Hello, in SUMO there is a large asymmetry between lane-changing to the left and lane-changing to the right (especially when lane-changing to overtake slower vehicles). This is mainly due to German regulations which demand that vehicles should not overtake on the right on motorways. There are several ways to reduce this assymetry - in the latest svn version ( http://sumo-sim.org/wiki/Downloads#Nightly_Snapshots) there is the option --lanechange.overtake-right which permits overtaking on the right in all traffic situations - in src/microsim/MSLCM_LC2013.cpp there are constants called CHANGE_PROB_THRESHOLD_RIGHT (SUMOReal)2.0 CHANGE_PROB_THRESHOLD_LEFT (SUMOReal)0.2 Currently, the threshold for lane-changing due to being slowed down is much higher for changing to the right than for changing to the left. If you lower that value, vehicles will change to the right for overtaking purposes faster. (We plan to make such parameters configurable eventually - http://sumo-sim.org/trac.wsgi/ticket/1136)
regards, Jakob 2014-05-06 16:04 GMT+02:00 Judhakitt Gun Tirathumcharoen < [email protected]>: > Dear Jakob, > > > Perhaps it was long time about my previous emails. I am not sure if you > have time to take a look in my configurations which I sent with 2 emails > earlier. If you have a little time, please kindly help me to see my issue > about "motorcycle overtaking in to flows of cars":- > - from the screenshot, the white motorcycle is an obstacle > - red motorcycles reaching to the white motorcycle and they stop. Then > they wait for some times to overtake in to flow of cars > > ---------------------------- > Objective:- > - I would like make the motorcycle overtaking immediately without stopping > at the obstacle (white motorcycle) > - To simulate at higher speed of car and motorcycle > - Shockwave generated by motorcycle overtaking will be analysed from SUMO > output. > > I attach my sumo configuration in mocyc.doc, which you have to rename to > mocyc.zip, then unzip it. (Googlemail blocks this kind of attachment > somehow). > What I did was I reduced speed of car and motorcycle to 3 m/s and 7 m/s > respectively in order that motorcycle could overtake (even motorcycle stops > at obstacle before overtaking). But it does not correspond to the real > world where cars and motorcycle are much faster than that. > ---------------------------- > <vType minGap="1.5" sigma="0.3" maxSpeed="7.0" width="1.0" length="2.0" > guiShape="motorcycle" vClass="motorcycle" decel="3.0" accel="3.5" > id="Motorcycle"/> > <vType minGap="2.5" sigma="0.3" maxSpeed="3.0" width="2.0" length="4.0" > guiShape="passenger" vClass="private" decel="3.0" accel="2.5" id="Car"> > > <carFollowing-IDM/> > > </vType> > ---------------------------- > Questions:- > Is it possible to let motorcycles overtaking immediately? or > Do I have to modify SUMO source codes, such as lateral movement model, > longitudinal movement model, lane changing model, or car-following model, > etc? If so, could you please advise me where I should change the parameters > so that:- > - motorcycle can overtake immediately > - motorcycle can overtake even car flow are driving faster (faster than 3 > m/s, perhaps car should be faster than 13.8 m/s) > ---------------------------- > > > > Have a nice day. > > > > Best Regards, > > > Judhakitt > > > -----ข้อความต้นฉบับ---- > จาก: Judhakitt Tirathumcharoen [mailto:[email protected]] > ส่ง: Thursday, 27 March, 2014 17:54 > ถึง: Jakob Erdmann > สำเนาถึง: [email protected] > เรื่อง: Re: [sumo-devel] Implementing motorcycle mobility model in SUMO > > Dear Jakob, > > > I sent you a email with input files in zip, but it cannot not be delivered > because google mail server blocks. I try to send you agian the