Patent Application
20060025896
The present invention relates to a vehicle dynamics control system, and a method for stabilizing a vehicle in critical driving situations according to the present invention.
Vehicle dynamics control systems, such as ESP (electronic stability programs), are used for the purpose of improving the controllability of motor vehicles in critical driving situations, for example, upon oversteering when negotiating curves, and stabilizing the vehicle. Known vehicle dynamics control systems include a control unit, in which a control algorithm for executing a float angle and/or yaw rate regulation is stored, as well as an array of sensors which provide measured values about the current driving condition of the vehicle. Different setpoint variables are calculated from the driver selection, in particular the steering wheel position, the accelerator pedal position, and the brake operation. In the event of too high a deviation of the actual behavior from the setpoint behavior of the vehicle, the electronic stability program intervenes in the vehicle operation and produces a compensating yaw moment, which counteracts the yaw movement of the vehicle. For this purpose, the vehicle dynamics control system typically operates the vehicle brakes and/or the engine controller as final controlling elements.
Modern vehicles also increasingly include active rear-wheel steering systems, which may also intervene in the vehicle operation for the purpose of vehicle stabilization. Systems of this type typically include a separate control unit and a steering control element, using which the steering angle of the rear wheels may be adjusted. The control algorithm of the rear-wheel steering system typically also determines different setpoint values of driving condition variables, such as a setpoint yaw rate or a setpoint float angle, and calculates a required stabilization intervention (the superimposed steering angle) from the system deviation. The calculated steering angle changes are implemented using a steering control element and influence the driving behavior of the vehicle.
Since both the electronic stability program ESP and the active rear-wheel steering system (RWS) perform stabilization interventions, this may result in the two systems mutually impairing one another and, in the worst case, the driving safety being endangered.
It is therefore an object of the present invention to coordinate the stabilization interventions of the vehicle dynamics control system and the rear-wheel steering system.
An aspect of the present invention is providing an expanded vehicle dynamics control system (VDM), which may also address a steering control element of the rear-wheel steering system, in addition to the brake system and the engine controller, and providing this VDM system with only one single control algorithm, which generates a controller output variable (e.g., a yaw moment), from which both an actuation request for a final control element (i.e., the brake system or the engine controller) of the vehicle dynamics control system and also for the steering control element of the rear-wheel steering system are generated. This type of central regulation may be implemented particularly easily and is safe and reliable.
The corresponding control algorithm may be implemented, for example, in the control unit of the vehicle dynamics control system. The currently existing electronic stability program algorithm (ESP) must only be slightly supplemented and adapted for this purpose. Preferably, no separate electronic stability program is executed in the control unit of the rear-wheel steering system.
The expanded vehicle dynamics controller (VDM) preferably includes a distributor unit, which generates both an actuation for the brake system and the engine controller and an actuation for the steering control element of the rear-wheel steering system from the controller output variable.
The control unit of the expanded vehicle dynamics control system (VDM) and the control unit of the rear-wheel steering system (RWS) are preferably connected to a shared data bus (e.g., a chassis CAN), via which different types of steering angle information are transmitted. In addition, the VDM control unit is preferably connected to a further bus (e.g., a PT CAN), via which different types of sensor information of the VDM sensor system are transmitted in particular. Through the separate transmission of steering angle and other sensor information, the data transmission speed may be elevated and system security may be improved.
In an expanded electronic stability program (VDM), as was described above, some special control technology characteristics arise, which will be explained in greater detail in the following:
1. Adaptation of the Control Behavior of the Yaw Rate and Float Angle Regulators
Known vehicle dynamics controllers typically include a yaw rate regulator and a float angle regulator, which generate a regulator output variable as a function of their associated system deviation, from which the manipulated variable for the brake system (hydraulic modulator) and/or the engine controller (motronics) is calculated. The expanded vehicle dynamics control system (VDM) according to the present invention also calculates a manipulated variable for the rear-wheel steering system as a function of the system deviation. The following set of problems thus results in regard to the yaw rate and float angle regulation:
Basically, a control intervention of the yaw rate regulator in the rear-wheel steering simultaneously also has an influence on the float angle of the vehicle. In contrast to a braking intervention, the intervention in the rear-wheel steering causes an increase in the float angle while reducing the yaw rate. A control intervention of the float angle regulator in the rear-wheel steering, in contrast, causes an increase in the yaw rate while reducing the float angle. The interventions of the two regulators thus act in precisely opposite directions.
