VehicleComponents

AEB

The Autonomous Emergency Braking system checks if a collision is likely to occur in the near future and, if necessary, brakes to avoid the collision. In each timestep, the system evaluates all objects detected by a Sensor and calculates the time to collision (TTC) for this object based on the perceived movement of the object. If, for any object, the TTC is lower than the threshold of the component, then the component gets activated. The system deactivates if the TTC is larger than 1,5 times the threshold of the component.

Attribute

Type

Unit

Description

CollisionDetectionLongitudinalBoundary

Double

m

Additional length added the vehicle boundary when checking for collision detection

CollisionDetectionLateralBoundary

Double

m

Additional width added the vehicle boundary when checking for collision detection

TTC

Double

s

Time to collision which is used to trigger AEB

Acceleration

Double

m/s²

Braking acceleration when activated

<ProfileGroup Type="AEB">
    <Profile Type="AEB" Name="AEB1">
        <Double Key="CollisionDetectionLongitudinalBoundary" Value="4.0"/>
        <Double Key="CollisionDetectionLateralBoundary" Value="1.5"/>
        <Double Key="TTC" Value="2.0"/>
        <Double Key="Acceleration" Value="-2"/>
    </Profile>
    ...
</ProfileGroup>

DynamicsTrajectoryFollower

This module forces agents to drive according to a specific trajectory. The trajectory is defined in the scenario. This module is disabled by default and is activated if a trajectory from openSCENARIO is triggered. It is always important that the trajectories matches the current scenery file, otherwise the Agent could be placed outside of valid lanes. If the agent gets placed on a invalid position, it will be deleted.

All attributes are required.

Attribute

Type

Description

AutomaticDeactivation

Bool

If true, the trajectory follower relinquishes control of the vehicle after the final instruction in the TrajectoryFile. If false, it stops at the last point of the trajectory.

EnforceTrajectory

Bool

If true, the trajectory follower overrides external input related to the vehicle’s travel.

<ProfileGroup Type="DynamicsTrajectoryFollower">
    <Profile Name="BasicTrajectoryFollower">
        <Bool Key="AutomaticDeactivation" Value="true"/>
        <Bool Key="EnforceTrajectory" Value="true"/>
    </Profile>
</ProfileGroup>

FMU Wrapper

The FMU Wrapper provides a connection to arbitrary FMUs (Functional Mock-up Unit). An FMU has to be compatible with the FMI 1.0 or the FMI 2.0 specification (Functional Mock-up Interface) and has to be ABI (Application Binary Interface) compatible with the opSimulation binary.

Additional reading about FMI is provided by the FMI standard website at https://fmi-standard.org/. For interfacing the FMUs in openPASS, the Modelon FMI Library is used, which is recommended on the FMI standard website. See https://jmodelica.org/.

FMU package format

FMI defines a packaging format for FMUs. The used container format is ZIP. It basically contains - among other parts - the compiled FMU code (as *.dll or *.so, depending on the platform) and the modelDescription.xml. Latter provides meta-data about the FMU, i. e. - Author information - Model name, identifier and description - Generation timestamp - Name and datatype of model variables (inputs and outputs)

Architectural overview

The wrapper is instantiated as a component of an agent. It reads the input variables for the FMU from the simulation and provides it the FMU and reads the output of the FMU and forwards it via signals to other agent components.

|op| FMU wrapper architectural overview

Framework channels

The wrapper can use input and output signals via Channels as every other agent component does. Framework channels (signals) can provide data and can also be written to by the wrapper. In addition, the wrapper is able to access the c AgentInterface and c WorldInterface methods.

FMI variables

Communication with the FMU happens via FMI variables (inputs and outputs). The wrapper will read in available variables from modelDescription.xml in the FMU package. These variables need to be mapped to variables and signals of openPASS in the VehicleComponentProfile.

