4 ms·

> Heck I see still drives on trains, I guess until we can end that position how do we expect cars to go it alone? (and train drivers are responsible for some ma

> Heck I see still drives on trains, I guess until we can end that position how do we expect cars to go it alone? (and train drivers are responsible for some major accidents both with people and cargo) I've always wondered why there hasn't been more progress on this front. Driverless trains have been around for 50 years now [0]. Anyone care to comment further? [0] https://en.wikipedia.org/wiki/Automatic_train_operation https://en.wikipedia.org/wiki/Automatic_train_operation

negativeview 12y ago
At least here in Minneapolis, a good portion of the train operates down the middle of public streets. It follows red lights and has to deal with people pulling into intersections waiting on a turn signal, stupid pedestrians, etc. Even WITH drivers two people (that I know of) have been hit in the last two months. The train line wasn't set up with driverless uses in mind in the first place, and the city is likely worried about liability and cost in switching over. Even if the number of accidents goes down with full automation, cities aren't likely to believe that until enough other places prove it for them.
- aragot 12y ago
  Sure, but wouldn't automation provide something more? If people know it's automatic, they're less enclined to assume the driver will stop; There is also less risk that the computer-driver is at fault; and last but not least, the video of the accident could be tweeted immediately so everyone is convinced that the biker/pedestrian/car is at fault. As opposed to human drivers, who are more reluctant to be filmed because it invades their privacy.
  - negativeview 12y ago
    All of those things are logical and make sense, but are unproven. Politicians are notoriously risk adverse when it comes to those kinds of risks.
- lucaspiller 12y ago
  My comment was more aimed separate railways - what you have sounds more like a tram. Here they are building a new tramline, and there has already been a collision with a car before it even opened :D
astazangasta 12y ago
Also, before we get to driverless cars, how about smarter traffic lights, that can say "ok, these guys are waiting for no reason. Instant green!" Smaller, easier problem, super helpful.
akgerber 12y ago
A driverless train needs to be entirely grade-separated, which generally restricts it to metros. Big cities generally have legacy metros where it would be difficult and expensive to update signaling to allow automated operation. And driverless metros generally still need a train operator to handle the doors and passenger safety, so the train/signaling upgrades are mostly worthwhile when the line is so busy that being able to run a couple more trains per hour pays off. And ATO requires a level of competence in operations and maintenance that not all organizations are capable of: https://en.wikipedia.org/wiki/June_22,_2009_Washington_Metro_train_collision https://en.wikipedia.org/wiki/June_22,_2009_Washington_Metro...
monort 12y ago
There are a lot more people per driver in the train, than in the car, so you can't expect to get a great saving on wages. For the same reason I don't see why ubiquitous driverless buses should come before driverless taxis.

negativeview 12y ago

At least here in Minneapolis, a good portion of the train operates down the middle of public streets. It follows red lights and has to deal with people pulling into intersections waiting on a turn signal, stupid pedestrians, etc. Even WITH drivers two people (that I know of) have been hit in the last two months. The train line wasn't set up with driverless uses in mind in the first place, and the city is likely worried about liability and cost in switching over. Even if the number of accidents goes down with full automation, cities aren't likely to believe that until enough other places prove it for them.

aragot 12y ago
Sure, but wouldn't automation provide something more? If people know it's automatic, they're less enclined to assume the driver will stop; There is also less risk that the computer-driver is at fault; and last but not least, the video of the accident could be tweeted immediately so everyone is convinced that the biker/pedestrian/car is at fault. As opposed to human drivers, who are more reluctant to be filmed because it invades their privacy.
- negativeview 12y ago
  All of those things are logical and make sense, but are unproven. Politicians are notoriously risk adverse when it comes to those kinds of risks.
lucaspiller 12y ago
My comment was more aimed separate railways - what you have sounds more like a tram. Here they are building a new tramline, and there has already been a collision with a car before it even opened :D

astazangasta 12y ago

Also, before we get to driverless cars, how about smarter traffic lights, that can say "ok, these guys are waiting for no reason. Instant green!" Smaller, easier problem, super helpful.

akgerber 12y ago

A driverless train needs to be entirely grade-separated, which generally restricts it to metros. Big cities generally have legacy metros where it would be difficult and expensive to update signaling to allow automated operation. And driverless metros generally still need a train operator to handle the doors and passenger safety, so the train/signaling upgrades are mostly worthwhile when the line is so busy that being able to run a couple more trains per hour pays off. And ATO requires a level of competence in operations and maintenance that not all organizations are capable of: https://en.wikipedia.org/wiki/June_22,_2009_Washington_Metro_train_collision https://en.wikipedia.org/wiki/June_22,_2009_Washington_Metro...

monort 12y ago

There are a lot more people per driver in the train, than in the car, so you can't expect to get a great saving on wages. For the same reason I don't see why ubiquitous driverless buses should come before driverless taxis.