Method for performing roll stabilization of a vehicle
By using estimated future measurement signals and dynamic filtering, the method improves vehicle roll stabilization by compensating for actuator dynamics and noise, enhancing driving comfort and dynamic behavior.
Patent Information
- Application Number
- DE102016104944
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2016-03-17
- Publication Date
- 2026-02-05
- Estimated Expiration
- 2036-03-17
AI Technical Summary
Existing methods for vehicle roll stabilization are limited by actuator dynamics and measurement noise, which affect the dynamic behavior and driving comfort.
The method employs estimated future measurement signals, particularly wheel height and body height, using an observer and Kalman filter to compensate for actuator limitations and reduce noise, enabling dynamic filtering and improved roll stabilization.
This approach enhances the dynamic behavior and driving comfort by compensating for actuator limitations and reducing measurement noise, particularly in high-frequency excitation scenarios.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
The invention relates to a method and a device for performing roll stabilization of a vehicle.In the prior art, it is known to perform roll stabilization on a vehicle. A corresponding method is known, for example, from DE 102 54 392 A1, which describes a method and a device for optimum driving dynamics control for a vehicle. The device comprises at least one actuator, by means of which a manipulated variable is applied, wherein a controller gain for a basic system model of the vehicle is determined, wherein deviations of at least one variable system parameter of the vehicle from a basic value are detected during operation, wherein the system parameter is identified and an identified system model is determined and the controller gain is adapted to the identified system model.DE 10 2015 205 369 A1, DE 10 2006 039 353 A1 and DE 102 21 718 A1 each disclose the subject matter of the preamble of claim 1.The method known from DE 10 2013 018 923 A1 is provided for controlling at least one actuator of a chassis of a motor vehicle. In this case, values of at least one measured variable for detecting the surroundings of the motor vehicle are ascertained by at least one sensor of the motor vehicle. The values determined are preprocessed. Furthermore, during preprocessing of the values, among other things, only those values required for controlling the actuator are determined. Only those values which are required or are required for controlling the actuator are transmitted from a control device assigned to the sensor to a control device assigned to the actuator. In addition, only the values provided for monitoring the actuator are filtered in time by the monitoring device. The control device provides actuating signals for the actuator from the values which are initially preprocessed and then temporally filtered.The object of the invention is to provide an improved method and an improved device for performing roll stabilization.The object of the invention is achieved by the method according to claim 1 and by the apparatus according to claim 10.An advantage of the described method and the described device is that a dynamic behavior of the roll stabilization is improved.It has been found that by using estimated values, the measurement signals of the vehicle leading in time, in particular a height level of a wheel and / or a height level of a vehicle body, can be used in order to be able to compensate for limited actuator dynamics of the actuator which is used for roll stabilization.For this purpose, a method for carrying out roll stabilization of a vehicle is provided, wherein a measurement signal of the vehicle is detected, wherein an estimated value for the measurement signal is determined for a future point in time with the aid of the measurement signal, wherein the estimated value is taken into account in the roll stabilization.In one embodiment, the roll stabilization is carried out with the aid of a control method, wherein the estimated value is used for ascertaining a controlled variable.In one embodiment, a wheel stroke and / or a wheel acceleration and / or a wheel carrier acceleration are used as the measurement signal, wherein a future wheel stroke and / or a future wheel acceleration and / or a future wheel carrier acceleration are determined as the estimated value.In one embodiment, the determination of the estimated value is carried out with the aid of an observer.In one embodiment, a dead time for a measurement signal is taken into account in the determination of the estimated value.In one embodiment, a noise signal is taken into account for a detection of the measurement signal when ascertaining the estimated value.In one embodiment, the estimate is determined using a Kalman filter.According to the invention, the estimated value is subjected to filtering, wherein the filtered estimated value is used for roll stabilization.The filtering is carried out as dynamic filtering, wherein the dynamic filtering brings about a compensation of a limited dynamic of an actuator of the roll stabilization. For this purpose, the dynamic filtering includes an inversion of a temporal dynamic of the actuator.In one embodiment, the estimate represents wheel acceleration. In addition, the estimated value can represent a wheel travel.By using the estimated future measurement signals, a transmission time of the measurement signals from the measurement to the consideration in the roll stabilization can also be compensated.For