FMI 1.0 supports these standard datatypes: - bool - integer - real - string

By using OSMP, three integer values can be used to support full osi messages

Configuration

Configuration of a particular FMU takes place in ProfilesCatalog.xml. An example of a static system configuration can be found here “sim/contrib/examples/Configurations/StaticOSMPSensorDataToTrafficUpdateStepper”.

The following parameters are always required for the FmuWrapper. Depending on the FmuHandler additional parameters may be needed.

Key

Type

Default

Description

FmuPath

string

-

Path to FMU file, either absolute or relative to the simulator’s configuration directory.

Logging

bool

true

If set to true, FMU initialization and execution task are logged to a text file.

CsvOutput

bool

true

If set to true, FMI outputs are logged to a CSV file.

<ProfileGroup Type="FMU1">
  <Profile Name="FMU">
    <String Key="FmuPath" Value="OSMPSDToTUS.fmu"/>
    <Bool Key="Logging" Value="true"/>
    <Bool Key="CsvOutput" Value="false"/>
    <String Key="Input_OSMPSensorDataIn" Value="SensorData"/>
    <String Key="Output_OSMPTrafficUpdateOut" Value="TrafficUpdate"/>
 <Bool Key="WriteJson_SensorData" Value="false"/>
 <Bool Key="WriteJson_TrafficUpdate" Value="false"/>
  </Profile>
</ProfileGroup>

Upon instantiation of the FMU wrapper, it will extract the FMU ZIP file to a temporary directory. Then the modelDescription.xml is parsed and the FMU is checked for compatibility.

If the parameter CsvOutput is set to true, a subfolder “FmuWrapper/Agent<ID>” will be created in the simulator’s “results” directory. “<ID>” is replaced with the agent id. FMI output data will be logged to a file inside this directory. The filename consists of the FMU’s name and extension “csv”. This output can then be used for visualization in a spreadsheet application or it may be processed in any other way.

Same goes for parameter Logging (having “log” as output file extension).

Primitive Datatypes

The FMU Wrapper allows to link Simulink models or any other FMU to openPASS. It lets the user link any input variables of the FMU to values of the Agent in the simulation and any output values of the FMU to signals, that are forwarded to other openPASS components. These mappings are defined with the following optional parameters.

Note

Be careful with the size of integer data types when used in Matlab/Simulink. The FMU integer data type shall always be 32 bit or bigger, e.g. for IDs.

Key

Type

Description

Parameter_varName

any

Mapping of a fixed value (bool, integer, double, string) to an FMU input:

  • varName references an FMI input variable.

  • The type of the parameter has to match the FMI variable type.

Parameter_AssignSpecial_varName

any

Mapping of a specific value of the simulation to an FMU input, assigned only once at FMU initialization:

  • varName references an FMI input variable.

  • The parameter is always of type string.

  • The value of this parameter has to be one of the types specified in the Special simulation values table below, which will also determine the required type of the FMI variable.

Input_varName

string

Mapping of a specific value of the simulation to an FMU input:

  • varName references an FMI input variable.

  • The value of this parameter has to be one of the types specified in the Input simulation values table below.

Output_varName

string

Mapping of a FMU output to a specific field in a specific signal:

  • varName references an FMI output variable.

  • The value of this parameter has to be one of the types specified in the Output simulation signals table below.

The allowed special simulation values are as follows:

Type

FMU Variable Type

Calculation

RandomSeed

Integer

The random seed of the current simulation run.

OutputPath

String

An output path unique to this FmuWrapper instance. The path will always refer to a directory below the simulator’s current result folder. This directory is not necessarily created by the FmuWrapper, depending on the setting of CsvOutput and Logging parameters (see FmuWrapper basic configuration).

MaxSteering

Real

The max_steering property of an agent’s front axle as defined in OpenSCENARIO (catalog).

SteeringRatio

Real

The steering ratio of the vehicle model. Has to be defined in the properties of the Entity in the OpenSCENARIO catalog with the name SteeringRatio and the value has to be a valid floating point expression.

NumberOfGears

Integer

The number of gears of the vehicle model. Has to be defined in the properties of the Entity in the OpenSCENARIO catalog with the name NumberOfGears and the value has to be a valid integer.

