I really don’t see the need to have a driver in a vehicle on rails.
In: 8
Responsibility mostly. We could build driverless trains, but laws don’t allow putting the life of so many people into the hands of an AI that could fail and derail the train. Humans can too, but laws haven’t caught up to the level of technology yet. (Though some countries change that slowly, and some AIs already drive trains with the driver just being there to oversee it)
Safety precaution. And convenience.
– You want someone in front who is able to swiftly make decisions on the possibility that the track control does something wrong. It seldom does, and if it does it’s not certain that it’s the train driver who is actually able to do something about it, that spots it in time.
– when the track control system goes down (because a server malfunctions, because a radio tower goes down, because a power feed for a network switch dies, because the radio equipment in the train engine dies. and so on) it’s going to create HORRIBLE delays for that specific train if you first need to get a driver to the train, before being able to instruct the train to manually override the (lack of or wrongful) instructions from the track control.
The driver is actually pretty cheap per hour, compared to what the train operator pays for track access per hour.
Safety precaution. And convenience.
– You want someone in front who is able to swiftly make decisions on the possibility that the track control does something wrong. It seldom does, and if it does it’s not certain that it’s the train driver who is actually able to do something about it, that spots it in time.
– when the track control system goes down (because a server malfunctions, because a radio tower goes down, because a power feed for a network switch dies, because the radio equipment in the train engine dies. and so on) it’s going to create HORRIBLE delays for that specific train if you first need to get a driver to the train, before being able to instruct the train to manually override the (lack of or wrongful) instructions from the track control.
The driver is actually pretty cheap per hour, compared to what the train operator pays for track access per hour.
Responsibility mostly. We could build driverless trains, but laws don’t allow putting the life of so many people into the hands of an AI that could fail and derail the train. Humans can too, but laws haven’t caught up to the level of technology yet. (Though some countries change that slowly, and some AIs already drive trains with the driver just being there to oversee it)
Safety precaution. And convenience.
– You want someone in front who is able to swiftly make decisions on the possibility that the track control does something wrong. It seldom does, and if it does it’s not certain that it’s the train driver who is actually able to do something about it, that spots it in time.
– when the track control system goes down (because a server malfunctions, because a radio tower goes down, because a power feed for a network switch dies, because the radio equipment in the train engine dies. and so on) it’s going to create HORRIBLE delays for that specific train if you first need to get a driver to the train, before being able to instruct the train to manually override the (lack of or wrongful) instructions from the track control.
The driver is actually pretty cheap per hour, compared to what the train operator pays for track access per hour.
Responsibility mostly. We could build driverless trains, but laws don’t allow putting the life of so many people into the hands of an AI that could fail and derail the train. Humans can too, but laws haven’t caught up to the level of technology yet. (Though some countries change that slowly, and some AIs already drive trains with the driver just being there to oversee it)
There is a wide array of redundant systems in our world. Generally, redundancy is a reliable way to increase safety by adding more checks on a system at critical moments.
For trains, it’s a lot cheaper to pay someone to monitor the trains’ performance than it would be to pay lawsuits from crashing trains.
There is a wide array of redundant systems in our world. Generally, redundancy is a reliable way to increase safety by adding more checks on a system at critical moments.
For trains, it’s a lot cheaper to pay someone to monitor the trains’ performance than it would be to pay lawsuits from crashing trains.
A number of light rail lines in the world are autonomous, for example the DLR in London.
The main issue that prevents the automation of heavy rail is the complexity of the railway system. The autonomous light rail systems mostly are simple two-way lines with no points or branch lines or interactions with non-autonomous vehicles.
You’d also need to install a significant amount of control equipment and remote sensing which would get ludicrously expensive for cross-country lines.
There is a wide array of redundant systems in our world. Generally, redundancy is a reliable way to increase safety by adding more checks on a system at critical moments.
For trains, it’s a lot cheaper to pay someone to monitor the trains’ performance than it would be to pay lawsuits from crashing trains.
Latest Answers