In both cases, the operating point of one regulator is shifted as a function of the actuator intervention of the other regulator. The two regulators may thus mutually amplify one another and impair the stability of the vehicle. It is therefore suggested that the regulation behavior of the yaw rate regulator preferably be adapted as a function of the share of the slip angle regulator in the manipulated variable for the rear-wheel steering system and the regulation behavior of the slip angle regulator be adapted as a function of the share of the yaw rate regulator. For this purpose, for example, the sensitivity of the regulators may be varied correspondingly.
To influence the regulator sensitivity, either the control threshold of the regulator may be adapted or the system deviation itself may be corrected. Thus, for example, the system deviation of the yaw rate regulator may be set as a function of the share of the float angle regulator in the manipulated variable for the rear-wheel steering system and the system deviation of the float angle regulator may be set as a function of the share of the yaw rate regulator.
A correction unit is preferably provided for adapting the system deviation, which generates a corrected system deviation from the system deviations of each of the regulators, which then forms the basis for the yaw rate and/or float angle regulation. The correction unit preferably defines a dead zone, i.e., a range of the system deviation in which the system deviation is set to zero, and a range in which the original system deviation is reduced by a predefined absolute value. The operating point displacements of one regulator because of the control interventions of the particular other regulator are thus compensated for.
2. Activation/Deactivation of the Yaw Rate Regulator in Specific Ranges as a Function of the Slip Angle
In the event of high slip angles of the rear axle wheels, only a very weak effect or even no effect may be exerted on the driving behavior of the vehicle through the change of the rear-wheel steering angle, since the tire transverse forces change only slightly at large slip angles. In the range of larger slip angles (e.g., between 3° and 5°), in particular at a low coefficient of friction (e.g., on snow), hardly any further steering effect may be achieved by turning the steering wheel further. In order to nonetheless be able to stabilize the vehicle, it is therefore suggested that in a range of larger slip angles (which may differ depending on the road surface substrate), the electronic stability program be allowed to intervene more strongly in the vehicle operation via the brake system and/or the engine controller than in a range of smaller slip angles.
According to a preferred embodiment of the present invention, the yaw rate regulator or float angle regulator is thus activated in a predefined slip angle range (in particular at large slip angles) and deactivated in another range (in particular at small slip angles), or at least its effect is reduced. According to the present invention, in particular, stabilizing interventions in the rear-wheel steering are not suppressed, since in this case the steering behavior of the vehicle would change. This would no longer be controllable by the driver or would at least place great demands on the driver. The stability interventions in the rear-wheel steering are therefore preferably not interrupted in the range of higher slip angles.
For the purpose of activating/deactivating braking interventions, for example, a clearing signal (CRS) may be generated, using which stability interventions in the brake system are permitted and/or suppressed. The clearing signal is preferably a function of the slip angle having maximum adhesion at a predefined road surface coefficient of friction. (The road surface coefficient of friction is typically estimated by the control algorithm.) The slip angles having maximum adhesion at different coefficients of friction are preferably mathematically approximated using a characteristic curve.
3. Calculating the Superimposed Steering Angle from the Regulator Output Variable
The expanded vehicle dynamics control system (VDM) preferably includes a computer unit, using which the corresponding manipulated variable (superimposed steering angle) is calculated from the component of the moment relative to the center of gravity which is to be implemented by the rear-wheel steering. In order to ensure that this manipulated variable does not assume too high or incorrect values in any case, thus endangering the driving safety, one or more of the following processing steps may be executed:
The raw value of the superimposed steering angle obtained from the conversion of torque into superimposed steering angle is preferably scaled and limited as a function of the estimated coefficient of friction. For this purpose, a device is preferably provided which defines a dead zone, i.e., sets the superimposed steering angle to zero for small steering angle changes and reduces the superimposed steering angle by a predefined value in the remaining range.