GearRatioN

Real

The ratio of the Nth gear. Has to be defined in the properties of the Entity in the OpenSCENARIO catalog with the name GearRatioN and the value has to be a valid floating point expression. N is allowed to be in the range 1-9.

The allowed inputs (simulation values) are as follows:

Type

FMU Variable Type

Calculation

VelocityEgo

Real

Absolute velocity (length of the velocity vector) at reference point

AccelerationEgo

Real

Longitudinal acceleration at reference point

CentripetalAccelerationEgo

Real

Centripetal acceleration at reference point

SteeringWheelEgo

Real

Angle of the steering wheel (in radian)

AccelerationPedalPositionEgo

Real

Position of the acceleration pedal in the interval [0, 1]

BrakePedalPositionEgo

Real

Position of the brake pedal in the interval [0, 1]

DistanceRefToFrontEdgeEgo

Real

Distance between the reference point and the front of the agent (static)

PositionXEgo

Real

X position of the reference point

PositionYEgo

Real

Y position of the reference point

YawEgo

Real

Yaw of the reference point

LaneEgo

Integer

Lane id of the front center on the route (0, if off route)

PositionSEgo

Real

S position of the reference point on the route (0, if off route)

PositionTEgo

Real

T position of the reference point on the route (0, if off route)

ExistenceFront

Boolean

true, if there is a object in front on the own lane (any range), false otherwise

PositionXFront

Real

X position of front object reference point (0, if no front object)

PositionYFront

Real

Y position of front object reference point (0, if no front object)

YawFront

Real

Yaw of front object reference point (0, if no front object)

PositionSFront

Real

S position of front object reference point on ego route (0, if no front object)

PositionTFront

Real

T position of front object reference point on ego route (0, if no front object)

RelativeDistanceFront

Real

Net distance to front object along route (0, if no front object)

WidthFront

Real

Width of front object (0, if no front object)

LengthFront

Real

Length of front object (0, if no front object)

DistanceRefToFrontEdgeFront

Real

Distance between the reference point and the front of the front object (0, if no front object)

VelocityFront

Real

Absolute velocity of front object at reference point (0, if no front object)

LaneFront

Integer

Lane id of the reference point of the front object on the ego route (0, if no front object)

ExistenceFrontFront

Boolean

true, if there are at least two objects in front on the own lane (any range), false otherwise

PositionXFrontFront

Real

X position of second front object reference point (0, if no second front object)

PositionYFrontFront

Real

Y position of second front object reference point (0, if no second front object)

RelativeDistanceFrontFront

Real

Net distance to second front object reference point (0, if no second front object)

VelocityFrontFront

Real

Absolute velocity of second front object reference point (0, if no second front object)

LaneFrontFront

Integer

Lane id of the reference point of second front object reference point (0, if no second front object)

LaneCountLeft

Integer

Number of lanes to the left of front center of type Driving, Exit, Entry, OnRamp or OffRamp

LaneCountRight

Integer

Number of lanes to the right of front center of type Driving, Exit, Entry, OnRamp or OffRamp

SpeedLimit_X

Real

Speed limit in effect in distance X meters from front center (999, if no speed limit)

RoadCurvature_X

Real

Road curvature in distance X meters from front center

  • reference point: Center of the rear axle

  • front center: Center of the front of the bounding box of the object

If the FmuWrapper is linked to at least one sensor with InputId “Camera”, the following additional inputs are available. The objects seen by this sensor(s) are sorted by distance from the agent and accessed by indices starting from 0. For each object the values listed in the following table are available where X is the index of the object (between 0 and 9). If there are less objects than X, a default value is set (-1 for the Id, 0 for the other values). Only the list of objects is taken from the sensor. The values are then calculated by the FmuWrapper (not from the SensorData).

