Method for detecting a side wind event, and control device

Abstract

The invention relates to a method for detecting a side wind event in a motor vehicle, wherein at least one operating parameter is measured, an average value and a standard deviation are calculated, a tolerance range is calculated, and the side wind event is detected on the basis of whether the operating parameter lies outside of the tolerance range at at least one measurement time. The invention additionally relates to a corresponding control device.

Description

[0001] Method for detecting a crosswind event and control device

[0002] The invention relates to a method for detecting a crosswind event in a motor vehicle and an associated control device.

[0003] Crosswind events can occur in motor vehicles when they are in motion. A crosswind event is typically understood to be a sudden, strong change in a crosswind. This can occur, for example, on bridges or when the motor vehicle drives into or out of the slipstream of a larger vehicle. Such crosswind events can cause the motor vehicle to experience a sudden change in course, which is also perceived by the driver, who typically reacts with abrupt countersteering. This can lead to dangerous situations. It would therefore be desirable to be able to react automatically to a crosswind event. To do this, it is necessary to reliably detect a crosswind event.

[0004] It is therefore an object of the invention to provide a method for detecting a crosswind event in a motor vehicle, which is alternative to or better than known embodiments. It is further an object of the invention to provide an associated control device. This is achieved according to the invention by a method and a control device according to the respective main claims.

[0005] Advantageous embodiments can be found, for example, in the respective subclaims. The content of the claims is incorporated into the description by express reference.

[0006] The invention relates to a method for detecting a crosswind event in a motor vehicle. The method comprises the following steps:

[0007] Measuring at least one operating parameter of the motor vehicle, calculating a mean value and a standard deviation of the operating parameter over a predetermined period of time and / or over a predetermined number of measured values,

[0008] Calculating a tolerance range based on the mean and standard deviation, and

[0009] Detecting a crosswind event based on the operating parameter being outside the tolerance range at at least one measurement point.

[0010] By providing tolerance ranges, it is possible to react dynamically to different situations in which the vehicle finds itself. In particular, the tolerance ranges can be dynamically adjusted to avoid false detections as much as possible. For example, if a vehicle drives over a particularly bumpy road for a certain period of time, it is not surprising that certain measured values ​​of operating parameters are subject to greater fluctuations than if the vehicle drives over a particularly smooth road. Accordingly, the threshold for detecting a crosswind event should also be raised. This prevents crosswind events from being incorrectly detected, which could lead to unwanted interventions or warnings.

[0011] A crosswind event is typically understood to be a strong change in the crosswind to which a motor vehicle is exposed. This can represent either a significant increase or a significant decrease. This typically leads to a direct influence on the course of the motor vehicle, which is not caused by a steering movement and is compensated for by the driver, for example, through strong countersteering. The operating parameters can be various parameters that can be measured by sensors. Determining an operating parameter in an implemented model is equivalent to measuring. A period of time can be specified, the length of which can be predetermined and, in particular, which can be dynamically adjusted. Each calculation of the mean and / or standard deviation therefore considers a predetermined period in the past.Likewise, a specific, predefined number of measured values ​​can be considered. For example, measured values ​​can be taken at predetermined times and / or at predetermined intervals. The known formulas can be used to calculate the mean and standard deviation. Calculation rules for the tolerance range depending on the mean and standard deviation, which have proven particularly advantageous for the application described here, are described in more detail below.

[0012] In particular, the aforementioned steps of the method can be repeated continuously. This can be understood, in particular, as continuous measurements and continuously updating the mean value, standard deviation, and tolerance ranges. This can be done, for example, at predetermined times or at predetermined intervals. After each newly obtained measured value of an operating parameter, it can be compared with the tolerance range, and a crosswind event can be detected if necessary.

[0013] In particular, each operating parameter can be assigned a separate tolerance range. This can thus be calculated or specified specifically for each operating parameter.

[0014] According to an advantageous embodiment, the or one operating parameter is a yaw rate of the motor vehicle. According to an advantageous embodiment, the or one operating parameter is a lateral acceleration of the motor vehicle. The operating parameters yaw rate and lateral acceleration in particular have proven particularly advantageous in determining crosswind events, since crosswind events are immediately and reliably noticeable in these operating parameters. In principle, however, other operating parameters can also be used alternatively or additionally. In particular, yaw rate and lateral acceleration can be used as operating parameters. According to an advantageous embodiment, the tolerance range has an upper limit, which is calculated by adding the standard deviation multiplied by an upper factor to the mean value.According to an advantageous embodiment, the tolerance range has a lower limit, which is calculated by subtracting the standard deviation multiplied by a lower factor from the mean. The specified calculation instructions follow the usual "point before line" rule. Thus, the standard deviation is first multiplied by the upper or lower factor, and then the resulting value is added to or subtracted from the mean.

[0015] In particular, it can be provided that the lower factor and / or the upper factor are taken from a table that specifies the lower factor and / or the upper factor depending on the number of measured values ​​used to calculate the standard deviation. This allows, in particular, to take into account the fact that the tolerance range should be larger or smaller depending on the number of available measured values. In particular, this also allows for the inherently increased inaccuracy with a small number of measured values ​​to be taken into account.

