Please be aware that 'ACC' and 'CACC' are just one possible implementation of (cooperative) automated cruise control. They are not the "definite" models for these things. If the models behave differently from your expectations, this may reflect the very specific situations they were calibrated for by the original model authors.
In my own tests, I observe that the capacity of 100% CACC with tau=0.6 is up to 16% higher compared to the default. However, this is heavily influenced by the choice of insertion speed value (with values 'desired' and 'avg' performing well and value 'max' performing far worse than the default model. I suggest that you do your own parameter calibration until the observed results reflect your expectation. To model different reaction times, use different actionStepLength values for the different types. regards, Jakob Am So., 30. Jan. 2022 um 06:03 Uhr schrieb Pradeep Viyaluru Harinath < [email protected]>: > Hello , > > I am analyzing a freeway section and trying to implement the ACC and CACC > model in my simulation. I observed that as the number of vehicles with ACC > and CACC model increased the capacity of freeways started to decline and it > was lower than the manual driven vehicles. I have given Tau as 0.6 for CACC > vehicles and 0.9 for ACC vehicles , correspondingly I reduced the > simulation step length as well to 0.1. I wanted to know two things here how > do I implement the reaction time for ACC or CACC vehicle say 0.1 sec and > should I use the simulation step length as 1 and use the action step length > for a particular vehicle type to improve the speed of simulation. Further > why is my capacity decreasing even with ACC and CACC modes? > > > > Regards, > > V.H.Pradeep > _______________________________________________ > sumo-user mailing list > [email protected] > To unsubscribe from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
