cmcmanis%pepper@Sun.COM (Chuck McManis) (08/11/89)
I have an interesting problem. At one point someone wondered if there
were any traffic simulators around. It sounded like an interesting
project, sort of a battling robots kind of thing, but the "cars" would
be programmed with destinations and driver tendencies rather than
weapons and hunting algorithims. I'm probably going to write it in
Modula with co-routines for each car. But the interesting thing is
that where I can get freeways and highways, and vehicles drawn to
scale, my On/Off ramps look "funny." and don't seem to be correct.
So I was wondering if there are some standards for building them
that make them all look about the same. Things like maximum degree
of arc, width, accelleration/decelleration lane lengths (although that
is one of the things I'll be varying in my simulation). The goal being
that I should be able to set up an "intersection" of some type and
put car "feeders" and "meters" on it at the entrances to and exits
from the screen respectively. And see how many cars can get through
my intersection. Then by varying things like meter lights, traffic
signals, ramp design I'd like to see how that effects my throughput.
Of course I'd also like to simulate some nearby intersection to see
how closely my model matches reality etc. Any pointers or hints would
be appreciated.
--Chuck McManis
uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.
"A most excellent barbarian ... Genghis Kahn!"451061@UOTTAWA.BITNET (Valentin Pepelea) (08/12/89)
Chuck McManis <cmcmanis%pepper@sun.com> writes in <120811@sun.Eng.Sun.COM> > I have an interesting problem. At one point someone wondered if there > were any traffic simulators around. It sounded like an interesting ... > that make them all look about the same. Things like maximum degree > of arc, width, accelleration/decelleration lane lengths (although that > is one of the things I'll be varying in my simulation). The goal being > that I should be able to set up an "intersection" of some type and > put car "feeders" and "meters" on it at the entrances to and exits > from the screen respectively. ... > Then by varying things like meter lights, traffic > signals, ramp design I'd like to see how that effects my throughput. > Of course I'd also like to simulate some nearby intersection to see > how closely my model matches reality etc. Gee, this sounds awfully like queue length simulation. Basically all you have a bunch of nodes across the highway, with entry and exit ramps as vectors that have different arrival (exit) rates and different probabilities. You may change the arrival rate of new cars as you like. Usually using a formula involving exponential probabilities. There already is a simulation package around, called QNAPLS. Just last term (Winter) I had to use it to simulate average queue lengths of tasks waiting for a processor in a multiprocessor system. I didn't though, this was one of those things engineering students team around for, to get rid of quickly. :-) Valentin _________________________________________________________________________ The godess of democracy? "The Name: Valentin Pepelea tyrants may destroy a statue, Phonet: (613) 231-7476 but they cannot kill a god." Bitnet: 451061@Uottawa.bitnet Usenet: Use cunyvm.cuny.edu gate - Confucius Planet: 451061@acadvm1.UOttawa.CA
frost@sdsu.edu (Richard Frost) (08/12/89)
Traffic simulation is not new. There are 3 SHELVES of books on the subject at most university libraries, two professional journals devoted to nothing else, and hundreds of models appearing as research papers in many mathematical journals. -- Richard Frost E-mail: frost@sdsu.edu - a scientist is someone who learns more and more about less and less until they know everything about nothing! (Ziman)
hasker@m.cs.uiuc.edu (08/12/89)
/* Written 5:49 pm Aug 10, 1989 by cmcmanis%pepper@Sun.COM in m.cs.uiuc.edu:comp.sys.amiga */ /* ---------- "Freeway design?" ---------- */ So I was wondering if there are some standards for building them that make them all look about the same. Things like maximum degree of arc, width, accelleration/decelleration lane lengths (although that is one of the things I'll be varying in my simulation). /* End of text from m.cs.uiuc.edu:comp.sys.amiga */ Try calling your local highway department. I'll bet they have a notebook listing such things. From working for the local Soil Conservation Service office I know they have manuals for designing culverts, bridges, road beds, etc. Sort of fun browsing; I never imagined that someone would go to the trouble of writing a manual for such mundane things, or that it would be so thick! Rob Hasker hasker@cs.uiuc.edu
pfaff@mercury.asd.contel.com (Ray Pfaff - Oakwood 457 934-8162) (08/14/89)
>There already is a simulation package around, called QNAPLS. Just last term >(Winter) I had to use it to simulate average queue lengths of tasks waiting >for a processor in a multiprocessor system. I didn't though, this was one of >those things engineering students team around for, to get rid of quickly. :-) Sorry to waste the bandwith on this, but my mail to the person that posted this bounced back to me. Any idea where I could get this package? I need to do some queuing simulations for a multiprocessor project that I am doing.