Kompany Car
Well-Known Member
- Joined
- 19 Sep 2015
- Messages
- 3,612
Yes and yes and much better than a human driver.Drones to detect a cow round a bend? Differentiate between a track worker at the side of the track and one on the track?
Detecting a cow around a bend, a human driver couldn't detect a cow around a bend or a vehicle incursion. See Great Heck rail accident for reference. Unfortunately trains have a lot of kinetic energy when they are travelling at even only a few mph never mind 125mph. A Pendolino would take around 1 mile to stop from 125mph.
The paper that you referenced sets out the technical challenges and is a good peer reviewed piece. It makes the case for it to be technically feasible but potentially not cost effective if you retain a driver.Even if technically feasible, the vulnerability of automatic systems (to failure or malicious action) is an issue. See the failure of air traffic controls. Also components would be fail-safe, potentially stranding trains.
As regards malicious actions, I could (not that any sane person would want to) tamper with the train detection system, point detection circuits and/or SSI (solid state interlocking) and potentially cause havock on areas of the railway. The ability to do it remotely on more automated system is governed by the security of the communications backbone and ensuring that the system is running on an isolated and properly partitioned network.
You are correct in saying systems need to be fail safe which does, like the existing railway, mean bringing things to a halt. The challenge that cannot be overcome with technology or not in a sensible way is the interaction with passengers which is why you still need, not necessarily a driver, but someone to manage people when boarding and disembarking and in the event of needing to detrain passengers. Bottom line is that a large percentage of the population are for want of a better word, idiots. It takes very little time, particularly on hot days, for people stuck on a stationary train to start wanting to break the windows and climb out.
On DC systems fully automated driving makes a lot of sense from an energy saving point of view. AC trains can, albeit with losses, regenerate back onto the grid when braking, DC trains cant and need a receptive load i.e. when a train is braking, you need a train accelerating. Having it fully automated allows this and in deep tube it would mean you are not dumping as much waste heat into the environment which is exceptionally challenging to cool.
Could you make the whole transport network driverless, yes with the will and correct guiding minds. But we are talking about something that will take 20yrs or more and even then there will be some scenarios where you still require a driver.
As regards the opinion piece and obviously its my own opinion. Its very much playing to an audience, offering no thought around the challenges and how they could be overcome, clearly written by someone rather closed minded, which is odd for a person who calls themself a historian.