attachment > in this email. The file mocyc.doc must be renamed to mocyc.zip, then please > unzip it. > > > Best Regards, > > > Judhakitt > > > Quoting Jakob Erdmann <[email protected]>: > > > If the flow of cars is to dense the vehicle may have trouble changing > > to the left. SUMO currently cannot distinguish changing to the left at > > on-ramps from changing to the left in other settings so there are some > > constrains on what the cars on the left lane will do to help the ego > > contrary to what you might expect (this is to avoid jamming on > motorways). > > regards, > > Feel free to send me input files to take a look though. It might still > > be something else. > > Jakob > > > > > > > > 2014-03-25 14:26 GMT+01:00 Judhakitt Tirathumcharoen < > [email protected]>: > > > >> Dear Jakob and Manish, > >> > >> > >> Thank you very much for your fast response. It will be very > >> impressive to try your simtram in sumo-0.17. K=3 is very reasonable > >> to simulate the real situation. May I know when approximately simtram > >> will be available to download. > >> > >> Regarding to Q3, I forgot to mention in my scenario that, in the 2 > >> lanes, there are flow of cars passing on the left lane, while the > >> vehicle on right lane try to overtake the obstacle. What I was trying > >> to see is a vehicle overtaking the obstacle on the right lane into > >> the flow of cars, which are passing on the left lane. > >> May I send my input files and screenshots to Jakob in the separate > >> email > >> > >> > >> Best Regards, > >> > >> > >> Judhakitt > >> > >> > >> > >> > >> Quoting manish chaturvedi <[email protected]>: > >> > >> Missed to answer the second question. Sorry about that. Here it > follows: > >>> > >>> *Regarding slow down of simulation:* > >>> > >>> > >>> In Simtram, each lane is divided in to number of strips > (configurable).. > >>> > >>> And at all the places in the sumo code where we do iterations for > >>> lanes (e.g. for updating vehicle positions and all), we need to do > >>> nested loops for number of strips as well. > >>> > >>> Hence, it seems that if we have k strips in a lane, our simulation > >>> is k times slower than the actual Sumo. k=3 is reasonable for > >>> simulations involving two wheelers. > >>> > >>> Sincerely, > >>> Manish > >>> > >>> > >>> > >>> On Tue, Mar 25, 2014 at 2:42 PM, manish chaturvedi > >>> <[email protected] > >>> >wrote: > >>> > >>> Sir, we need to acknowledge that we are/were not part of the > >>> Simtram > >>>> team.. > >>>> > >>>> We got the source code from Simtram Team and started adapting it > >>>> for > >>>> sumo-0.17 for our own research work and were able to do it > successfully. > >>>> > >>>> We will upload the adapted simtram version for sumo-0.17 and share > >>>> the link with user community soon. > >>>> > >>>> Currently we are working on documentation part it may take about 2 > >>>> weeks time. > >>>> > >>>> Sincerely, > >>>> > >>>> Manish > >>>> > >>>> > >>>> > >>>> > >>>> > >>>> On Tue, Mar 25, 2014 at 12:21 PM, Jakob Erdmann < > >>>> [email protected]> wrote: > >>>> > >>>> Hello Manish, > >>>>> I'm happy to hear that you remain active. > >>>>> Can you tell us when the downloads will be active again? > >>>>> Regarding your report on slow simulations: > >>>>> At what number of vehicles do you experience drastic slowdown? > >>>>> How much slower compared to regular SUMO is SimTRaM? > >>>>> > >>>>> best regards, > >>>>> Jakob > >>>>> > >>>>> > >>>>> 2014-03-25 5:24 GMT+01:00 manish chaturvedi <[email protected]>: > >>>>> > >>>>> regarding Q1 of Judhakitt: > >>>>> > >>>>>> > >>>>>> We have adapted SIMTRAM for sumo-0.17 version on fedora and its > >>>>>> working fine.. > >>>>>> > >>>>>> We started with the fresh sumo-0.17 