Type

FMU Variable Type

Calculation

SensorFusionObjectId_X

Integer

Id of the object

SensorFusionNumberOfDetectingSensors_X

Integer

Number of sensors detecting the object

SensorFusionRelativeS_X

Real

Distance between reference points along route (NaN, if object not on route)

SensorFusionRelativeNetS_X

Real

Net distance along route (NaN, if object not on route)

SensorFusionRelativeT_X

Real

Lateral obstruction for front center (NaN, if object not on route) (see GetObstruction)

SensorFusionRelativeX_X

Real

Relative distance between reference points in x in world coordinates

SensorFusionRelativeY_X

Real

Relative distance between reference points in y in world coordinates

SensorFusionRelativeNetLeft_X

Real

Lateral obstruction for leftmost point (NaN, if object not on route)

SensorFusionRelativeNetRight_X

Real

Lateral obstruction for rightmost point (NaN, if object not on route)

SensorFusionRelativeNetX_X

Real

Net distance between bounding boxes in x in world coordinates

SensorFusionRelativeNetY_X

Real

Net distance between bounding boxes in y in world coordinates

SensorFusionLane_X

Integer

Lane of front center

SensorFusionVelocity_X

Real

Absolute velocity at reference point

SensorFusionVelocityX_X

Real

Velocity in x at reference point in world coordinates

SensorFusionVelocityY_X

Real

Velocity in y at reference point in world coordinates

SensorFusionYaw_X

Real

Yaw in world coordinates

The FMU wrapper can output one or more of these signals: AccelerationSignal, LongitudinalSignal, SteeringSignal and DynamicsSignal

The name of the signal field has to be specified after the signal name. This means the output type is one of the following:

Type

FMU Variable Type

Enum Values

ComponentState

Enum

Undefined, Disabled, Armed, Acting

AccelerationSignal_Acceleration

Real

LongitudinalSignal_AccPedalPos

Real

LongitudinalSignal_BrakePedalPos

Real

LongitudinalSignal_Gear

Int

SteeringSignal_SteeringWheelAngle

Real

DynamicsSignal_Acceleration

Real

DynamicsSignal_Velocity

Real

DynamicsSignal_PositionX

Real

DynamicsSignal_PositionY

Real

DynamicsSignal_Yaw

Real

DynamicsSignal_YawRate

Real

DynamicsSignal_YawAcceleration

Real

DynamicsSignal_SteeringWheelAngle

Real

DynamicsSignal_CentripetalAcceleration

Real

DynamicsSignal_TravelDistance

Real

CompCtrlSignal_MovementDomain

Enum

Undefined, Lateral, Longitudinal, Both

CompCtrlSignal_WarningActivity

Bool

CompCtrlSignal_WarningLevel

Enum

INFO, WARNING

CompCtrlSignal_WarningType

Enum

OPTIC, ACOUSTIC, HAPTIC

CompCtrlSignal_WarningIntensity

Enum

LOW, MEDIUM, HIGH

CompCtrlSignal_WarningDirection

Enum

If one of these fields of a signal (except ComponentState) is mapped to an FMU variable, all fields of this signal have to be mapped. If the ComponentState is mapped to a FMU variable, it is used for all signals, otherwise it defaults to Acting.

OSI Data

OSMP (OsiSensorModelPackaging) is a package layer specification for the Open Simulation Interface (OSI). It allows to pass input to the FMU as OSI messages as well as receive output as OSI message. For more information on OSMP see https://github.com/OpenSimulationInterface/osi-sensor-model-packaging.

The FmuHandler has the following additional (optional) parameters:

Key

Type

Description

Init_var_name

string

var_name references an FMU variable (as defined in FMU’s modelDescription.xml) to which a specific OSI message is sent during initialization Allowed values: GroundTruth

Input_var_name

string

var_name references an FMU variable (as defined in FMU’s modelDescription.xml) to which a specific OSI message is sent Allowed values: SensorView, SensorViewConfig, SensorData, TrafficCommand

Output_var_name

string

var_name references an FMU variable (as defined in FMU’s modelDescription.xml) from which a specific OSI message is received Allowed values: SensorViewConfigRequest, SensorData, TrafficUpdate

Parameter_var_name

any

The value of the parameter is assigned to the FMU variable var_name

Parameter_transformation[mapping ]_name

string/string/any*

Same as Parameter_name but with an preceding transformation according to a mapping.
Currently, only mappings between the same types are supported.
*;When using TransformList as transformation, the type of the data is expected to be a string and the string must be a comma separated list of values.

