Hello, 1. Unfortunately, No. 2. It doesn't make a difference because a new route is created for the new list of edges as well.
Also, you are right regarding your assumption regarding the way vehicle and route input has to be specified to start the simulation with initial traffic. >From a technical perspective, another vehicle attribute 'departEdge' which would allow you to use the same route for vehicles starting on different edges should not be too hard to implement. You could implement this yourself, or wait until someone else implements this (no telling when that might be). Currently you have another option: You can also save the simulation state at a selected time and then start anther simulation using this state. See http://sumo-sim.org/wiki/SUMO (options --save-state.* --load-state.*) This feature is a rather recent addition so be sure to tell us if you notice odd behavior. regards, Jakob 2014/1/13 Pieter Loof <[email protected]> > Hello, > > Thank you for the quick answer. It is still not clear to me though, what > are the suitable input files? Let me phrase my case and questions in more > detail: > > In TraCI, as far as I can see now, when creating a new vehicle it has to > receive an existing route before it can be added. But if I want to put > vehicles on every edge, then I need to add an enormous amount of routes, > because I need routes with every edge of the network as starting point. > Which brings up the following questions: > 1. Is it possible that vehicles can start at any edge of their route, such > that I only need a limited set of routes, and then I can put vehicles on > several edges of this route? > 2. If the answer to 1 is NO, then I wonder about whether an alternative to > the described method makes any difference: insert every vehicle using the > same default route, then change the route of each vehicle by providing a > list of edges (and change the location of the vehicle). Does this make a > difference internally, or does the 'changeRoute' method also add a new > Route to the list of existing routes anyway? > > But since TraCI is slower, I'd like to have a simulation with initial > traffic without using TraCI. The only method I can currently think of is > creating/generating an additional xml file with a list of vehicles. But > from sumo-generated vehicle files I see that each defined vehicle should > have a route that is defined in a route-file, which means that I also have > to generate an enormous amount of routes (a few routes times the number of > edges in the network)? > 3. Is there a smarter way of doing this (without using TraCI) and is there > any help or documentation on how to do this? And does question 1 has the > same answer for this case? > > Best regards, > Pieter > > > On 12 January 2014 17:15, Jakob Erdmann <[email protected]>wrote: > >> Hello, >> when you start a simulation you cannot only define flows but also >> individual vehicles with their positions and speed. This allows you to >> initialize your network with any desired traffic state. (If you only have >> the number of vehicles on an edge you need to make up plausible positions). >> You will need to write a script to transform your information table into >> suitable input files. >> >> When running your simulation under control of a TraCI script you can also >> place vehicles on any position and with any speed you desire. Thus you >> could fill your simualation in the first step and then continue your >> simulation from there. >> >> regards, >> Jakob >> >> >> 2014/1/11 Pieter Loof <[email protected]> >> >>> While working with SUMO, so far I see that each simulation starts with an >>> empty road network. When starting the simulation, vehicle flows start >>> generating vehicles at the boundaries of the network. But now I want to >>> start a simulation with initial traffic. The idea is to perform a >>> predictive simulation, in order to find out how the traffic evolves, >>> using >>> the current traffic conditions as the initial network situation. >>> >>> So suppose for example that I have a large information table, with for >>> each >>> network edge the number of vehicles on it and the average speed. Then one >>> of the SUMO modules should be able to read this information and generate >>> the initial traffic on the network. >>> >>> Now I wonder whether this is possible with TraCI and whether this is >>> possible without TraCI? If so, please provide more information on how to >>> do >>> this. If not, then I think this is very important future work, since >>> starting simulations with an empty network is highly unrealistic in >>> certain >>> cases, it makes it virtually impossible to reproduce a received live >>> traffic state and thus virtually impossible to make traffic forecasts. >>> >>> ------------------------------------------------------------------------------ >>> CenturyLink Cloud: The Leader in Enterprise Cloud Services. >>> Learn Why More Businesses Are Choosing CenturyLink Cloud For >>> Critical Workloads, Development Environments & Everything In Between. >>> Get a Quote or Start a Free Trial Today. >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk >>> _______________________________________________ >>> sumo-user mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/sumo-user >>> >> >> > ------------------------------------------------------------------------------ CenturyLink Cloud: The Leader in Enterprise Cloud Services. Learn Why More Businesses Are Choosing CenturyLink Cloud For Critical Workloads, Development Environments & Everything In Between. Get a Quote or Start a Free Trial Today. http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk _______________________________________________ sumo-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sumo-user
