-
Notifications
You must be signed in to change notification settings - Fork 29
Short Term Parameters
The driver_param.xml file sets parameters to define driver behaviour and characteristics.
The various parameters that define the driving behaviour in SimMobilityST are mentioned in the general_driver_model parameters section. Each column represents the following population percentile 0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95
The first section contains parameters that describe the boundaries of the car following regime. A vehicle is in a car following regime when the time headway between the lead and the following vehicles is between the Car Following upper and lower headway thresholds. The following are the parameters that determine driving behaviour for a vehicle in car following regime.
<param name="min_response_distance" value="3"/>
The min_response_distance is the minimum space headway between the lead and following vehicles.
<param name="CF_parameters_1" value="0.0401 0.740 0.205 0.650 0.494 0.0"/>
CF_parameters_1 holds the array of parameters that is used to calculate the acceleration of the following vehicle in a car following regime.
<param name="CF_parameters_2" value="-0.374 0.000 0.650 0.750 0.605 0.0"/>
CF_parameters_2 holds the array of parameters that is used to calculate the deceleration of the following vehicle in a car following regime.
<param name="Car_following_acceleration_add_on" value="-1.7945 -1.1307 -0.7358 -0.4203 -0.1370 0.1370 0.4203 0.7358 1.1307 1.7945"/>
The Car_following_acceleration_add_on has values that correspond to specific population percentile. If the following vehicle in a car following regime falls under a population percentile, that corresponding value is added to acceleration calculated using CF_parameters_1 and the resulting value is the new acceleration of the vehicle.
<param name="Car_following_deceleration_add_on" value="-0.8092 -0.5099 -0.3318 -0.1895 -0.0618 0.0618 0.1895 0.3318 0.5099 0.8092"/>
Same as Car_following_acceleration_add_on, Car_following_deceleration_add_on has values that correspond to specific population percentile. If the following vehicle in a car following regime falls under a population percentile, that corresponding value is added to deceleration calculated using CF_parameters_2 and the resulting value is the new deceleration of the vehicle.
<param name="FF_Acc_Params_b0" value="10.0"/>
<param name="FF_Acc_Params_b2" value="0.3091"/>
<param name="FF_Acc_Params_stddev" value="1.15"/>
When the headway is greater than the threshold, the driver has the freedom to attain its desired speed. In such a case, the above parameters are used to calculate the free flow acceleration rate.
####Desired Speed
<param name="speed_factor" value="1.261"/>
<param name="speed_limit_add_on" value="-0.3125 -0.1969 -0.1281 -0.0732 -0.0238 0.0238 0.0732 0.1281 0.1969 0.3125"/>
The parameter speed_factor specified above are used to calculate the desired speed of the vehicles that are constrained by the posted speed limit. speed_limit_add_on has values that correspond to specific population percentile. If the vehicle falls under a population percentile, that corresponding value is added to desired speed that is calculated using speed_factor and the resulting value is the new desired speed of the vehicle.
<param name="target_gap_acc_parm"
value="0.604 0.385 0.323 0.0678 0.217 0.583 -0.596 -0.219 0.0832 0.170 1.478 0.131 0.300"/>
The target gap acceleration model is the acceleration that a vehicle executes to reach an acceptable gap for lane change. The parameter target_gap_acc_parm is used for this model.
The acceleration of the subject vehicle is specified by a set of possible accelerations and decision rules. In general a vehicle accelerates or decelerates in order to:
- react to the vehicles ahead (in the same or adjacent lane)
- perform a lane changing or merging manoeuvre
- respond to events e.g. red signals and incidents
- achieve its desired speed
This section provides the acceleration and deceleration thresholds of the vehicle types used by SimMobility based on grade and speed. Each column represents the following speed bins (m/s) defined in "speed_scaler" below and Each row represents a vehicle class
<param name="max_acc_car1" value=" 6.00 5.50 5.00 4.50 4.00"/>
<param name="normal_deceleration_car1" value=" -2.50 -2.25 -2.00 -1.75 -1.50"/>
<param name="max_deceleration_car1" value="-9.00 -8.50 -8.00 -7.50 -7.00"/>
<param name="max_acceleration_scale" value=" 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20"/>
<param name="normal_deceleration_scale" value=" 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20"/>
<param name="max_deceleration_scale" value=" 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20"/>
For each new vehicle created in SimMobility, a stochastic factor is drawn from the above distribution.