version and incorporated all > >>>>>> the changes from simtram source code and it adapted fine. > >>>>>> > >>>>>> Only problem we are facing is, with large number of vehicles, the > >>>>>> simulation slows down drastically.. > >>>>>> > >>>>>> Please let us know if further assistance needed.. We will be more > >>>>>> than happy to reply.. > >>>>>> > >>>>>> Sincerely, > >>>>>> > >>>>>> Manish and Viral > >>>>>> > >>>>>> > >>>>>> > >>>>>> > >>>>>> On Sun, Mar 23, 2014 at 7:08 PM, Jakob Erdmann < > >>>>>> [email protected]> wrote: > >>>>>> > >>>>>> Hello, > >>>>>>> regarding your questsions... > >>>>>>> > >>>>>>> Q1: putting two vehicles next to each other on the same lane is > >>>>>>> currently not possible. There is a sumo version called SiMTraM > >>>>>>> writen by Indian researchers which does this ( > >>>>>>> http://www.civil.iitb.ac.in/~vmtom/SiMTraM_Web/html/index.html) > >>>>>>> but it appears the project is inactive (at least the download > >>>>>>> links are broken). > >>>>>>> Also, it is based on a very old version of SUMO so it can be > >>>>>>> expected to contain a long list of issues which have since been > >>>>>>> fixed. > >>>>>>> > >>>>>>> Q2: there is a switch in the code in > >>>>>>> src/:microsim/MSLCM_LC2013.cpp #define OVERTAKE_RIGHT_FORBIDDEN > >>>>>>> true. By setting this to false, overtaking on the right will be > >>>>>>> possible. (We should probably turn this into an option, see > >>>>>>> http://sumo-sim.org/trac.wsgi/ticket/1212) > >>>>>>> > >>>>>>> Q3: in my tests vehicles overtake on the left without slopping > >>>>>>> (I see a small slowdown of 0.2m/s which maybe shouldn't be there > >>>>>>> but nothing larger. Can you send input files that let us > >>>>>>> reproduce a simulation with implausible behavior? > >>>>>>> > >>>>>>> best regards, > >>>>>>> Jakob > >>>>>>> > >>>>>>> > >>>>>>> 2014-03-21 12:43 GMT+01:00 Judhakitt Tirathumcharoen < > >>>>>>> [email protected]>: > >>>>>>> > >>>>>>> > Dear Sirs, > >>>>>>> > > >>>>>>> > > >>>>>>> > My name is Judhakitt and I am studying post-master in ITS at > >>>>>>> Eurecom, > >>>>>>> > France. > >>>>>>> > I have been working on impact of motorcycle mobility by using > >>>>>>> > SUMO > >>>>>>> in > >>>>>>> > the past 4 months and supervised by Prof. Jerome Härri. > >>>>>>> > At the moment, in my semester project (duration March-June > >>>>>>> > 2014), > >>>>>>> the > >>>>>>> > objective is to model the motorcycle mobility in SUMO and > >>>>>>> > would like to ask you for suggestions. > >>>>>>> > > >>>>>>> > (Remark: "overtaking vehicle" in this case means motorcycle) > >>>>>>> > > >>>>>>> > Q1) Is it possible to put more than one vehicle (for example > >>>>>>> > vehicle type is motorcycle, half-width, and length of > >>>>>>> > passenger car) in the same lane at the same time? > >>>>>>> > Some research papers show that up to 3 motorcycles can be on > >>>>>>> > the > >>>>>>> same > >>>>>>> > lane at the same time in the real world. What I have tried to > >>>>>>> > model before was to create virtual sub-lane only for > >>>>>>> > motorcycle (cars are not allowed), and another lane for car > >>>>>>> > and motorcycle. However there can just only be 2 motorcycles > >>>>>>> > at the same position (one in > >>>>>>> sub-lane, > >>>>>>> > and one car/or one motorcycle in the normal lane). > >>>>>>> > > >>>>>>> > Q2) Are there any possibility to let vehicle on left lane to > >>>>>>> overtake > >>>>>>> > on the right lane? > >>>>>>> > I have noticed that a vehicle cannot overtake on the right > >>>>>>> > side. For example there are 2 lanes on the same direction > >>>>>>> > (called left and > >>>>>>> right > >>>>>>> > lane). If one vehicle suddenly stops on the left lane, the > >>>>>>> > following vehicles are stopping and