Settings and parameters¶
Settings and parameters give model control to the external simulator. This section covers three different types of settings and parameters as given in Table 14. The model settings can be given as command line arguments to control what model components are used in the OTS model. This holds for both background traffic generated by OTS, and vehicles spawned by the external simulator. For specific spawned vehicles different model components can be used. In the Vehicle message there is a parameters field that can be used for this. Any parameter ID that starts with -- is recognized as a setting. Model settings that are part of the Vehicle message are only used for the specific vehicle. Model settings given as command line arguments are used for all vehicles, unless overridden for a specific vehicle in the Vehicle message.
Table 8.14: Setting types.
| Settings | Description | Command line | Vehicle message |
|---|---|---|---|
| Transceiver | These control the technical operation of the OTS transceiver | yes | no |
| Model | Settings that enable or disable model components | yes | yes |
| Parameters | Parameter values used on the model components | no | yes |
To get an overview of all settings, including default values, one can call the transceiver with the --help command line argument. For example:
java.exe -jar OtsTransceiver.jar --help
The following sections show available settings across the different types. The last section explains the relevance of parameter values on vehicles that are externally controlled.
Transceiver settings¶
Transceiver settings are given as command line arguments and control the technical operation of the OTS transceiver. Depending on the implementation of the transceiver, different settings apply.
General¶
The settings in Table 8.15 apply to all implementations of the OTS transceiver (AbstractOtsTransceiver).
Table 8.15: Generic transceiver settings.
| Setting | Description |
|---|---|
| --gui | Whether to show the GUI |
| --idPrefix | Prefix of ID of generated vehicles |
| --useRoadName | Whether to use the road name to identify origins and destinations |
The --useRoadName setting applies to the use of an OpenDRIVE network. By default it is assumed that origins and destinations in the OD matrix are defined by road IDs. With this setting the road name can be used instead. Note that OpenDRIVE does not guarantee unique road names.
Sim0mq transceiver¶
The settings in Table 8.16 apply to the Sim0mq implementation of the OTS transceiver (OtsTransceiverSim0mq).
Table 8.16: Sim0mq transceiver settings.
| Setting | Description |
|---|---|
| --federationId | Federation ID to receive/sent messages |
| --otsId | OTS ID to receive/sent messages |
| --clientId | Client ID to receive/sent messages |
| --bigEndian | Big-endianness |
| --port | Port number |
| --logLevel | Logging level: OFF, ERROR, WARN, INFO, DEBUG or TRACE |
Model settings¶
The settings in Table 8.17 apply to all implementations of the OTS transceiver and govern what model components are used. Some settings are valid depending on the value of the used Fuller implementation. See the java documentation of LmrsFactory for a table indicating this applicability.
When model settings are sent as part of the Vehicle message, the values have to be of the right type.
- Most settings are of type
boolean. - Setting --fractionOverEstimation is of type
double. - Setting --primaryTask is of type
String. - Settings with given options should be
String.
Table 8.17: Model settings.
