Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* splitting street vehicle - based legs in preparation for parking decision * debugging discrepancy between R5 path traversal assumption (end of first to start of last) and MATSim/BEAM (end of first to end of last) * add test to ensure durations make sense within trips * make splitting optional and add test coverage * RHM doesn't need to do parking * updating protocols * framing the pass of from driver to ChoosesParking trait * add choosesparking class * getting simple zonal parking infrastructure selection algorithm implemented * wiring up creation of parking manager and introducing messaging from person to manager * put taz coords into EPSG and some debugging * completing the parking process and debugging, runs through beamville now * resetting beamville conf * tweaks * WIP parking specification from csv * WIP parking specification from csv - thresholdForMakingParkingChoiceInMeters parameter added to config * WIP parking specification from csv - fix to compilation issue * WIP parking specification from csv - fix to compilation issue * WIP parking specification from csv - using thresholdForMakingParkingChoiceInMeters * WIP: Parking system * WIP: Parking System - generating/reading parking csv file with stall values * update lfs ip * Parking system: events logged to out events file * Parking system: working on score * Parking System: Wip sampleLocationForStall * Parking system: Writting area in taz-centers.csv * Parking system: refactor sampleLocationForStall to use exponential distribution * Parking system: sampleLocationForStall using exponential distribution * #261 - Missing implementation tasks * #262 WIP: events output review for test * WIP #262: make sure at least some parking used, arrival and departure should be from same parking 4 tuple * In the middle of adding parking back into work flow after major DriveVehicle refactor. Stuck not on how to handle parking generically instead of assuming along Persons park. * making code compile * go through PersonAgent before going into ChoosesParking * fixing error * add parking scenarios * fix test typo * Limited parking scenario * Adding ReleasingParkingSpot state * Adding initial park for vehicles * Making parkingSpec pass: adding initial parking for vehicles in beamMobsim * get rid of ChoosesParkingData because it adds nothing and confuses tracing transitions * getting it back to where I started today * debugging * adapt to no car route found in parking planning * Pulling up StateTimeout for BasePersonData case in ChoosesParking for ReleasingParkingSpot state. Making parkSpec pass. * Throwing ParkEvents after last PathTraversalEvents plus test to verify this * Specs for expensive, limited and empty parking should reduce driving. Spec for limited parking access should increase VMT * #320 - Ready to use in RideHailManager * #321 - dealing with DepotParkingInquiry in parking manager * Missing config and data for TazTreeMap * Removing repeated initialization after merge * make mustParkAtEnd be false as default * Add missing ChoosesParking trait during PersonAgent mixing * Change order for BeamWarmStart in BeamMobSim * several fixes and tuning of replanning * more fixes for failing tests though still failing * tests usually pass, but replanning is overly diminishing value of car for some reason, causing some tests to fail occasionally * nasty bug allowing walkers and bikers to travel at the speed of a car * merging * get SF light to run and fix injection silliness * fixing broken tests * reproduce TAZs with area column, produce parking data * fixed another test, needed to mock TAZTreeMap correctly * fixing mistke made during merge, need to pass flag to R5RoutingWorker to split legs for parking * ride hail wasn't being picked up * change in how config is passed in caused previous method of customizing to break now fixed * something has changed in router making expected travel times in TimeDependentRoutingSpec need to be adjusted * now there is one route out of 1000s that fails for CAR, so softening requirement in spec * fmt * ignore poorly formulated parking test
- Loading branch information