[SystemSafety] Interesting new publication about safety for autonomous vehicles

Peter Bernard Ladkin ladkin at causalis.com
Thu Jul 11 13:41:11 CEST 2019

On 2019-07-11 11:47 , paul_e.bennett at topmail.co.uk wrote:
> Could we do software interlocking such that when implemented would test
> out as robust as if we had done it mechanically? It is a tall ask I think, but
> one where this community should aspire to be able to achieve. 
For rail, this is essentially what companies such as Siemens (which has the contract for German
rail, DB), Bombardier and others do.

The criterion to which they must do it is known in German as MGS. That is, at least the equivalent
level of safety to the previously-installed system.

But rail control is further along. There are fixed blocks, and then there are moving blocks.
Moving-block systems are in place for some low-speed lines (Victoria, Jubilee in London, I
understand, as well as the DLR). For normal-speed intercity rail, it is part of ERTMS Level 3. I
don't think there is a Level 3 installation yet. There are quite a few Level 2 stretches in Europe.

https://en.wikipedia.org/wiki/European_Train_Control_System has a lot of info, but I haven't checked
it for accuracy.

Rather closer to the environmental situation with autonomous road vehicles, the Uni Paderborn had a
project called Railcab, in which small uncoupled vehicles would use a rail network in a
non-scheduled manner, merging and separating under inter-cab negotiation. Info that I know about is
only in German, e.g., https://de.wikipedia.org/wiki/Neue_Bahntechnik_Paderborn . The project did not
succeed in demonstrating high-density mature reliability, as far as I know; the proof-of-concept
tests were quite limited (as is the test track in topography). But then, the amount of resources
being put into it was orders of magnitude less than what is going into autonomous road vehicles

The two safety mechanisms road has going for it that are not available for rail-guided
transportation is trajectory modification and very rapid braking. I have no idea if or how traj-mod
is used in the semi-automatic "driver assistance" systems available nowadays. (Relatively) rapid
braking is of course used.


Prof. Peter Bernard Ladkin, Bielefeld, Germany
Je suis Charlie
Tel+msg +49 (0)521 880 7319  www.rvs-bi.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20190711/013ba7dc/attachment.sig>

More information about the systemsafety mailing list