| Setting | Description |
|---|---|
| --carFollowingModel | Car-following model: IDM, IDM_PLUS or IDM_PLUS_MULTI |
| --synchronization | Lane change synchronization: PASSIVE, PASSIVE_MOVING, ALIGN_GAP or ACTIVE |
| --cooperation | Lane change cooperation: PASSIVE, PASSIVE_MOVING or ACTIVE |
| --gapAcceptance | Lane change gap-acceptance: INFORMED or EGO_HEADWAY |
| --incentiveRoute | Mandatory lane change incentive for route |
| --incentiveGetInLane | Mandatory lane change incentive to join slow traffic at split and not block other traffic |
| --incentiveSpeedWithCourtesy | Voluntary lane change incentive for speed with courtesy |
| --incentiveCourtesy | Voluntary lane change incentive for cooperative lane changes |
| --incentiveQueue | Voluntary lane change incentive to join the shortest queue |
| --incentiveStayRight | Voluntary lane change incentive for trucks to stay in rightmost two lanes |
| --incentiveKeep | Voluntary lane change incentive to keep to the slow lane |
| --accelerationSpeedLimitTransition | Acceleration incentive to slow down prior to a lower speed limit |
| --accelerationTrafficLights | Acceleration incentive to approach traffic lights |
| --accelerationConflicts | Acceleration incentive to approach intersection conflicts |
| --accelerationNoSlowLaneOvertake | Acceleration incentive to not overtake traffic on the slow lane |
| --fullerImplementation | Implementation of Fuller: NONE, SUMMATIVE, ANTICIPATION_RELIANCE or ATTENTION_MATRIX |
| --primaryTask | ID of primary task under ANTICIPATION_RELIANCE |
| --anticipation | Enables temporal constant-speed anticipation |
| --estimation | Enables estimation of neighboring vehicles |
| --fractionOverEstimation | Fraction of drivers over-estimating speed and distance [0..1] |
| --carFollowingTask | Enables car-following task |
| --alternateCarFollowingTask | Enables alternate car-following task (exponential) |
| --freeAccelerationTask | Enables free acceleration task, useful when updateTimeAdaptation is true |
| --trafficLightsTask | Enables traffic light task |
| --signalTask | Enables signal task |
| --laneChangingTask | Enables lane-change task |
| --alternateLaneChangingTask | Enables alternate lane-changing task (lane change desire) |
| --cooperationTask | Enables cooperation task |
| --intersectionTask | Enables intersection task |
| --roadSideDistractionTask | Enables road-side distraction task |
| --speedAdaptation | Enables behavioral speed adaptation |
| --headwayAdaptation | Enables behavioral headway adaptation |
| --laneChangeAdaptation | Enables behavioral voluntary lane change adaptation |
| --updateTimeAdaptation | Enables behavioral update time adaptation |
| --socialPressure | Enables social pressure exerted to the (potential) leader |
| --tailgating | Enables tailgating due to social pressure towards the leader |
| --socioLaneChange | Enables lane changes due to social pressure from the (potential) follower |
| --socioSpeed | Enables speed increase due to social pressure from the follower |
Parameters¶
An overview of all supported parameters is given in Table 8.18. When parameter values are given in a vehicle message this should be in the form of a key-value pair, where the Parameter column value is the key as a String and the value is any object of the correct type as in the Type column. The Default column gives the default values for the parameters. There are a few parameters for which the OTS transceiver will assign a random value from a distribution. These are:
- vGain: LogNormal(mu = 3.379, sigma = 0.4). This value is only randomly drawn in case of social interactions. Otherwise this parameter receives the value as in the table.
- sigma: Triangular(min = 0.0, mode = 0.25, max = 1.0)
- fSpeed: Normal(mu = 123.7/120.0, sigma = 0.1)
- fSpeedGtu: Normal(mu= 85.0/80.0, sigma = 2.5/80.0). This reflects typical truck speeds.
Table 8.18: Sim0mq transceiver settings.
