position, why waste time on this? It's like putting power windows in cars when you can't even keep the doors from falling off the hinges, or worrying about computerized anti-lock brakes back when they still had mechanical cable brakes.
Just like BART
That's what's wrong here. If they can't get basic safety standards in place, high tech GPS position reporting and timely web sites aren't worth beans.
Infuriate left and right
"British Rail" hasn't existed in over a decade...
...as the onboard unit would have to send data at the rate of hundreds of location datagrams per minute. The point being, a location is a point in time - may not be relevant 30 seconds later, travelling at 150km/hr. Trains move quite quickly, and given past British rail mishaps, existing systems must have to be sped up a bit.
:-(
With that said, GPS/GPRS units would have to communicate fairly frequently. At the very most a location sent to the server(probably over GPRS as a UDP datagram) every 2 minutes.
In New Zealand, the GSM provider here (business plan) charges per 10,000KB packet, even for a 500byte datagram
Some rough maths:
A location data packet(charged at 10k) every 1 minute.
Thats 0.6 MB per hour.
Train runs, say, 10 hours per day, thats 6mb.
Per month thats 180mb.
In New Zealand, thats about $200 of data.
In my town, a taxi company uses it. The combined cost per month is $33,000 in data charges.
And thats on 5 min updates!
Anyone got some info on charges from other countries?
IE how much will is cost our pommie friends per month per train, running 10 hr/days, sending location every 5,2,1 minutes, 30 seconds?
In post Patriot Act America, the library books scan you.
GPS is not so much to let the train driver know where he/she is (although that is a valid use under some control systems) as it is to let the train controllers and the safety interlocking systems know where the train is. This is obviously an important consideration before trying to place another train into that location ;-)
Historically there are rail systems around the world that do without this basic functionality, particularly in areas of low density traffic. Instead, procedures were established by which a section was proven clear. The signaller at the entrance to the section (or 'block') would ring a bell to let the signaller at the exit from the block know that a train was entering the block. The signaller at the exit would then ring a bell back to the signaller at the entrance when the train cleared the block section.
This worked to some extent (and indeed continues to work in many places around the world), but was highly susceptible to human error. A couple of years ago, at Glenbrook in Sydney's west, a crash may have been averted had the signaller had some indication to remind him that the Indian Pacific had not yet cleared the section ahead of the suburban passenger train. As it was, he forgot to inform the second train of the first's existence, with the result that the second rammed into the back of the first.
The most common method for establishing a train's locations is the track circuit - a power source is placed on one end of a section of track, whilst a receiver measures the voltage at the other end of the section. When the train is on the section, the axles of the train short out the circuit, and the receiver measures a 0V potential. This is fed back to the local interlocking, and any signals which require this track to be clear in order to give a proceed aspect will go to stop, preventing any other trains from entering the section and hence avoiding a crash. This is what the article refers to as the "current block signalling system".
The problem with track circuits is twofold. As mentioned in the article, the accuracy is not great - track circuits can often extend over several kilometres, so the best that can be said is that the train is "somewhere within that circuit". Secondly, the track circuits are relatively difficult to maintain - maintenance crews must go to each track circuit in order to perform routine maintenance. Travel time being relatively unproductive, rail operators would much prefer that maintenance be achievable in one location.
GPS would overcome both of the problems listed above. The accuracy would be greatly increased, limited only by the accuracy of the GPS. Similarly, the equipment for the location would all be located on the rollingstock, rather than trackside, and could be brought to the maintainers. Further, with systems like the European Rail Traffic Management System (ERTMS), GPS can be used to run trains much closer together in what is called a 'moving block configuration'.
This translates into lower running/maintenance costs, combined with greater efficiency - is it any wonder BR are interested? Aren't you?
In few days GPS-enabled trains could create a map of the entire rail system.
Train drivers will be happy... no more asking directions!
2) They're not "committed" to using it. "Key industry figures" (lobby groups?) had a meeting on Tuesday about whether to implement it system-wide. RTFA.
3) It won't be ready for service in 2008, that's when Galileo will be operational. RTFA.
The GPS one is a prime example. The door systems on the modern trains (the ones with sliding doors that don't have to be slammed shut and opened by reaching out of a window and fumbling for a lever) are GPS actuated. These doors will not allow passengers to open them unless the location of the train can automatically be established as being within a few metres of a normal station platform stopping point. The upshot? When it's cloudy or there is any kind of reception fault (as when we get back into London's Victoria station and we're under 20 ft of steel-reinforced concrete) the doors cannot be opened without the driver entering the positioning coordinates manually.
A driver was telling me that there is no 'look just open the bloody doors - I've got a key' button. So journey's all over the south coast are now delayed by really stupid door faults. Ironically the most reliable trains are the slam-door variety I mentioned earlier (which are eminently usable despite feeling like Stephenson's Rocket - unless you are in a wheelchair and then you can pretty much forget it).
...and, on the seventh day, God switched off his Mac.
They used this on the line I used to get to work, Hastings to Cannon St.
It was great in the summer, but not so great in the winter. When there were clouds. And no GPS reception.
We had to sit outside each station for 5 minutes while the driver got 'authorisation' to open the doors, becasue the GPS didn't know we were sitting beside the station.
Wonderfull bit of over engineering.
We've not had "British Rail" for some time. For you USians who may be a little confused, here's a little history.
Before the 50s, the railways were all in lots of groups of competing companies. The government then nationalised these companies (OMG SOCIALISM!!!1) as, well, you can't really run a railway for profit and even if you could it's a son of a bitch to do so without killing hundreds of people. This new confederacy was British Rail, and had that pointy double arrow logo some of you may have seen.
In the 60s, the government axed all of the loss making lines in an attempt to make a profit. This involved axing all of the lines to small rural communities, and it made the railway much less attractive to people and made the car a de facto requirement.
In the 80s, Margaret Thatcher grew weary of nationalised industry, and while sitting on her stylised throne of evil ordered that the railways be privatised, which took effect in 1997, just when the new government was coming in. Clever that.
So now we're stuck with a bunch of fucking idiots who can't run trains on time and have to rely on satellites to do so. As I once heard someone describe, it's fairly simple to run a decent train service; you have a train, you have a track, the former rides on the latter. The problem is that we have lots of people trying to make money off it, which just won't work.
By summer it was all gone...now shesmovedon. --
The article doesn't say, so it's reasonable for someone non-British to assume it's still called that, but it's probably Network Rail (formerly Railtrack) doing this. They are a now part-re-nationalised company that looks purely after the rails, stations and other non-profit making infrastructure. The private rail companies still own and (sometimes) run the trains and are doing very nicely thank you very much Mr Major (A stunning example of how privatisation actually works: Public funding, Private profit).
Rant over
should be both
'cause if a car falls off the end, who knows?
every day http://en.wikipedia.org/wiki/Special:Random
An inertial navigation system would be an obvious fix to the limited availability of GPS. When GPS fixes are available, they can be used to zero the drift on the inertial platform. What's the going price for a complete inertial navigation system these days? Today we have laser ring gyros and micro-machined accelerometers, which should cut the cost. If they can install them on commercial aircraft, they should be able to put one on a train.
Mea navis aericumbens anguillis abundat