Allowed values:
transformation: Transform, TransformList
mapping: ScenarioName>Id

Example: Parameter_TransformList[ScenarioName>Id]_*name*

WriteJson_var_name

bool

If true the osi message specified by var_name is written to a json file

WriteTrace_var_name

bool

If true the osi message specified by var_name is written to the trace file

EnforceDoubleBuffering

bool

If true the wrapper will throw an error if FMU doesn’t use double buffering. Defaults to false.

The type of OSI messages the FmuHandler sends and receives is defined by its parameters. Only messages for which an FMU variable is given in the configuration are sent/received. An additional parameter defines whether the message should be logged as JSON file for every agent and every timestep (see table above).

Currently these messages are supported:

  • SensorView: SensorView generated from the GroundTruth with this agent is host vehicle.

  • SensorViewConfig, SensorViewConfigRequest: Configuration of a sensor according to OSMP.

  • TrafficCommand: Trajectory from openSCENARIO, that will be converted into a TrafficCommand.

  • SensorData: Output of a sensor. Can be input and/or output of an FMU. Received SensorData is forwarded to other components as SensorDataSignal.

  • TrafficUpdate: Will be converted to a DynamicsSignal.

  • GroundTruth: Will be used as groundtruth information for everything that exists in the simulation world.

FmuVariables

FmuVariables can have different variability and causality.

  • There are the following causalities: Input, Output, Parameter and CalculatedParameter

  • Input or outputs can have the variability constant, fixed, discrete or continuous

  • Parameter or CalculatedParameter can have the variability constant, fixed or tunable

In openPASS we have the initialization phase, which is only called once. In that phase first readValues is called. Then parameter values are synchronized between different config files. The following priority is used for the synchronization:

modelDescription < systemConfig

Afterwards still in the initialization phase writeValues is called. During the whole simulation in openPASS all the trigger functions are called each time step. For the FMU component we call WriteValues before trigger and readValues afterwards. For the first two time steps we have the following calls for read- and writeValues:

  1. ReadValues (Init)

  2. WriteValues (Init)

  3. WriteValues

  4. Trigger

  5. ReadValues

  6. WriteValues

  7. Trigger

  8. ReadValues

Depending on them, FmuVariables are written/read to/from the FMU on different occasions, which is shown in the following table.

Write- and ReadValues depending on variability and causality

ReadValues - Init

WriteValues - Init

WriteValues - Trigger

ReadValues - Trigger

Input - fixed

x

x

Input - discrete or continuous

x

x

x

Output - fixed

x

Output - discrete or continuous

x

x

Parameter - fixed

x

x

Parameter - tunable

x

x

x

CalculatedParameter - fixed

x

CalculatedParameter - tunable

x

x

Note

For alle kinds of Input, Output, Parameter and CalculatedParameter there exists the variablity “constant”. All constant values are only read into openPASS once at the initialization phase. Never will these values be written onto the FMU.

SensorGeometric2D

This sensor is selected, when a sensor is parameterized as ProfileGroup “Geometric2D”.