The size of the dead zone is preferably also a function of the (estimated) road surface coefficient of friction.
Through the cited measures, the robustness of the electronic stability program and the steerability of the vehicle may be improved in particular.
4. Special Embodiment of the Yaw Rate Regulator
The yaw rate regulator of the expanded VDM system is preferably implemented as a PID regulator. The stabilization behavior may thus be significantly improved in relation to a typical P regulator.
The regulating behavior of the I and D components of the yaw rate regulator also brings a certain set of problems with it, which are based in particular on the fact that the output signal of the I component must be set back to zero as rapidly as possible after an adjustment and the D component is relatively noise-sensitive. In order to prevent too strong a stabilization intervention by the I and D components of the PID regulator, the influence of the I and D regulator is preferably reduced as a function of the coefficient of friction. Disproportionately strong control interventions may thus be avoided in the event of road surfaces having low coefficients of friction in particular.
In addition, a device is preferably provided which permits relatively high manipulated variables of the I and D regulator at the beginning of a stability regulation and reduces the components of the I and D regulator as a function of the coefficient of friction after a first steering intervention. This has the advantage that, in particular during steady-state straight travel, during which the coefficient of friction of the road surface may be estimated relatively poorly by the regulator, a relatively high regulator amplification of the I and D components is first permitted at the beginning of the stability regulation and the regulator amplification is stabilized at lower regulator amplifications as a function of the coefficient of friction in the course of the stability regulation.
According to a preferred embodiment of the present invention, the regulating behavior of the PID regulator is therefore designed as a function of the coefficient of friction. In addition to the regulator amplification, other regulator parameters may also be a function of the coefficient of friction.
a, 7b shows the tire identifiers of a tire in the longitudinal and transverse directions of the tire for different road surfaces.
Furthermore, the control algorithm includes a distributor unit 6, which converts regulator output variable ΔMz into the components ΔLwHA, pWheelSet for individual subsystems 8a (rear-wheel steering system) and 8b (hydraulic unit and motronics), ΔLwHA being a superimposed steering angle (in the form of a steering angle change) for the rear-wheel steering and pWheelSet being a brake pressure for hydraulic system 15, 18.
Individual actuating requests ΔLwHA, pWheelSet are transmitted via interfaces 7a, 7b to control unit 1 of rear-wheel steering system 8a and electronic system 15 (
The structure and function of such a vehicle dynamics controller are sufficiently known from the related art (e.g., Bosch, Kraftfahrtechnisches Handbuch [Automotive Handbook], 23rd edition), so that in the following only the functions and, in particular, the differences from known regulators will be discussed: the actual values of the regulated status variables (yaw rate, float angle) are determined in “observer” 3. The setpoint values of the status variables are calculated in unit 4 for setpoint value calculation.
Higher-order state regulator 5 executes a yaw rate and float angle regulation in a known way and generates a regulator output variable ΔMz in the form of a yaw moment or a variable proportional thereto. A part of regulator output variable ΔMz is converted into a setpoint slip lambdaSo, which is supplied to lower-order braking and drive slip regulator 13. Setpoint slip lambdaSo calculated for the individual wheels is converted into corresponding manipulated variables pWheelSet, MSoEng for brake system 15, 18 and engine controller 16, 19, which set the required braking and/or driving forces at the individual wheels.
Furthermore, distributor unit 6 generates a partial center of gravity moment ΔMzx, which is to be implemented by rear-wheel steering 17, 20. This center of gravity moment ΔMzx is then converted by a computer unit 14 into a superimposed steering angle ΔLwHA. Superimposed steering angle ΔLwHA is finally added at steering control element 20 to the current rear-wheel steering angle.
The weighting of individual control components pWheelSet, MSoEng, ΔMzx calculated from regulator output variable ΔMz of state regulator 5 may basically be selected arbitrarily, depending on how strong the desired intervention of individual subsystems 8a, 8b is. Preferably, their distribution is a function of the coefficient of friction or the slip angle of the rear wheels, however.
