[SystemSafety] underground trains without drivers

[Peter Bernard Ladkin]

On 1/24/13 12:20 PM, Dewi Daniels wrote:
>> So, the big question. If moving-block systems work so well automated with
>> trains, why wouldn't they be adaptable to organising vehicles in road
> traffic?
>
> I think the main obstacle to the deployment of such systems is not
> technological, but societal.
>
> On 4 November 2011, 34 vehicles were involved in a collision on the M5
> motorway, resulting in 7 fatalities and 51 injuries.  This incident received
> a lot of press coverage at the time.  A total of 1,901 people died in road
> accidents in Great Britain in 2011 [2].
>
> Can you imagine if the 34 vehicles involved in the collision had been
> automated?  There would have been a public outcry and demands for the system
> to be withdrawn from service.

General point taken, but concerning this specific example you are missing a hypothesis, I think.
If the vehicles had been automated *and a similar thing happened* then......

Some of the sensorics used in today's cars also works in typical low-visibility situations. With 
platooning algorithms on all vehicles, they would likely have reduced speed at similar rates.

Where it won't work is with obstacles sensed closer to the vehicles than necessary for smooth braking.

I have an article on this accident and similar (which I call Auffahr accidents, after the German 
word for it) at http://www.rvs.uni-bielefeld.de/publications/Papers/20111230CPSV2.pdf There was a 
similar Auffahr accident on a German autobahn not too far from Bielefeld, near Münster, about a week 
later. Low-vis there was caused by fog.

PBL

-- 
Prof. Peter Bernard Ladkin, Faculty of Technology, University of Bielefeld, 33594 Bielefeld, Germany
Tel+msg +49 (0)521 880 7319  www.rvs.uni-bielefeld.de