<param name="speed_scaler" value="5 10 5"/>
<param name="acceleration_grade_factor" value="1"/>
<param name="tmp_all_grades" value="0"/>
This section contains other general factors that are used for acceleration/deceleration thresholds
Based on a headway threshold, a driver is assumed to be in one of the two following regimes: the car following regime and the free ow regime. If the current headway is less than the threshold, the driver is assumed to be in the car following regime and follow its leader. Speed selection and hence the acceleration decision is governed by the speed of the leader. Otherwise, the driver is assumed to be in the free flow regime in which case speed selection is governed by its desired speed.
A headway buffer is found for each vehicle and is a behavioural parameter that describes the aggressiveness of a driver for accepting a headway gap in lane changing, merging, and car-following. Some value between the lower and upper bound will be added to the minimum headway gaps for the population, which are constants also provided in
<param name="hbuffer_lower" value="0.80"/>
<param name="hbuffer_Upper" value="2.1775 2.4817 2.6627 2.8073 2.9371 3.0628 3.1926 3.3372 3.5182 3.8224"/>
<param name="yellow_stop_headway" value="1.0"/> <!-- s -->
<param name="min_speed_yellow" value="2.2352"/> <!-- m/s -->
<param name="driver_signal_perception_distance" value="150.0"/>
These parameters decide the acceleration of the vehicle while approaching (signalized) intersections.
<param name="dec_update_step_size" value="0.537 0.20 0.25 1.05 0.0"/>
<param name="acc_update_step_size" value="0.750 0.25 0.30 1.55 0.0"/>
<param name="uniform_speed_update_step_size" value="0.750 0.25 0.30 1.55 0.0"/>
<param name="stopped_vehicle_update_step_size" value="0.537 0.20 0.25 1.05 0.0"/>
Driver behaviour (e.g. responses to stimuli) is updated periodically during simulation depending on the state of the vehicle (i.e. depending on whether the vehicle is decelerating, accelerating, travelling at uniform speed, or is stopped). At each call of the corresponding function (i.e. at the end of each decision instant) a new update state size is drawn from a truncated log-normal distribution.
<param name="visibility_distance" value="10"/>
The visibility_distance parameter is used in the car follow model to reduce the follower reaction time when subject vehicle brakes (in meter)
<param name="MLC_PARAMETERS" value="
100.0 #lower bound(meter)
190.0 #delta(meter)
2.0 #minimum time in lane (sec) "/>
The probability of starting a mandatory lane change or staying in the current lane model requires that the vehicle distance from the downstream node (or lane drop) be greater than a distance 'lower bound'. This probability is computed using the parameters that are mentioned above. A vehicle must remain in a lane for a minimum of minimum time in lane seconds before considering another lane change.
<param name="lane_utility_model" value="
4.2656 # current lane constant
0.3213 # right lane constant
-1.1683 # right most lane dummy
0.0 # currently unused
0.0633 # front vehicle speed (all lanes)
-1.0 # bus following dummy
0.0058 # front vehicle spacing
-0.2664 # heavy neighbour in lane
-0.0088 # density in lane
-3.3754 # tailgate
10 # gap behind threshold for tailgate dummy (meters)
19 # density threshold for tailgate dummy
-2.3400 # one lane change required
-4.5084 # two lane changes required
-2.8257 # each additional lane change required
-1.2597 # next exit, one lane change required
-0.7239 # next exit, each additional lane changes required
-0.3269 # distance to exit "/>
This model describes drivers’ choice of lane they would want to travel in. A driver chooses one of up to three alternatives: to stay in the current lane (CL) or to target changing either to the right lane (RL) or to the left lane (LL).