| Parameter | Type | Default | Description |
|---|---|---|---|
| s0 | Length |
3.0 m | Car-following stopping distance |
| a | Acceleration |
1.25 m/s2 | Maximum (desired) car-following acceleration |
| b | Acceleration |
2.09 m/s2 | Maximum comfortable car-following deceleration |
| bCrit | Acceleration |
3.5 m/s2 | Maximum critical deceleration, e.g. stop/go at traffic light |
| b0 | Acceleration |
0.5 m/s2 | Maximum adjustment deceleration, e.g. when speed limit drops |
| Tmin | Duration |
0.56 s | Minimum car-following headway |
| Tmax | Duration |
1.2 s | Maximum car-following headway |
| tau | Duration |
25.0 s | Headway relaxation time |
| t0 | Duration |
43.0 s | Look-ahead time for mandatory lane changes |
| x0 | Length |
295.0 m | Look-ahead distance |
| fSpeed | Double |
1.0 | Lane speed limit adherence factor |
| fSpeedGtu | Double |
1.0 | GTU type speed limit adherence factor |
| vCong | Speed |
60.0 km/h | Speed threshold below which traffic is considered congested |
| lcInfo | Length |
2.0 km | Length over which lane change information is considered |
| delta | Double |
4.0 | Acceleration flattening exponent towards desired speed |
| dFree | Double |
0.365 | Free lane change desire threshold |
| dSync | Double |
0.577 | Synchronized lane change desire threshold |
| dCoop | Double |
0.788 | Cooperative lane change desire threshold |
| vGain | Speed |
69.6 km/h | Anticipation speed difference at full lane change desire |
| socio | Double |
1.0 | Sensitivity level for speed of others |
| alpha | Double |
0.8 | Fraction of primary task that can be reduced by anticipation reliance |
| beta | Double |
0.6 | Fraction of auxiliary tasks that can be reduced by anticipation reliance |
| TScrit | Double |
0.8 | Critical task saturation |
| TSmax | Double |
2.0 | Maximum task saturation |
| TC | Double |
1.0 | Task capability |
| over_est | Double |
1.0 | Over estimation (1.0) or under estimation (-1.0) |
| h_exp | Duration |
3.83 s | Exponential decay of car-following task by headway |
| beta_T | Double |
1.0 | Maximum adaptation headway scaling |
| beta_v0 | Double |
1.0 | Maximum adaptation desired speed scaling |
| SAmin | Double |
0.5 | Minimum situational awareness |
| SAmax | Double |
1.0 | Maximum situational awareness |
| TRmax | Duration |
2.0 s | Maximum reaction time |
| dt_min | Duration |
0.3 s | Minimum update time |
| dt_max | Duration |
2.0 s | Maximum update time |
| beta_d | Double |
1.0 | Voluntary lane change desire scaling |
| tau_min | Duration |
0.32 s | Minimum perception delay |
| tau_max | Duration |
1.19 s | Maximum perception delay |
| td_scan | Double |
0.0279 | Scanning task demand |
| td_signal | Double |
0.2 | Signal task demand |
| x0_d | Length |
119.98 m | Distance discount range for task demand |
| td_b | Double |
0.75 | Maximum task demand due to ego distance to intersection |
| beta_yl | Length |
25.12 m | Exponential decay of conflict task by ego distance |
| td_c | Double |
0.25 | Maximum task demand due to time-to-conflict-point of conflicting vehicle |
| beta_con | Duration |
12.13 s | Exponential decay of conflict task from time-to-conflict-point |
Parameters for externally controlled vehicles¶
It is important to carefully consider what parameters to use on vehicles that are externally controlled (HYBRID or EXTERNAL mode). For HYBRID mode this influences the suggested behavior. In any case parameters determine interactions with surrounding traffic. For example, an externally controlled vehicle with a low fSpeed and high vGain value will not be perceived to exert social pressure on its leader, regardless of distance. Interactions are complex and dependent on many factors, both inter- and intra-driver. The following points are guidelines to understand and influence the interactions. Whether these interactions occur depends on the used model components. In particular whether social interactions and behavioral adaptations within the perception model are used. Furthermore interactions depend on circumstances in traffic.
- fSpeed/fSpeedGtu determine the base-level desired speed, a higher value causes more social pressure (and tailgating).
- vGain reflects a driver's insensitivity to its own speed gain, a lower value causes more social pressure (and tailgating) and less response to social pressure from a follower.
- sigma indicates sensitivity to social pressure, a higher value causes an increased desired speed and lane change desire in case of social pressure.
- Social effects on desired headway, desired speed and lane change desire may be diminished in case of task over-saturation. The strength of this diminishing is governed by circumstances and sensitivities beta_T, beta_v0 and beta_d.