If necessary, further signals (not shown), such as an “operating status” signal or a “status” signal may also be transmitted between control units 1 and 2 for the security software of vehicle dynamics controller 29 and a clearing software, using which vehicle dynamics controller 29 may be activated and/or deactivated.
RWS control unit 1 includes a control function 27, which calculates rear-wheel steering angle Lw_dr desired by the driver as a function of set steering wheel angle LwS and wheel velocity vWheel. As long as vehicle 10 is in a stable state, this steering angle Lw_dr is set by a steering angle regulator 28 at the rear axle. In contrast, if vehicle 10 is in an unstable situation, vehicle dynamics controller 29 generates a superimposed steering angle ΔLwHA in the form of a steering angle change which is transmitted to RWS control unit 1, where it is linked to rear-wheel steering angle Lw_dr desired by the driver. Rear-wheel steering angle LwSo resulting therefrom then forms the new setpoint value for steering angle regulator 28.
Cited steering angle information Lw_dr, LwHA, ΔLwHA is transmitted via a data bus, which is also referred to as a chassis CAN. VDM control unit 2 is additionally connected to a second data bus PT CAN, via which different sensor signals of ESP sensor system 11 are input in particular. The separate bus connection between both control units 1, 2 allows particularly rapid and reliable transmission.
The two regulators 30 and 31 receive associated system deviation evGi (yaw rate) and eBeta (slip angle) and they generate a corresponding center of gravity moment ΔMzGi or ΔMzBeta, respectively. Regulator output variables ΔMzGi and ΔMzBeta are processed in block 32, and a center of gravity moment ΔMz is generated therefrom, which is typically a setpoint yaw moment ΔMGiSo.
Finally, distributor unit 6 distributes center of gravity moment ΔMz to the individual subsystems, specifically the brake system and the engine controller (combined in block 8b) and rear-wheel steering system 8a, actuation requests being output in the form of a center of gravity moment ΔMz1 and a setpoint slip lambdaSo. Variables ΔMz1, lambdaSo are then converted into corresponding manipulated variables ΔLwHA, pWheelSet, MSoEng in blocks 13 and 14.
In an expanded electronic stability program (VDM), as was described above, some special control technology characteristics arise, which will be explained in greater detail in the following:
1. Adaptation of the Regulating Behavior of the Yaw Rate and Float Angle Regulators.
In principle, a control intervention of yaw rate regulator 30 in the rear-wheel steering also has an influence simultaneously on float angle and/or slip angle alHA of vehicle 10. In contrast to braking interventions, the intervention in the rear-wheel steering causes an increase in float angle beta and/or slip angle alHA while reducing yaw rate vGi. A regulating intervention in float angle regulator 31 on the rear-axis steering, in contrast, causes an increase in yaw rate vGi while reducing the float angle. The interventions of both regulators 30, 31 thus act in precisely opposite directions. This will become clearer from the following example:
In the event of too high a yaw rate vGi of vehicle 10, the rear wheels are influenced in the same direction as the front wheels in order to reduce yaw rate vGi. However, the float angle and slip angle alHA on the rear axle are thus increased. This means that operating point deviations occur at float angle regulator 31. This may in turn result in float angle regulator 31 intervening in the driving operation and causing a deflection of the rear wheels in the opposite direction to reduce slip angle alHA. Regulators 30, 31 may thus mutually amplify one another and endanger the driving safety.