Parameter

Type

Unit

Description

DetectionRange

Double

m

Detection range

EnableVisualObstruction

Bool

Activates 2D sensor obstruction calculation

FailureProbability

Double

Probability object is not detected although it is visible

Latency

Double

s

Delay the sensor output

DetectionDelayTime

Double

s

Time an object needs to be in detection range before it is detected (optional)

MaxDropOutTime

Double

s

Time after which delay for undetected object starts anew (optional)

OpeningAngleH

Double

rad

Horizontal opening angle

RequiredPercentageOfVisibleArea

Double

Required percentage of an object within the sensor cone to trigger a detection

<ProfileGroup Type="Geometric2D">
  <Profile Name="Standard">
    <Double Key="DetectionRange" Value="300"/>
    <Bool Key="EnableVisualObstruction" Value="false"/>
    <Double Key="FailureProbability" Value="0"/>
    <NormalDistribution Key="Latency" Max="0.0" Mean="0.0" Min="0.0" SD="0.0"/>
    <Double Key="OpeningAngleH" Value="0.35"/>
    <Double Key="RequiredPercentageOfVisibleArea" Value="0.001"/>
    <Double Key="DetectionDelayTime" Value="0"/>
    <Double Key="MaxDropOutTime" Value="0"/>
  </Profile>
</ProfileGroup>

Note

Sensors also need a mounting position, defined w.r.t. the coordinate system of the vehicle (center of rear axis). See also VehicleProfiles.

ReceiverCar2X

This type is selected, when a sensor is parameterized as ProfileGroup “ReceiverCar2X”.

Parameter

Type

Unit

Description

FailureProbability

Double

Probability object is not detected although it is visible

Latency

Double

s

Sensor latency

Sensitivity

Double

W/m²

Sensitivity of the sensor

<ProfileGroup Type="ReceiverCar2X">
  <Profile Name="Standard">
    <NormalDistribution Key="Latency" Mean="0.0" SD="0.0" Min="0.0" Max="0.0"/>
    <Double Key="FailureProbability" Value="0"/>
    <Double Key="Sensitivity" Value="1e-5"/>
  </Profile>
</ProfileGroup>

The moving object is detected if the received SignalStrength is greater than the Sensitivity of the ReceiverCar2X .

The received SignalStrength is calculated by ss_{\text{received}} = \frac {ss_{\text{sender}}} {4 \cdot pi \cdot d \cdot d}, where the symbols meanings are:

Symbol

Description

ss_{\text{received}}

Received strength of the signal [1/m²]

ss_{\text{sender}}

Sent strength of the signal [1/m²]

d

Distance between the receiver agent and sender [m]

Note

Sensors also need a mounting position, defined w.r.t. the coordinate system of the vehicle (center of rear axis). See also VehicleProfiles.

VehicleDynamics

Components of this group can be used to model the vehicle dynamics. The vehicle dynamics model has a modular design. If necessary, the individual components can be replaced by the user with their own models. The vehicle dynamics model consists of six components listed below:

Component

Short Description

ActionSteeringSystem

The steering model transfers the driver’s input into the vehicle’s wheel angle

ActionPowertrain

The powertrain model converts the accelerator pedal position into wheel drive torques, under consideration of the selected gear

ActionBrakeSystem

The brake model converts the brake pedal position into wheel brake torques

DynamicsTireModel

The tire model converts the predetermined drive and braking torques of the tires into tire longitudinal and lateral forces, under consideration of the wheel angles

DynamicsMotionModel

The motion model calculates the translational and rotational vehicle movement with the calculated tire forces

DynamicsChassis

The chassis model determines the dynamic wheel loads via the vehicle’s longitudinal and lateral acceleration

The following figure gives an overview of the driving dynamics components and their signals:

|op| Vehicle dynamics overview

ActionSteeringSystem

The steering model obtains the “SteeringRatio” property from the VehicleCatalog and uses it to calculate the steering angle of the front wheels. Both wheels are turned at the same angle. Steering elasticities are currently not taken into account. The following parameter can be used to set a static toe:

Attribute

Type

Unit

Description

StaticToe

VectorDouble

rad

Static toe of the wheels (A positive value corresponds to a toe-in; wheels are indexed from the front left in the vector)

ActionPowertrain

The powertrain model contains an engine model and a gear model. The type of the powertrain can be set using the following parameters:

Attribute

Type

Unit

Description

TypeDrivetrain

String

Type of drivetrain; A selection can be made between front-wheel drive (FWD), rear-wheel drive (RWD) and all-wheel drive (AWD)

FrontRatioAWD

Double

Distribution of the drive torque to the front axle in the case of all-wheel drive (AWD); Range 0-1

The wheel speed is converted into an engine speed [Hz] according to the axle ratio and the transmission ratio of the selected gear. The axle ratio and gear ratios are obtained from the VehicleCatalog (“AxleRatio” & “GearRatio”). The average value of the powered wheels is used for the determination of the engine speed.