estimating the measurement signals leading in time, an observer, in particular a Kalman filter, can be used. In this case, noise components can be reduced substantially without phase distortion. In addition, speed signals for a wheel and / or a body can be determined substantially without phases.The invention is explained in more detail below with reference to the figures. It shows FIG. 1 is a schematic illustration of a vehicle, FIG. 2 shows a schematic illustration of an axle of a chassis of the vehicle of FIG. 1, FIG. 3 shows a schematic illustration of essential elements of the method for roll stabilization, FIG. 4 shows a schematic illustration for a model of the Kalman filter, and FIG. 5 shows a schematic illustration of signal profiles of measurement signals at the output of the Kalman filter.The driving behavior of a motor vehicle is determined by a plurality of forces and moments which act on the motor vehicle in the direction of motor vehicle axles, namely in the direction of a longitudinal axis, a transverse axis, a vertical axis, and a roll axis. The roll axis is understood to mean an axis which extends through the roll centers of the front axle and the rear axle of the vehicle in the longitudinal direction of the vehicle. A rolling movement of the vehicle about the roll axis is referred to as rolling.FIG. 1 shows a schematic illustration of a vehicle 10 which has a chassis 11 and a body 12. Of the vehicle 10, a front axle 13 and a rear axle 14 are shown as essential parts. The front axle 13 has front wheels 15. The rear axle 14 has rear wheels 16. The front wheels 15 and the rear wheels 16 are respectively fixed to wheel suspensions 17. Between the wheel suspensions 17 for the front wheels 15 there are provided a first and a second stabilizer 18, 19. An actuator 20 is arranged between the stabilizers 18, 19. Between the wheel suspensions 17 of the rear wheels 16 there is arranged a first and a second stabilizer 18, 19, respectively. An actuator 20 is likewise provided between the stabilizers 18, 19 of the rear axle 14. The actuators 20 can be designed as electrical, hydraulic and / or piezoelectric actuators.FIG. 2 shows a schematic illustration of the structure of the front axle 13. In the example shown, the actuator 20 is operatively connected to the first stabilizer 18. The stabilizer 18 is rotatably attached to the body 12 via a first bearing 22. In addition, one end of the first stabilizer 18 is operatively connected to the wheel suspension 17. Depending on the rotational position of the first stabilizer 18, the wheel suspension 17 is changed in the height position with respect to the vehicle body 12. A front wheel is rotatably mounted on the wheel suspension 17.The actuator 20 is also connected to the second stabilizer 19. The second stabilizer 19 is rotatably attached to the body 12 via a second bearing 24. One end of the second stabilizer 19 is operatively connected to a second wheel suspension 17. Depending on the rotational position of the second stabilizer 19, a height position of the second wheel suspension 17 relative to the vehicle body 12 is changed. A second front wheel is rotatably mounted on the second wheel suspension 17. The first and the second stabilizer 18, 19 have a bend 25, 26 between the respective bearing 22, 24 and the end of the stabilizer or the associated wheel suspension. Due to the bending 25, 26, a rotation of the stabilizers 18, 19 leads to a change in the height position of the wheel suspensions 17.The actuator 20 is designed to exert a force forming a roll moment on the body 12 and the wheel suspensions 17 via a rotation of the first and / or the second stabilizer 18, 19. Depending on the selected embodiment, both stabilizers 18, 19 can be rotated in opposite directions by actuator 20 at the same time, so that the wheel suspension of first stabilizer 18 experiences, for example, a force that exerts a rolling moment that causes upward in the direction of body 12. At the same time, the wheel suspension of the second stabilizer 19 can be subjected to a force which is subjected to a rolling moment which leads downward away from the body 12. Moreover, depending on the embodiment chosen, only one of the stabilizers can also be rotated in the angular position. The actuator 20 can be designed in the form of an angle-controlled actuating unit. For example, the actuator 20 can be designed as a BLDC motor with a planetary gear.A control unit 30 is provided for the actuation of the actuator 20. The control device 30 is able to detect a height position of the wheel, a relative wheel travel of the wheel with respect to the vehicle body, a wheel acceleration of the wheel and / or a body acceleration of the vehicle body via sensors 31, 32. For improving driving comfort, the control device 30 regulates actively acting wheel excitations with the aid of the actuator 20. For this purpose, height level signals of the wheels, a relative wheel travel of the wheels, a wheel acceleration of the wheels and / or a body acceleration of the vehicle body can be used as measurement signals. The control device has characteristic curves, tables or programs which define a corresponding actuation of the actuator 20 as a function of a measurement signal.With the aid of the following method, measurement noise of the measurement signals and / or a dead time of the measurement signals can be reduced, in particular avoided. Furthermore, limited system dynamics will be compensated for by the following method. The frequency is usually between the natural roll frequency of the vehicle body and