[PPL-devel] Variable and Polyhedron dimension

Enea Zaffanella zaffanella at cs.unipr.it
Sat Apr 5 07:08:47 CEST 2014


On 04/04/2014 05:02 PM, Stefan Schupp wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
>
> I am starting to use the PPL for a project dealing with state set
> representations for hybrid systems. I have a question regarding the
> dimension of polyhedra and variables. From what I can conclude, each
> variable can be assigned to a coordinate axis. Suppose I utilize 4
> variables with each assigned to a different dimension:
>
> Variable a(0);
> Variable b(1);
> Variable c(2);
> Variable d(3);
>
> When I construct a polytope with the first two variables, it has a
> dimension of 2, as intended. When creating the same polytope with
> other variables (e.g. c and d), I get a dimension > 2 (in this case
> 4). Is it intended to return the dimension of the variable assigned to
> the "latest/largest" coordinate axis, although the number of used
> variables/dimensions might be smaller than the returned value (the
> polytope still uses only 2 variables)?

Hello Stefan.

Variable are just "names" of space dimensions.
For a polyhedron or other PPL entity of space dimension k,
its set of space dimensions is { i \in N | i < k }.
So if you put Variable d (naming space dimension 3) in a 
constraint/generator,
you will obtain a constraint/generator having space dimension 4.
Namely,
    a + 3*b >= 0    is a constraint having space dimension 2
whereas
    c + 3*d >= 0    is a constraint having space dimension 4.
Also note that
    a + 0*d >= 0    would also have space dimension 4
(the fact that the coefficient for d is zero does not matter).

Hope this helps; if you need more details, take a look to the documentation.

Regads,
Enea.


> Best regards,
> Stefan
>
> - -- 
> Stefan Schupp M.Sc.
> RWTH Aachen University
> Computer Science Department, Informatik 2
> D-52056 Aachen, Germany
> http://www-i2.informatik.rwth-aachen.de/i2/schupp/
> Tel.: +49 241 80 21243
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG/MacGPG2 v2.0.22 (Darwin)
> Comment: GPGTools - https://gpgtools.org
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
>
> iQEcBAEBCgAGBQJTPslxAAoJEFPSi5GyofDjxdsH/jri+KfYwXgrh1XmG1kxkQX7
> pDU9i+V0u7zhttW0Wbsh/wWOOKtBc29JXcWBX3sy8AnYDhWFOyzcp4kJ10M+4wQA
> OmIKIt3X510YIpcxqzYbohdMzdZ/+3knyf06uzPdB3jQX1BBu9D7Y8AwdB/Mp/Tw
> Q5FaBmBcV4oXHA7COoYtnxuYyIEcmitUNAovz5/BOEWluM/j43pPd0ChDxBpY+YE
> A/K/aifiExYJNFFQpSz2G0+1j2ooFWeyM0OQzp1HVL61R1lhuIW8UXyLv8YJd/Kv
> H85d9ieYnYoEVCw7qNMhdg1jgFmt4NiNkL69Pwwxmau5TgZmbRu/j21W9mC7ymc=
> =i/vm
> -----END PGP SIGNATURE-----
> _______________________________________________
> PPL-devel mailing list
> PPL-devel at cs.unipr.it
> http://www.cs.unipr.it/mailman/listinfo/ppl-devel
>




More information about the PPL-devel mailing list