To coordinate both regulators 30, 31, it is suggested that the regulating behavior of yaw rate regulator 30 be set as a function of the share of slip angle regulator 31 and control request ΔLwHA for the rear-wheel steering and the regulating behavior of slip angle regulator 31 be set as a function of the share of yaw rate regulator 30. A possibility for coordinating both regulators 30, 31 is illustrated in
Two correction units 26a and 26b are provided for correcting system deviations evGi and eBeta, respectively; the correction units calculate a corrected system deviation evGi, eBeta from actual system deviations evGi0, eBeta0, which are then supplied to regulators 30, 31. Correction units 26a, 26b define a dead zone ToZo, i.e., a range of the system deviation in which system deviation evGi, eBeta is set to zero, and a range in which the actual system deviation is reduced by a predefined absolute value. If actual system deviation evGi0 or eBeta0 is located within the dead zone, whose boundaries are predefined by the values ±ToZoGi and ±ToZoBeta, corrected system deviations evGi and eBeta supplied to regulators 30, 31 are set to zero. Outside the dead zone, actual system deviations evGi0 and eBeta0 are reduced by value ToZoGi and ToZoBeta, respectively. Operating point deviations of yaw rate regulator 30 and float angle regulator 31 may thus be compensated for.
The calculation of dead zones ToZoGi and ToZoBeta is schematically illustrated in
The change of yaw rate ΔvGI and the change of float angle and/or slip angle ΔalHA are calculated in blocks 22 and 23. Because of an intervention of float angle regulator 31 yaw rate deviation ΔvGI results in this case from:
ΔvGI=−ΔLwHABeta*vGiSo/(Lw−LwHA) (1)
and because of an intervention of yaw rate regulator 30 slip angle deviation ΔalHA results in:
ΔalHA=−ΔLwHAvGi (2)
In this case, Lw is the front-wheel steering angle, LwHA is the rear-wheel steering angle, and vGiSo is the yaw rate without the adjustment of float angle regulator 31.
Actuators 20 of the active rear-wheel steering typically operate very rapidly; however, the operating point deviations are not established immediately. The inertia of the actuator system and the entire vehicle may be simulated by a suitably adapted low-pass filtering 24, 25. Operating point deviations ΔvGI and ΔalHA are therefore each supplied to a low-pass filter 25a, 25b, at whose output values ToZovGi and ToZoBeta for the above-mentioned dead zones are output. Filter time constant tau is set as a variable here as a function of the curve of operating point deviations ΔvGI, ΔalHA using units 24a and 24b. In this case, different time constants tau are selected in particular for signals ΔvGI and ΔalHA, which become larger and smaller.
The operating point deviations from equations (1) and (2) may, for example, be added directly to the corresponding setpoint values. Preferably, however, the absolute value of each of operating point deviations ΔvGI and ΔalHA is determined and the value for a dead zone ToZovGi and ToZoBeta, respectively, is calculated therefrom. In this case:
ΔvGI=|LwHABeta*vGiso/(Lw−LwHA)|=ToZovGi (3)
and
ΔalHA=|ΔLwHAvGi|=ToZoBeta (4)
In this case, vGiSo is the setpoint yaw rate, Lw is the front axle steering angle, and LwHA is the rear axle steering angle.
Equations (1)-(4) may be derived from the known linear single track model. Accordingly, the following equation applies for setpoint yaw rate vGiSo:
υGiSo=((Lw−LwHA)*v)/(1*(1+(υ/υch)2)) (5)
with υ being vehicle velocity, 1 being wheelbase, and υch being characteristic velocity.
Through differentiation, the following equation results from equation (5):
ΔυGiSo=(−ΔLwHA*v)/(1*(1+(υ/υch)2)) (6)
with
υGi being yaw rate change and ΔLwHA being change of the rear-wheel steering angle.
After rearranging and equating the equations (5) and (6), the change of yaw rate ΔvGiSo as a function of a steering angle change at the rear axle results:
ΔvGiSo=−ΔLwHA*vGiSo/(Lw−LwHA) (7)
The equations of the linear single track model also provide a statement about the slip angle at the rear axle, in which the following applies:
alHA=−LwHA+Beta−vGiActual*1HA/v (8)
where
After differentiation, the following equation results for the slip angle change, i.e., float angle change ΔalHA as a result of a steering angle change ΔLwHA at the rear axle:
ΔalHA=−ΔLwHA (9)
2. Activation/Deactivation of the Yaw Rate Regulator in Specific Ranges as a Function of the Slip Angle
a shows the tire identifier (μ/slip characteristic curve) in the longitudinal direction of the tire for different road surfaces. In this case, reference numeral 61 identifies the coefficient of friction curve for a dry road surface, reference numeral 62 for a wet road surface, reference numeral 63 for snow, and reference numeral 64 for ice.
b shows the tire identifier (μ/slip angle characteristic curve) in the transverse direction of the tire for different road surfaces. In this case, reference numeral 65 identifies a dry road surface, reference numeral 66 identifies snow, and reference numeral 67 identifies ice.