queuing up, but they do > >>>>>>> > not overtake to > >>>>>>> the > >>>>>>> > right. > >>>>>>> > I discuss with Prof. Härri and he told me that SUMO blocks > >>>>>>> > this by law. There should be some part in source code that > >>>>>>> > allows vehicle to overtake on the right side. > >>>>>>> > > >>>>>>> > Q3) Why a vehicle stops at a static obstacle, without simple > >>>>>>> > overtaking continuously? > >>>>>>> > For example, there are 2 lanes, right and left. A car (or any > >>>>>>> vehicle) > >>>>>>> > stops on the right lane for long time. (In this case, vehicles > >>>>>>> > are allowed to overtake, not blocking by law). When the > >>>>>>> > following > >>>>>>> vehicles > >>>>>>> > >>>>>>> > approaching to this stopping car, they are stopping before > >>>>>>> > trying to overtake. I try to simulate in different scenario > >>>>>>> > with different speed, acceleration, but I never make the > >>>>>>> > following vehicles > >>>>>>> overtake > >>>>>>> > suddenly, continuously without stopping after the stopping car. > >>>>>>> > > >>>>>>> > > >>>>>>> > Please kindly feedback or let me know if you need anything > >>>>>>> > from my > >>>>>>> side. > >>>>>>> > > >>>>>>> > > >>>>>>> > Best Regards, > >>>>>>> > > >>>>>>> > > >>>>>>> > Judhakitt > >>>>>>> > > >>>>>>> > > >>>>>>> > > >>>>>>> > > >>>>>>> ------------------------------------------------------------ > >>>>>>> ------------------- > >>>>>>> > This message was sent using EURECOM Webmail: > >>>>>>> http://webmail.eurecom.fr > >>>>>>> > > >>>>>>> > > >>>>>>> > > >>>>>>> > > >>>>>>> ------------------------------------------------------------ > >>>>>>> ------------------ > >>>>>>> > Learn Graph Databases - Download FREE O'Reilly Book "Graph > >>>>>>> > Databases" is the definitive new guide to graph databases and > >>>>>>> their > >>>>>>> > applications. Written by three acclaimed leaders in the field, > >>>>>>> > this first edition is now available. Download your free book > today! > >>>>>>> > http://p.sf.net/sfu/13534_NeoTech > >>>>>>> > _______________________________________________ > >>>>>>> > sumo-devel mailing list > >>>>>>> > [email protected] > >>>>>>> > https://lists.sourceforge.net/lists/listinfo/sumo-devel > >>>>>>> > > >>>>>>> > >>>>>>> ------------------------------------------------------------ > >>>>>>> ------------------ > >>>>>>> Learn Graph Databases - Download FREE O'Reilly Book "Graph > >>>>>>> Databases" is the definitive new guide to graph databases and > >>>>>>> their applications. Written by three acclaimed leaders in the > >>>>>>> field, this first edition is now available. Download your free > >>>>>>> book today! > >>>>>>> http://p.sf.net/sfu/13534_NeoTech > >>>>>>> _______________________________________________ > >>>>>>> sumo-devel mailing list > >>>>>>> [email protected] > >>>>>>> https://lists.sourceforge.net/lists/listinfo/sumo-devel > >>>>>>> > >>>>>>> > >>>>>> > >>>>>> > >>>>> > >>>> > >>> > >> > >> > >> ------------------------------------------------------------ > >> ------------------- > >> This message was sent using EURECOM Webmail: > >> http://webmail.eurecom.fr > >> > >> > > > > > > > ------------------------------------------------------------------------------- > This message was sent using EURECOM Webmail: http://webmail.eurecom.fr > > ------------------------------------------------------------------------------ Is your legacy SCM system holding you back? Join Perforce May 7 to find out: • 3 signs your SCM is hindering your productivity • Requirements for releasing software faster • Expert tips and advice for migrating your SCM now http://p.sf.net/sfu/perforce _______________________________________________ sumo-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sumo-user
