Advance scout robot for trains

Submitted by brad on Wed, 2004-01-21 14:25
Topic: 
Transportation

Another transportion item, because last night the train I was on hit a car stalled on the tracks (the occupant is OK, though was hit by the car when the train bashed it.)

Since trains do hit things, why aren't solutions to this more common in our data network world? A laser detector over the grade crossings would be simple enough.

At dinner, my friend Kurth Reynolds made a suggestion that I have improved. How about a small robot, equipped with camera and other sensors, which travels far enough in front of the train that if it sees a problem on the track, can send a signal back to the train in time to stop it. Trains take a while to stop, which is one of the reasons they can't do anything when they see a car or person ahead on the tracks.

You can't be too far ahead or you enter the "space" of the earlier train on the track, though during any tight conflicts you can of course give up this "foresight" and bear through (or slow down.)

If you have a human driving the train, you can show them video of what's ahead of the robot and give them time for a decision. Some decisions (Robot hits something or derails) would be automatic. Of course the robot might hit the car stalled on the tracks (though it can stop much, much faster than a train) but do far less damage.

The robot would be tall enough to go over the suicides who are "sleeping" on the track, but light enough so a car hit by it would survive.

Simpler for shorter runs like commuter trains would just be cameras along the track beaming to the oncoming trains. The engineer could be seeing a mile ahead at all times. Hey, if x10 can sell 2 broadcasting video cameras for $80 (WARNING: Don't buy from their web site, you will be spammed to death) I bet this can be made affordable.

This is important because some people don't think we should have rail with grade crossings. Without grade crossings, rail becomes vastly more expensive.

Some updates five years later: Some have worried the robot could hit workers or cars. Today, we are more comfortable we can build robots which would use LIDAR and never hit anything that wasn't running onto the track. The robots would also be light and perhaps have airbags to soften the blow against something rushing onto the track. When coming to a grade crossing, the robot would actually stop at the crossing and wait for the guard to come down (for the train, if the path is clear) and continue to monitor the crossing and report if something stops in it. Then it would speed up again and start going down the track to assure it is far enough ahead of the train.

Share on:
  • Reddit logo
  • Facebook logo
  • Twitter logo
  • LinkedIn logo

Comments

Paul Baclace

Thu, 2004-02-12 07:35

Moreover, a swarm of robot rail escorts would be a good way to prevent collisions with anything, not just stalled cars. This would be especially good for avoiding high speed train collisions which have the most fatalities. It would also be effective when a signaling system was faulty or an operator asleep or the track was damaged. I'm thinking that the swarm would communicate directly with the train it was escorting to detect objects in the path. A very simple logic could be employed--if the swarm of escorts are in contact with the train, integrate over their distances to determine the window of safety. If some percentage of escorts cannot be contacted, reduce speed, assess situation.

Anton Sherwood

Fri, 2005-12-23 16:15

Are you proposing a robot with legs that would step over "sleepers"?

Kurth

Thu, 2006-01-26 11:02

An additional optimization would be for such a robot to maintain a variable distance depending on the speed and stopping distance of the train. That way it can stay near by when the train is at stations so as not to trip the automated crossing guards, but stay well ahead when the train rarely tips 80mph.

I'd also like to give credit to Keith Henson for the original idea, putting a giant sticky airbag/cow-catcher on the front of the train. We then discussed how long it would take thrill seekers to start jumping in front of trains for fun.

Keith was a founding member of L5, and now on the run from Scientology persecution:
http://en.wikipedia.org/wiki/Keith_Henson

A Fisher

Wed, 2009-12-02 02:32

Two words, Totally impractical,

Firstly it would be far far too distracting for the driver, the drivers first priority is to read the line for the signals ahead, could you imagine driving your car and then also having a screen which shows you the road 1/2 mile ahead?

for a train in the UK doing 125 mph, the "robot" would have to, braking distance + thinking time distance ahead of the train, which would put it two miles away.

Check out the HSE Report into the fatal car strike of a train in the UK,

An extract:

9. Ufton Level Crossing is a standard AHB crossing. This type of crossing
is protected by road traffic light signals and a lifting half-barrier on both sides
of the railway. The crossing equipment is initiated automatically by an
approaching train at the “strike in point” (1743 metres before the crossing),
and the lowering of the barriers is preceded by the display of road traffic light
signals. When lowered, the barriers only extend across the entrances to the
crossing, leaving the exits clear. The strike in point for Down trains
approaching Ufton crossing is correctly positioned to give a minimum warning
time for road users of 27 seconds for trains running at line speed.

(Actual time should be 30 seconds +/- 10%)
so as 1 743 meter = 1.083 049 988 1 miles a train travelling at 125 mph needs 1 mile and 1/4 to stop (from brake application, not including thinking time/distance)

the train has passed the point at which it can stop, BEFORE it even activates the level crossing.

Signalling, most signalling relays on track circuits which is a current running through the rails, the "Robots running on the rails" would cinflict with the signalling and activate it and short it.

