3
$\begingroup$

I started reading about timed automata lately. In the definition, they have locations instead of states as in automata theory. Can anyone explain why is this?

$\endgroup$
3
  • $\begingroup$ Do you mean Alur & Dill's timed automata? They call them states. If you are referring to some form of timed Petri net: Petri nets are said to contain places rather than states. This is because the system's state is distributed across the places. $\endgroup$ – reinierpost Oct 28 '15 at 17:11
  • $\begingroup$ @reinierpost Thank you for giving the original paper. Basically, I was reading it from some study materials from the web, e.g., fi.muni.cz/~xpelanek/IA158/TA-intro.pdf, lsv.ens-cachan.fr/Publis/PAPERS/PDF/BL-litron08.pdf $\endgroup$ – hola Oct 28 '15 at 18:07
  • $\begingroup$ It would help to edit your question to link to the specific references you are reading that use "locations" instead of "states": we want you to include all relevant information in the question, not just leave it in the comments. Also, if you now know enough (based on reinierpost's comment) to answer your own question, you might want to post an answer in the answer box below. $\endgroup$ – D.W. Oct 29 '15 at 23:49
5
$\begingroup$

It is mainly a matter of avoiding ambiguity.

At any moment in a Timed Automaton (TA), available transitions depend on the current location and the current clock valuation. Thus, the current state is the combination of the current control state (or location) and the current clock valuation.

If you see a Finite State Automaton as a TA with $0$ clocks, only one clock valuation is possible, and it is the same for every states. The clock valuation is thus irrelevant, and the current state is entirely determined by the current location. To simplify the vocabulary, we identify both terms.

The word "location" is often used because it is hard to mistake it for "state". But it is not uncommon to see "discrete state" instead of "location", and "extended state" instead of "state".

$\endgroup$
1
  • 1
    $\begingroup$ Note that this distinction also occurs in other extended automata models, like pushdown automata (state = location + stack contents). $\endgroup$ – Klaus Draeger Nov 3 '15 at 13:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.