Grade crossings, land slides, opposing traffic comprise much of the danger and safety of trains, cars, and people. I understand that BNSF and others are already considering flying drones for some track surveillance. Whistles are noisy and unpleasant (some of us kinda like them), gates are not entirely effective and often trouble prone. Landslides and track failures are a nightmare. Crews are wanting to be reduced to one person and even zero.
Has there been any discussion of a 'tracked drone', running maybe 90-120 seconds ahead of the train? It could not only ensure the track is OK, but monitor crossing and crossing signals, and in many cases simply block the crossing. I would see it as about 30 feet long, relatively light weight, hybrid to capitalize on fast acceleration and energy recovery from braking.
Crossing gates and signals are far to expensive in their current iterations, and I have long felt that modern electronics could do a lot better.
Railroads are all about reliability and simplicity. Drones are not simple or reliable.
Never too old to have a happy childhood!
A "drone," or in this case a remotely controlled speeder/track car, would simply hit the rock slide two minutes ahead of the train that was following.
As a rule, track cars have insulated wheels. By rule, they must stop at every crossing. I'm not sure how this is going to work if you have a train running 40-50 MPH behind it.
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
The answer is 'yes' it's been studied, and in a couple of contexts including having some of the crew in a vehicle like a glorified hi-rail truck controlling the whole rail consist by remote in "critical" areas. I will confess that the first time I saw a large-scale off-road RC car I thought 'put flanged wheels on that thing and you'll never cream stuff stuck on grade crossings again'.
There are some showstopping issues with the idea, though. First is the dramatically longer grade-crossing dead time, which reflects not only the dwell involved in initial activation to get the drone across but also the necessary 'block separation' (even with CBTC) to allow safe service braking should the drone detect anything. Second is the likelihood of mischief involving vandalism or theft of those expensive robot parts, especially if vehicles the size of the 'aerial' competition were to be used. More things will suggest themselves as you think about it.
One approach I was looking at involved a 'hybrid' device that could be either rail or airborne; it would be operated similar to a combination of old-time track walker and Sperry car up and down a given stretch of track including on desired headway to trains, shunted appropriately and able to read currents for signaling and to actuate grade crossing protection appropriately but be able to fly over actual signaled road crossings where necessary without needing to activate them. The current rules on what constitutes need for flag protection make this approach much more complex even if changed to allow the little robot vehicle to flash lights, plant markers or flares, etc. to act as the flagman, which seems like a reasonable extrapolation of the technology ... for otherwise slow movements kept prepared to stop, not long fast trains.
I continue with the idea that a combination of long-loiter UAVs and small devices communicating with the crew represents the best current use of the technologies.
A drone hitting a landslide or broken track obviously would trigger an alarm to the following train - be it one man or zero, and braking to a slower speed is (almost?) always better before an accident.
Res complications, all transportation systems have them already.
I think a better initial post on this would have been to list the problems.
Grade crossings are generally a problem and as trains go faster or get longer the problems increase. If they cannot be made fewer they need to be made safer (and quieter).
Grade crossing signals/gates are horribly expensive, complicated, and are known to fail. Modern technology likey can do better, anyone have links as to experts thoughts on this?
Whether it is a human brain or a computer driving the train, whatever is in the lead locomotive needs better information as to what is ahead. Is my guess of at least 60-90 seconds ahead about right? Cameras are getting so cheap they might be able to do much of this.
Cameras DO do this already, and there are accounts of pilot studies and tests in the FRA library.
The initial question has been how to handle the camera feeds cheaply and non-distractingly enough, with I think an important point: long-term deterrence of most 'dangerous activity' at crossings involves directed camera feeds to local law enforcement, not train crews themselves first, with the necessary empowerment to go after violators with actions that practically reduce 'recidivism' especially for multiple violators. The same is true in coordinating on-the-ground response to vehicles stuck on crossings long before a train crew first needs to decide the magnitude of a safe service stop...
Overmod - I have a small boat with a chart plotter, which does a really good job of showing what is ahead per charts. And with the radar on and split screen even what is actually ahead. But as my navigating class lady repeatedly pointed out, the most important two things when navigating are your eyes (and she would point at her two eyes).
She also repeatedly pointed out that driving a boat is not a video game. Glance at your screen, but only for a moment, your eyes belong on what is ahead. I did not find it difficult except on a few occasions. And at that point slow down, and figure out what is going on. Likely good advice when driving anything, or even just for living well in general.
Drone crosses road with no problem 90 seconds ahead of train. Impatient driver ignores crossbucks and flashing lights 0.1 seconds ahead of train. How does that help?
Our community is FREE to join. To participate you must either login or register for an account.