Characteristic curves 65-67 are continuous, having a positive gradient starting from the origin until a maximum coefficient of friction is reached and then are essentially flat or have a negative gradient. Those slip angles alpha at which the tires have a maximum adhesion in the transverse direction are referred to in this case as alHAmax.
For stability regulation using rear-wheel steering, this characteristic curve means that at small slip angles (alpha<alHAmax) the lateral forces may be expediently modulated, while at large slip angles (alpha>alHAmax) hardly any change or no change of the tire transverse forces may be achieved through a steering angle change ΔLwHA, since the gradient of characteristic curves 65-67 is nearly zero in this range. In the range of larger slip angles, it is therefore necessary to permit stronger ESP stability interventions in brake system 15, 18 and in engine controller 16, 19. The steering intervention in rear-wheel steering 17, 20 is expressly not suppressed in this case, since an interruption of the active rear-wheel steering would result in an altered drivability of the vehicle and would irritate the driver.
The individual steps of the calculation of clearing signal CRS are illustrated in
With operational rear-wheel steering, whose status signal Stat is taken into account at node 34, function block 35 generates clearing signal CRS through simple threshold value comparison. If current slip angle alHA is greater than threshold value alHAmax, signal CRS is set to “true” and therefore ESP interventions are permitted. Otherwise, signal CRS is set to “false” and ESP interventions are thus suppressed. Signal CRS is a Boolean signal.
3. Calculating the Superimposed Steering Angle from the Regulator Output Variable
The algorithm includes a low-pass filter 36, which is implemented here as a Pt1 filter and generates a filtered moment signal ΔMzF. The low-pass filtering of the center of gravity moment change ΔMz is shown using a constant filter time constant, but may also be optionally implemented as a function of the coefficient of friction. Signal ΔMzF is converted into a raw value ΔLwHA0 for steering angle change ΔLwHA using function 37. Raw value ΔLwHA0 of the superimposed steering angle is then scaled as a function of the coefficient of friction using a function 38, smaller values ΔLwHASc basically being generated in the event of larger coefficients of friction μ at node 42. Scaling 38 allows adaptation specific to the customer and vehicle in particular.
Scaled superimposed steering angle ΔLwHASc is finally reduced using a function 40, which in turn defines a dead zone ToZo, in which superimposed steering angle ΔLwHA is set to zero. The size of dead zone ToZo is a function of the coefficient of friction, it basically being smaller at larger coefficients of friction than at smaller coefficients of friction. The functional relationship between the size of dead zone ToZo and coefficient of friction μ is predefined by a function 39. Dead zone ToZo causes robustness of the regulation against signal noise and parameter oscillations in particular.
4. Special Embodiment of the Yaw Rate Regulator
Float angle regulator 31 is implemented as P regulator 47 and generates a moment change ΔMzBeta from system deviation eBeta between actual and setpoint slip angles alHA. The regulator components of regulator 30 and 31 are processed in block 46 and a resulting moment ΔMz relative to the center of gravity is calculated. This moment ΔMz is then in turn distributed to the individual subsystems.
P components ΔMzP, eBeta, D component ΔMzD, and I component ΔMzI are added at node 57. A linkage of components ΔMzP, ΔMzD, and ΔMzI as a function of the driving situation or a calculation of amplification factors 48, 49, 55, 56 as a function of the driving situation may also be provided, for example. In this case, the cited variables may be linked as a function of the coefficient of friction, the vehicle velocity, or other status variables, for example.