<param name="critical_gaps_param" value="
0.2 # lead gap constant
-0.231 # lead gap negative speed difference
-2.700 # lead gap positive speed difference
1.112 # lead gap standard deviation
0.5 # lag gap constant
0.000 # lag gap negative speed difference
0.2 # lag gap positive speed difference
0.742 # lag gap standard deviation
6.0 # max speed difference "/>
Once a driver has decided to change lanes, he/she will examine adjacent gaps and deem them acceptable or unacceptable for merging. Below are parameters to describe driver (critical) gap acceptance threshold.
<param name="GAP_PARAM_0" value="-1.23, -0.482, 0.224, -0.0179, 2.10, 0.239"/> #back
<param name="GAP_PARAM_1" value="0.00, 0.00, 0.224, -0.0179, 2.10, 0.000"/> #adj
<param name="GAP_PARAM_2" value="-0.772, -0.482, 0.224, -0.0179, 2.10, 0.675"/>
These parameters help in making the choice of the target gap (forward, backward or adjacent) depending on the gap utilities
<param name="nosing_param" value="
1.0 # time horizon (second) for constant acc
0.5 # constant
0.6 # coef for number of lanes
0.1 # coef for time (in minutes) since tagged
0.5 # max prob one on ramp lane drop, incident, red LUS, etc
1.0 # max prob two in connection to next link
30.0 # max yielding time (seconds)
100.0 # max stuck time (seconds)
600.0 # max distance for nosing
40.0 # min distance for nosing "/>
The main threshold parameters for the yielding and nosing states are defined in this section.
<param name="kazi_nosing_param" value="-3.159 0.313 -0.027 2.050 0.028 0.6"/>
<param name="CF_CRITICAL_TIMER_RATIO" value="0.5"/>
The kazi_nosing_param is directly used in calculating the probability to nose when distance from a point is less than a threshold. The CF_CRITICAL_TIMER_RATIO is used to change driver's reaction time when in a nosing behavior
<param name="LC_Yielding_Model" value="
0.80 # previously not yielding
1.00 # previously yielding "/>
These are parameters that apply to vehicles that currently yield to nosing vehicles
<param name="MLC_Yielding_Probabilities" value="
0.13 # None
0.71 # Upto one
0.13 # two
0.03 # three "/>
These parameters are used to compute probabilities to yield to other vehicles
<param name="intersection_attentiveness_factor_min" value="0.3"/>
<param name="intersection_attentiveness_factor_max" value="0.6"/>
These parameters define the minimum and maximum attentiveness factors of the drivers at intersections.
<param name="minimum_gap" value="1.25"/>
<param name="critical_gap_addon" value="1.0 0.5"/>
<param name="impatience_factor" value="0.2"/>
The above parameters are used to evaluate gap and decide acceleration
<param name="tailgate_separation_time" value="1"/>
Separation time between vehicles following one another (in sec) (also known as T1)
<param name="conflict_separation_time" value="2.46"/>
Separation time between vehicles with conflicting trajectories (in sec) (also known as T2)
<param name="min_speed" value="0.1"/>
<param name="LC_Discretionary_Lane_Change_Model_MinTimeInLaneSameDir" value="1"/>
<param name="LC_Discretionary_Lane_Change_Model_MinTimeInLaneDiffDir" value="1"/>
<param name="check_stop_point_distance" value="200"/>
<param name="density" value="0.0"/>
min_speed is the minimum speed to consider for a moving vehicle. The next two parameters decide the minimum time required between lane changing decisions. check_stop_point_distance defines the visibility of the vehicles at stopping points. density is the default value that is calculated by SimMobility.