The whole idea is just impractical.

Anonymous

Wed, 2009-12-02 02:41

What about track workers??

those poor guys out on the track working in between trains, are they going to have dodge these robots shooting around at over 100mph as well?? at best they'd take someones legs off, then these he open footpath crossings where the public can walk across a railway line along a footpath, how are the to hear/see a robot train coming at them at night doing 70pmh?

Then there's the crossing accidents its supposed to prevent, given that the major problem is people swerving around half barrier level crossings they aren;t on the track or possing a risk when the robot passes, but they swerve around the barriers and get hit by a train, they are in between the trai nand robot and never have been detected.

Mention was made of the robots being light enough to not kill drunks asleep on the tracjk, how is a vehicle that light to stay on a canted track at 125mph??

brad

Wed, 2009-12-02 09:53

Don't have too many grade crossings, do they?

The robot would be light and able to stop quickly. It might even be able to squeeze the rails to stop faster than friction can stop it, so that it did not hit track workers -- who should not be working on live track in any event. It could also have airbags on it so that if it hit you, it did not kill you, but I agree, this is still an issue.

Anonymous

Wed, 2009-12-02 23:47

There are manyt level crossing in the UK where line speed is 100mph plus,

http://www.dailymail.co.uk/news/article-1184437/Jail-cabbie-drove-terrified-children-125mph-rail-crossing-barriers-closed.html

http://news.bbc.co.uk/1/hi/scotland/tayside_and_central/8325933.stm
Transport police have launched an investigation after a level crossing barrier came down onto a bus travelling from Stirling to Bridge of Allan..................

http://www.bbc.co.uk/insideout/content/articles/2009/02/03/east_s15_w4_levelcrossings_video_feature.shtml
There were 15 accidental deaths on level crossings in 2008 according to Network Rail’s figures........................
Network Rail says that every year 2,000 people are reported to misuse level crossings with motorists ignoring warning lights or weaving round barriers.

In the above instance, a robot would not detect a car "weaving" around the barriers.

http://www.networkrail.co.uk/aspx/2292.aspx
There are over 7,600 Level Crossings both on public and private land that cut across the UK railway network....................

Public footpaths are a huge problem as well:
http://www.echo-news.co.uk/news/4436552.Plea_to_close_railway_crossing_after_woman_s_death/

A Robot would not stop this,
http://images.google.co.uk/imgres?imgurl=http://images.mirror.co.uk/upl/m4/feb2009/8/4/5B59A1C0-DE43-B0B1-49E9BC10615C407B.jpg&imgrefurl=http://www.mirror.co.uk/news/top-stories/2009/02/09/&usg=__R46-Ync-Ok6sP0vpBgfH0-23UsE=&h=300&w=450&sz=45&hl=en&start=3&um=1&itbs=1&tbnid=tG6UTZEmg-Ez5M:&tbnh=85&tbnw=127&prev=/images%3Fq%3Dlevel%2Bcrossing%2B125mph%26hl%3Den%26cr%3DcountryUK%257CcountryGB%26safe%3Dactive%26sa%3DN%26um%3D1

the car has drove onto the corssing at the last possible moment, also reading about the problem in America of several tracks and cars driving on to a crossing as soon as a first train has passed only to get hit by a second on the other, again it wouldn;t stop that.

RE: track workers,
http://news.bbc.co.uk/nolpda/ifs_news/hi/newsid_7695000/7695683.stm
A lot of inspection and light maintainance work is done under live conditions, track workers have to cross live lines to get to other places and on the densley operated networks in the UK there can be little chance of getting a "block" (Trains stopped) for inspection, having robots flying around the network is a fundamentally floored idea. It would import a greater risk to the railway than that to which it designed to prevent. It would also work in a minority of a very small number of incidents, yet require expense and impractical workings far beyond any savings or safety increase.

brad

Thu, 2009-12-03 09:35

But we are now at a level where robots on the tracks could, due to a quick ability to stop, never hit anybody. However, they would tell the train to start slowing down if they detected things on the track of course. To have workers working on the live line the workers would need to have a device to signal the robot not to worry about them, that they see it and know the train is coming.

Anonymous

Thu, 2009-12-03 13:05

Scout robots would be additional traffic on the line that would provide an additional point of failure and present traffic management problems. At present the UK rail network is underutilised because train stopping systems aren't effective and trains have to be spaced further apart when future need means they need to fit closer.

Wouldn't it be possible to lay a wire and measure the distortion in its electromagnetic field? I recall seeing a perimeter security system demonstrated on TV that used this principle. Satellites can be built with cameras that look at multiple small points. GPS location and remote car disabling? I'm sure some of those ideas are feasible.

brad

Fri, 2009-12-04 01:41

There are many possible ideas here. I blogged this one years ago but I presume it got some more links recently. The main concern is whether the detection solution, be it a robot or sensors, causes no problems and is cheaper than the cost of trains hitting things.