the natural wheel frequency. With the aid of the proposed method, the driving comfort can be significantly improved, in particular in the case of short-wave, high-frequency excitation.FIG. 3 shows a schematic illustration of functions of the proposed method which are carried out by the control unit 30 with the aid of the actuator. Depending on the embodiment chosen, individual or all functions of FIG. 3 can also be executed by a further control unit. The program block 33 represents a recording of the measurement signals and / or a processing of the measurement signals. In program block 33, for example, a height position of the wheel, a relative wheel travel of the wheel with respect to the vehicle body, a wheel acceleration of the wheel and / or a body acceleration on the side of the wheel is detected with the aid of sensors 31, 32. The measurement signals can be detected for the wheels of the front axle and / or the wheels of the rear axle.Depending on the embodiment chosen, however, the measurement signals can also be processed to further measurement signals at program block 33. For example, a difference value for the height position between the right and left wheels, a difference value for the relative wheel travel of the right and left wheels with respect to a starting position or with respect to the vehicle body, a difference value for the wheel acceleration between the right and left wheels, a difference value for the body acceleration between the right and left sides of an axle can be calculated.At least one measurement signal and / or one further measurement signal are transferred to the second program block 34. In the second program block 34, a future estimated value is determined for at least one measurement signal or for at least one further measurement signal. A time difference that the estimated value leads the measurement time can be predefined or individually selected. Depending on the present situation, different time differences may be advantageous. The second program block 34 can be designed as an observer for the measurement signals and / or the further measurement signals. The second program block determines a future estimated value for at least one measurement signal and / or for at least one further measurement signal. For this purpose, a Kalman filter can be used, for example. In addition, the second program block 34 can be designed to reduce a noise component of the measurement signal.The estimated value for the measurement signal and / or the further measurement signal is subsequently passed on to a third program block 35. The third program block 35 represents a prefilter which is designed as a dynamic prefilter. Furthermore, the dynamic prefilter has an inversion of the actuator dynamics, namely an inversion of the dynamics of the control method for the roll stabilization.The dynamic prefilter can have the following transfer function, for example: K represents the amplification factor of the control behavior of the control of the roll stabilization carried out in the fourth program block 36. T represents the time constant, d represents the attenuation and s represents the complex frequency range. T E represents a second predetermined time constant.The third program block 35 passes on the prefiltered measured value to the fourth program block 36. The fourth program block 36 carries out a control for the wall stabilization. For this purpose, a setpoint value for the regulation of the roll stabilization is determined on the basis of the received estimated value for the measurement signal. The estimated value of the measurement signal can itself be used as a setpoint value. In addition, a setpoint value can be determined with the aid of the estimated value of the measurement signal on the basis of characteristic curves, tables and / or formulas. With the aid of the setpoint value, regulation of the roll stabilization is carried out.The control of the fourth program block 36 can have the following temporal behavior, for example:T describes a predetermined time constant, s describes the complex frequency range, d describes the attenuation, and K describes the gain of the control.However, other temporal control behavior can also be provided.FIG. 4 shows a schematic illustration of a model for the design of the Kalman filter. The detected and / or determined measurement signal or further measurement signal, which represents a relative acceleration ra of the wheel, for example, is integrated in a first integrator 41 with respect to time to a relative speed rv. The relative speed rv is integrated in a second integrator 42 to form a relative wheel travel rs. In addition, an estimated value Sra for the relative acceleration is determined by means of a calculation block 43. For this purpose, the function e Tas is used. In addition, a dead time Tt is optionally added up with the aid of a first adder 44. The dead time Tt has been experimentally determined and represents a time that passes between the detection of the measurement signal and the use of the measurement signal in the roll stabilization. Subsequently, a noise signal Rs is subtracted by means of a second adder 45, which noise signal Rs is superimposed on the measurement signal. The noise signal Rs was also determined experimentally. Thus, an estimated value 46 for a future relative acceleration of the wheel is obtained. It is possible to dispense with both taking account of the dead time Tt and taking account of the noise signal Rs.Based on the relative wheel path rs, a future estimated value Srs for the relative wheel path is calculated