\omega_{engine} = \omega_{wheels,avg} \cdot  i_{axle} \cdot i_{gear,selected}

The maximum possible engine torque [Nm] is limited by the engine power [W] or the maximum engine torque [Nm]. The engine power and the maximum engine torque are obtained from the VehicleCatalog (“MaximumEnginePower” & “MaximumEngineTorque”).

M_{engine,max,current} = \begin{cases}
 \frac{P_{engine,max}}{\omega_{engine}}     & \text{ if } \frac{P_{engine,max}}{\omega_{engine}} < M_{engine,max} \\
 M_{engine,max}  & \text{ if } \frac{P_{engine,max}}{\omega_{engine}} >= M_{engine,max}
     \end{cases}

When 98% of the maximum speed of the motor is reached (“MaximumEngineSpeed” in the VehicleCatalog), the engine torque is linearly reduced to 0.

The maximum engine torque is scaled via the accelerator pedal position (input). This value is calculated back to the total wheel drive torque via the gear ratio.

M_{wheels,current} = M_{engine,max,current} \cdot  position_{accelerator pedal} \cdot  i_{axle} \cdot  i_{gear,selected}

The wheel total drive torque is evenly distributed to the wheels of an axle according to the definition of the drive type. With all-wheel drive, the entire wheel drive torque is distributed statically over the defined ratio.

ActionBrakeSystem

The brake model is a linearized model. The brake pedal position is used as input. As output, the model returns the braking torques of the wheels as a vector. The model considers a response time [ms] and linear factors [m/s³] for the increase and decrease of the braking force.The distribution of braking force between the front and rear axles can be defined statically.

Attribute

Type

Unit

Description

FrontAxlePercentage

Double

Distribution of the brake torque to the front axle in the case of all-wheel drive (AWD); Range 0-1

BrakeDecelerationInclineRate

Double

m/s³

Linear Rate of braking force increase

BrakeDecelerationDeclineRate

Double

m/s³

Linear Rate of braking force decrease

BrakeResponseTimeMs

Double

ms

Brake response time

The maximum braking force of the system is determined from the maximum possible deceleration and the mass of the vehicle and is scaled by the brake pedal position (Input). The maximum possible deceleration and the vehicle mass are obtained from the VehicleCatalog (“maxDeceleration” & “mass”).

F_{brake,max} = a_{deceleration,max} \cdot m_{vehicle} \cdot  position_{brake pedal}

When the brake is applied, a deceleration is calculated after the response time has elapsed. Then the braking force is built up linearly until the maximum or requested braking force has been reached.

F_{brake,current} = rate_{incline} \cdot (t_{brake} -  t_{response})

When the brake is released, the braking force is dissipated with the decline rate until it has dropped to zero. After that, the response time builds up again. The braking force is divided among the axles according to the parameter “FrontAxlePercentage”. Another input allows you to request a prefill that reduces the response time without braking

DynamicsChassis

The chassis model determines the vertical forces of the four wheels from the longitudinal and lateral acceleration of the vehicle. Constant spring and damper rates are taken into account, which can be defined by the following parameters per axis:

Attribute

Type

Unit

Description

SpringCoefficient

VectorDouble

N/m

Constant spring coefficient for each axis

DamperCoefficient

VectorDouble

Ns/m

Constant damper coefficient for each axis

DynamicsTireModel

The tire model is freely configurable and includes a degressive behaviour. The tire forces are modeled according to Rill using the TMEasy model. The following parameters can be set for the tire model per axis :