Anonymous

Fri, 2009-12-04 13:00

Didn't notice it was an old topic. That can happen when someone gives an old one a boost.

It's worth noting that bad decisions and penny pinching undermined the UK railway in the past. However, the need to economically rebalance the UK, deal with car traffic (and enabled crime), and climate change and oil use are putting these questions at the top of the agenda.

I've been one of many voices calling for a world class Shinkansen style rail network in the UK. It looks like the government is committed to realising that and it will go ahead even if there's a change of governing party. That may help explain new traffic with this this topic.

While I've argued that sci-fi is dead good engineering and customer service can help reshape the economy and culture, and a rail project like this is going to have some effect. Some of the change is obvious but it may also help reinvigorate sci-fi as people look forward and their expectations raise.

brad

Fri, 2009-12-04 13:08

Shinkansen style trains are a 50 year old improvement on a 150 year old technology. Hardly world class. Trains are only as efficient as their load factors. The UK does have trains with decent load factors, though not all of them, so special attention must be paid. And even trains with typical load factors are not as efficient as ultralight vehicles such as single person electric city vehicles.

Anonymous

Fri, 2009-12-04 17:06

First, the dropped a post because you couldn't be fucked to fix a broken link. Now you're trying to be a clever bastard. Are you going out of your way to piss me off? Go to fucking hell, Brad.

Overmod

Sun, 2009-12-06 15:09

Trackworkers will be covered under the same civil regs that govern train movements; you may assume that where organized trackwork is being done, there will be appropriate slow orders (and the robot vehicle would be running an appropriate distance ahead of its train at the restricted speed, sounding its audible warnings and running the lights and strobes per FRA regulations).

One may also presume that if the person (or AI or whatever) monitoring the sensor suite on the robot sees someone walking lineside, or has a 'worker transponder' signal indicating someone is near the track but may be invisible, the appropriate action(s) can be taken to reduce train speed, get the train crew ready for vigilance, etc.

One may also presume that the robot will be very low as well as light; tipover on any sensible superelevation would be extremely unlikely even in high crosswinds. If there is a problem with vertical stability a la Mercedes SLR of recent memory, there are technologies that can be used to enhance downforce without causing ballast-throwing or similar issues. This isn't a problem.

No, the robot won't help at all with the vast majority of common grade-crossing incidents. Its greatest use would be in detecting stalled vehicles or other 'hard' line obstructions (for instance, rockslides, broken rails, vandals' objects) with enough lead time for conventional braking stops. The issue for me is whether it is cost-effective to have All Those Robots for All Those Trains, with all the overhead and extra facilities and foreground attention 'that that implies' (to paraphrase Kipling).

I would argue that it makes little sense to have dedicated robots for this purpose when a more 'enabled' vehicle could serve many other purposes, notably a light crew, service, and repair vehicle. This would be large enough to have seats for crew changes and maintenance workers, have the ability to run effectively both on-road and off-road as well as be rail-capable, have external power and supplies for a wide range of common and uncommon failures on trains, and carry appropriate PPE and light transport (I think 'modified dirt bike' as one example) for people who have to walk up alongside stopped trains to fix issues like derailments.

Anonymous

Sun, 2009-12-06 16:37

Trackworkers will be covered under the same civil regs that govern train movements; you may assume that where organized trackwork is being done, there will be appropriate slow orders (and the robot vehicle would be running an appropriate distance ahead of its train at the restricted speed, sounding its audible warnings and running the lights and strobes per FRA regulations).

There are no "appropriate slow orders" in the UK,

Its a huge expense and import of more problms than it would ever be likely to solve,

I posted the initial idea on the works forum and it got many many replies of derision,

it's an impractical idea.

Overmod

Sun, 2009-12-06 17:41

Yes, my reply was primarily "US-American" centered. I think it's pretty clear from the other posts that usefulness of the 'robots' in Great Britain is not great enough to justify the various expenses. (My own preference is a combination of better sensing technologies on the locomotives and a distributed network of lineside cameras, proximity sensors, etc. that the locomotive PTC links to as needed.)

I presume that when you say "it's a huge expense... " you're referring to the use of robots, not to the practice of slow orders. What is the procedure in the UK when trackwork is being conducted on one of the lines and trains must pass? Surely there is something akin to a check that is used to restrict speed passing the area of work, or when passing over newly-laid track where the ballast has not settled in yet? I can't imagine 125mph trains running at unrestricted track speed past a work area (nor can I imagine taking the whole multiple-track path completely out of service while work is being done).

I was attempting to illustrate why some of the stated reasons for *technical* non-workability could be addressed with valid technical solutions. Not to support the overall principle in all cases and all regions.

Anonymous

Wed, 2009-12-02 23:58

Track workers on live lines adn the problems,

http://www.railwaypeople.com/rail-news-articles/a-near-miss-report-and-a-track-welder-is-killed-1334.html

Add new comment