The addition at node 57 results in a raw value ΔMz0, which is limited at node 58 as a function of vehicle velocity vFz. In this way, in particular at low vehicle velocities, manipulated variable ΔMz may be reduced and therefore smaller control interventions may occur. A corresponding reduction function is shown in block 52. The resulting signal is also limited by a limiting function 59 for reasons of safety. In this way, it may be ensured that impermissibly large control requests may be suppressed. The overall regulator may also be deactivated and/or cleared at node 60 using a signal F. Finally, variable ΔMz may be provided for the further computing sequence within vehicle dynamics controller 29 and, as described above, manipulated variables ΔLwHA, pWheelSet, and MSoEng may be derived for the different subsystems.
Algorithm 71 is used to reduce system deviation evGi and may, for example, include a function having a dead zone ToZo. The size of the dead zone is in turn predefined by a parameter PToZo 75. The resulting value of system deviation evGi is then linked at node 76 to a reduction factor RedID and a signal evGi′ is generated. Signal RIC also has an influence on the size of dead zone ToZo in algorithm 71. In addition, the size of dead zone ToZo may also be determined as a function of the driving condition or other influencing variables.
Subsequently, undesired signal components are filtered out of signal evGi′ using filters 77 and 79. The resulting signals are then limited to maximum values by functions 78 and 80. This is again performed for reasons of safety.
The actual regulator functions of I regulator 44 and D regulator 45 are illustrated in blocks 81 and 82. The D regulating algorithm is implemented here as a second-order low-pass filter. Regulator parameters natural frequency omega0 and damping d are supplied by block 83 and 84.
I regulator 81 is implemented here as a first-order high-pass filter. Time constant THP is variable and is predefined by block 74 as a function of signal evGi′, which is in turn a function of the coefficient of friction. Time constant THP is essentially determined as follows: a base signal for the selection of high-pass filter time constant THP is generated from signal evGi′ using absolute value calculation 73 and differentiation 72. The selection algorithm is shown as block 74. It is first queried therein whether the status of the RIC signal is active. If not, a higher value THP2 is selected for time constant THP of high-pass filter 81. If clearing signal RIC is low and/or inactive, it is checked whether signal evGi′ has a positive or negative gradient. In the case of a positive gradient, very small value THP0 is selected for time constant THP, and in the case of a negative gradient, a larger value THP1 is selected for time constant THP, with THP2>THP1>THP0.
Filter functions 77, 79 upstream from I and D regulators may, for example, have constant parameters. Alternately, it is also possible to set one or more of the filter parameters as a function of the driving situation, in particular vehicle velocity vFz, lateral acceleration ay, or another driving condition. Resulting signal components ΔMzI and ΔMzD are then supplied to vehicle dynamics controller 29 for further processing.
Since the coefficient of friction is determined from the vehicle acceleration, the signal value in the event of steady-state straight-line travel is approximately zero. Only in the event of stronger longitudinal or lateral acceleration does coefficient of friction μ assume the actual value near 1. This behavior is rather unfavorable for determining a suitable regulator amplification.
The second branch includes a function 85, using which a setpoint yaw rate vGiSo is calculated according to the linear single track model. In this model, vehicle velocity vFz and front axle steering angle lw enter as input variables. The signal of the setpoint yaw rate is then limited in block 86 as a function of the coefficient of friction via complex filtering algorithms and an output signal LimvGi is generated. The input and output signals of filter algorithm 86 are subjected in block 89 to an expanded quotient calculation, which prevents division by zero and keeps the value range from being exceeded. In case of travel on a road surface having a high coefficient of friction, the quotient from block 89 results in values near one. In block 90, these values are weighted to make calibration possible and reduction factor RedID2 is generated. Finally, in block 88, the maximum value is selected from both reduction factors RedID1 and RedID2 and output as a value RedID.
This method has the advantage in particular that at the beginning of a stability regulation, in particular starting from steady-state straight-line travel, value RedID is not too low and is stabilized at an exact function of the coefficient of friction in the course of the stability regulation. The extent of the amplification reduction is settable during the calibration of the control algorithms in this case.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2004 036 565.2 | Jul 2004 | DE | national |