in a second calculation block 46 using a second function. The estimated value of the relative wheel path rs can be reduced by a noise signal at a third adder 47. Thus, an estimate of the relative wheel travel is obtained. With the aid of these methods, future estimated values can be determined for different measurement signals and / or further measurement signals, which estimated values are used by the roll stabilization in the fourth program block 36.Depending on the embodiment chosen, the estimated value for the relative acceleration and / or the estimated value for the relative wheel travel of the Kalman filter can be fed to the prefilter at the third program block 35 and subsequent regulation at the fourth program block 36. With the aid of the supplied measurement signal, the control of the fourth program block 36 carries out a control of the height position of the wheel as a function of predetermined methods, in particular control methods and / or predetermined characteristic curves. In this case, a regulation of the angle of rotation and / or a regulation of the rotational speed and / or a regulation of the rotational acceleration can be transmitted from the control unit 30 to the actuator 20. For this purpose, the control device 30 can determine a setpoint rotational angle and / or a setpoint rotational speed and / or a setpoint rotational acceleration as a function of the estimated value and as a function of characteristic curves and / or tables and / or formulas and use it for regulating the roll stabilization with the aid of the actuator.Depending on the selected embodiment, other methods can also be used to determine a future estimated value for a setpoint value of a regulation for the roll stabilization.FIG. 5 shows, as an example, an upper diagram in which the acceleration a of the wheel for an offset jump in the acceleration signal is shown with a first characteristic curve 61 and the estimated value for the acceleration of the wheel is shown with a second characteristic curve 62. The estimated value was determined with the aid of the described Kalman filter.In a lower diagram of FIG. 5, the estimated value for the path s of a wheel with a third characteristic curve 71 and the measured path s of the wheel with a fourth characteristic curve 72 are shown over time t. The estimated value of the path s, i.e. the third characteristic curve, was determined with the aid of the Kalman filter according to FIG. 4. It can be seen from the diagram that the estimated value of the path leads the measured value of the path by the time difference Δt. The time difference can be selected accordingly depending on the application.With the aid of the described method, improved roll stabilization can be achieved, since estimated values for future measured values are used for the regulation of the roll stabilization. A temporal development of a measurement signal is thus estimated for the future and used for roll stabilization.
Claims
Method for carrying out roll stabilization of a vehicle, wherein a measurement signal of the vehicle is detected, wherein an estimated value for the measurement signal is determined with the aid of the measurement signal for a future point in time, wherein the estimated value is taken into account in the roll stabilization, characterized in that the estimated value is subjected to filtering which is carried out as dynamic filtering, wherein the dynamic filtering brings about compensation of a limited dynamic of an actuator of the roll stabilization and has an inversion of a temporal dynamic of the actuator and the filtered estimated value is used for the roll stabilization.Method according to Claim 1, wherein the roll stabilization is carried out with the aid of a control method, and wherein the estimated value is used for the determination of a controlled variable.Method according to one of the preceding claims, wherein a wheel stroke and / or a wheel acceleration and / or a wheel carrier acceleration are used as the measurement signal, and wherein a future wheel stroke and / or a future wheel acceleration and / or a future wheel carrier acceleration are determined as the estimated value.Method according to one of the preceding claims, wherein the determination of the estimated value is carried out with the aid of an observer.Method according to one of the preceding claims, wherein a dead time for a measurement signal is taken into account in the determination of the estimated value.Method according to one of the preceding claims, wherein a noise signal is taken into account for a detection of the measurement signal in the determination of the estimated value.Method according to one of the preceding claims, wherein the estimated value is determined with the aid of a Kalman filter.The method of any preceding claim, wherein the estimate represents a wheel acceleration.The method of any preceding claim, wherein the estimate represents a wheel path.An apparatus configured to perform a method according to any preceding claim.
Citation Information
Patent Citations
Device and method for influencing the spring force characteristics of an active chassis of a motor vehicle
DE102006039353A1
Method for controlling an actuator
DE102013018923A1
Method for operating a suspension system
DE102015205369A1
Motor vehicle, in particular passenger car, with a roll stabilization device
DE10221718A1
Regulating vehicle dynamics involves detecting variable system parameter deviation from base value, identifying system parameter, determining identified system model, adapting control gain to model
DE10254392A1