Why don't we have more fully automated trains? Part I

John Hopton for redOrbit.com – @Johnfinitum

The cause of the tragic Amtrak derailment in Philadelphia on May 12 is still unknown. However, a recent press release from the National Transportation Safety Board (NTSB) stated that “An examination of the signals systems has revealed no anomalies or malfunctions,” and the possibility of human error is being investigated.

According to a breakdown of historical derailment causes from Vox, human error accounts for almost one third of incidents. Given that high figure, there seems to be a good argument for more automated systems and computerization.

And yet, not only in the U.S. but around the world (even in highly developed countries like Japan and South Korea), fully automated systems are largely limited to urban networks and smaller systems such as airports.

Staff from the UK’s National Railway Museum, the main rail museum in the country, gave us their perspective as to why train operation still relies so much an human control.

“In essence, I believe part of the argument is that people don’t like the idea of it,” says Museum Librarian Karen Baker. “The Underground (subway) trains in London are all controlled by computers but with a “driver” inside – to close the doors and respond to emergencies.”

We worry more about human absence than human error

Bob Gwynne, Associate Curator of Rail Vehicles, agreed: “I think people don’t like the idea. The only driverless trains I can think of in the UK are the Docklands light railway (a small urban rail system in London). In theory, any train with in-cab signalling system could be made ‘driverless’ but in practice most people on a (powerful, high speed train) would worry about being on a train doing 180 mph with no one ‘up front’. Hence the only driverless trains I know of are on metro systems which are relatively slow speed.”

He continues: “Incidentally, trains have had processors (a form of computer control) in them for years, but ‘cruise’ control’ – very uncommon – is a car concept and surely one that is increasingly out-moded with busy roads and better ‘distance sensors’. Train drivers have long had the capacity on some locomotives to get the loco’s controls ‘set’ and then just leave them there unless there is an adverse signal, the main difference being that (when railways) are not flat, the driver tends to need to keep adjusting the motor settings to keep the train moving at a set speed, unless this has been set.”

“Also there will always be a ‘dead man’s handle’ – more often actually a pedal that you put your foot on. These days, driver controls often have the brake/accelerator combined, and once that’s set the train does the rest. Because this can lead to the driver being distracted, there is a ‘vigilance’ button which is the ‘ping’ you hear if you are near the cab. The driver needs to cancel this every time it pings or the brakes come on.”

So while there is an element of computerization, it appears it is limited because, overall, passengers still have more faith in humans than technology.

In Part II, which we”ll run tomorrow, MIT researcher Hannah Groshong will discuss why driverless trains, and even more automated train controls, have struggled to gain traction in the US.

—–

Follow redOrbit on Twitter, Facebook, Google+, Instagram and Pinterest.

redOrbit.com
offers Science, Space, Technology, Health news, videos, images and
reference information. For the latest science news, space news,
technology news, health news visit redOrbit.com frequently. Learn
something new every day.”