Attribute

Type

Unit

Description

MuTireMaxXFRef

VectorDouble

Normalized scaling factor for maximum longitudinal force at reference vertical force

MuTireMaxX2FRef

VectorDouble

Normalized scaling factor for maximum longitudinal force at double reference vertical force

MuTireSlideXFRef

VectorDouble

Normalized scaling factor for sliding longitudinal force at reference vertical force

MuTireSlideX2FRef

VectorDouble

Normalized scaling factor for sliding longitudinal force at double reference vertical force

SlipTireMaxXFRef

VectorDouble

Longitudinal slip at maximum longitudinal force at reference vertical force

SlipTireMaxX2FRef

VectorDouble

Longitudinal slip at maximum longitudinal force at double reference vertical force

SlipTireSlideXFRef

VectorDouble

Longitudinal slip at sliding longitudinal force at reference vertical force

SlipTireSlideX2FRef

VectorDouble

Longitudinal slip at sliding longitudinal force at double reference vertical force

F0pXFRef

VectorDouble

N

Initial slope of longitudinal force at reference force

F0pX2FRef

VectorDouble

N

Initial slope of longitudinal force at double reference force

MuTireMaxYFRef

VectorDouble

Normalized scaling factor for maximum lateral force at reference vertical force

MuTireMaxY2FRef

VectorDouble

Normalized scaling factor for maximum lateral force at double reference vertical force

MuTireSlideYFRef

VectorDouble

Normalized scaling factor for sliding lateral force at reference vertical force

MuTireSlideY2FRef

VectorDouble

Normalized scaling factor for sliding lateral force at double reference vertical force

SlipTireMaxYFRef

VectorDouble

Lateral slip at maximum lateral force at reference vertical force

SlipTireMaxY2FRef

VectorDouble

Lateral slip at maximum lateral force at double reference vertical force

SlipTireSlideYFRef

VectorDouble

Lateral slip at sliding lateral force at reference vertical force

SlipTireSlideY2FRef

VectorDouble

Lateral slip at sliding lateral force at double reference vertical force

F0pYFRef

VectorDouble

N

Initial slope of lateral force at reference force

F0pY2FRef

VectorDouble

N

Initial slope of lateral force at double reference force

FRef

VectorDouble

N

Vertical reference force for the tire parameters

FRefNormalized

VectorBool

Should the reference force be scaled with the static vertical tire force?

Inertia

VectorDouble

kgm²

Inertia of tire

PneumaticTrail

VectorDouble

m

Pneumatic trail of tire

The normalized factors refer to the reference vertical force or to the double reference vertical force The input variables used by the model are tire drive and braking torques as well as the wheel angles and vertical wheel forces. All data is provided as vectors. The model determines tire forces in the longitudinal and lateral directions as well as the wheel self aligning torques. The wheel self aligning torque is formed from the product of the tire side force and the pneumatic trail. A linear interpolation is performed between the values for the reference force and the double reference force. If no degressive tire behavior is desired, the parameters for the double reference force must be set identically to the values for the reference force.

All forces are scaled with the coefficient of friction from the VehicleCatalog (“FrictionCoefficient”).

All further information about the model can be found in the following sources:

https://www.tmeasy.de/

Rill, Georg. (2013). TMeasy – A Handling Tire Model based on a three-dimensional slip approach.

DynamicsMotionModel

The motion model converts the tire forces (input) into a translational and rotational movement of the vehicle. The air resistance of the vehicle is taken into account. For the dynamic calculation, the center of gravity position is taken from the VehicleCatalog (“XPositionCOG”,”YPositionCOG”), which indicates the distance of the center of gravity to the center of the rear axle. If this data is not given, the center of gravity is positioned on half wheelbase. For air resistance, the properties “AirDragCoefficient” & “FrontSurface” from the VehicleCatalog are used.

For the equations of motion, see relevant vehicle dynamics books such as:

Kücükay, Ferit (2022), “Grundlagen der Fahrzeugtechnik”, page 1067 ff