[0016] In particular, the table can specify the lower and / or upper factor such that the rate of false detection of a crosswind event is below a specified value. To this end, the table can specify the lower and / or upper factor according to a Student's t-distribution. This distribution was developed specifically to allow a tolerance range to be specified, even with a small number of measured values, which leads to a defined rate of false positive detections. This rate can be specified, and the table can be populated with entries accordingly.

[0017] According to an advantageous embodiment, the crosswind event is only detected when the operating parameter is outside the tolerance range for at least a predetermined number of at least two measured values. A higher number than two measured values ​​can also be used here. This ensures that a crosswind event is only detected when the operating parameter is outside the tolerance range not just for a single measured value, but for several consecutive measured values, so that a single, possibly random, exceedance does not automatically lead to the detection of a crosswind event.

[0018] In particular, it can be provided that the crosswind event is only detected when a lateral acceleration initially lies outside a tolerance range and only in a subsequent predefined time window does a yaw rate lie outside a tolerance range. This staggering has proven particularly advantageous for the reliable detection of a crosswind event while simultaneously maintaining a low error rate. Therefore, the

[0019] Lateral acceleration is monitored, and if it falls outside its tolerance range, a subsequent time window is specified during which the yaw rate is monitored. If the yaw rate falls outside its tolerance range within this time window, the crosswind event is detected. Otherwise, the fact that the lateral acceleration was outside the tolerance range does not lead to the detection of a crosswind event.

[0020] In response to the detection of a crosswind event, the method may in particular further comprise the following step:

[0021] Correcting a vehicle's course so that the course corresponds to that before the crosswind event.

[0022] The correction can be performed, for example, by means of braking intervention, steering intervention, and / or by means of intervention by one or more transverse torque-distributing actuators. This allows the vehicle to be stabilized again. In particular, the vehicle's course can be corrected independently of any reaction by the driver, thus relieving the driver's workload and preventing human error. In particular, the correction described herein can be performed so quickly that it occurs before a driver reacts and independently attempts to correct the crosswind event.

[0023] Examples of implementations of transverse torque distributing actuators are:

[0024] • Transverse torque distribution on the rear axle via a gearbox, which superimposes torque on the rear axle differential via clutches.

[0025] • TwinClutch system or two-clutch system, which can gradually decouple two output shafts individually using clutches.

[0026] • Torque vectoring via combined interventions on the brake and engine.

[0027] • For single-wheel electric motors, the distribution can be done directly via the control.

[0028] • A superposition of differential torques using electric motors.

[0029] • Use of any existing rear axle differential lock.

[0030] The invention further relates to a control device configured to carry out a method as described herein. The invention further relates to a non-volatile computer-readable storage medium on which program code is stored, the execution of which causes a processor to carry out a method described herein. With regard to the method, all embodiments and variants described herein can be used.

[0031] In other words, this article specifically examines the problem area of ​​yaw rate / yaw angle control when compensating for external disturbances such as crosswinds. The main problem here is that the influence of disturbances such as crosswinds can be very significant; for example, a crosswind event can cause the vehicle to shift by more than half a meter. At the same time, the vehicle's reaction can be so great that the driver overreacts. Compensation for these disturbances should therefore preferably be carried out through strong control interventions within a very short time after the disturbance is detected. In contrast to typical ESC interventions, which are only permitted for restricted driving maneuvers (cornering), the functionality should be active especially for straight-ahead driving and gentle cornering, i.e. in the vast majority of all situations at high speeds. The "crosswind disturbance" use case itself occurs rather rarely.In addition, there are a multitude of other disturbing factors, such as uneven road surfaces, road gradients, driver influences, and the like, which should not lead to control intervention. This places high demands on the robustness of the detection system in order to prevent erroneous interventions to such an extent that significantly fewer erroneous interventions occur than useful cases. Only a limited number of sensors (typically ESC sensors) are available for this purpose.

[0032] The procedure disclosed herein can be used, in particular, even if it is not limited to straight-ahead driving. It can therefore also be applied when cornering.

[0033] In particular, a stochastic approach can be used to minimize faulty interventions. In particular, the stochastic parameters of yaw rate and lateral acceleration (mean, standard deviation) are evaluated. Using the Student distribution, dynamic activation thresholds are defined according to a predefined error rate, above which an event is assessed as "exceptionally severe and requiring control given the situation." By combining two signals (yaw rate and lateral acceleration), the error probability is reduced to the product of both error probabilities. In addition, a strict condition can be imposed on the temporal shift of both events. The activation thresholds can also be raised by known external disturbances (driver steering, vehicle acceleration, wheel disturbances, torque fluctuations, etc.) that can generate similar patterns.By taking into account crosswind changes in particular and the stochastic evaluation of mean and standard deviation, the function can also be used in stationary cornering and is not limited to straight-ahead driving.

[0034] The control objective is typically, in contrast to the yaw rate, to correct the yaw angle in a highly dynamic manner. For this purpose, a yaw angle model is used to determine the yaw angle deviation since the first sign of a disturbance. If the control requirement is robustly identified, control takes place using a combination of a PID controller and a deadbeat controller. A very high operating point with a constant and yaw rate-dependent component can ensure highly dynamic corrective intervention. Control is terminated when the vehicle angle has reached the original angle, i.e., the vehicle is traveling in the original direction. Crosswind detection can also be used for crosswind corrections using other actuators, for example, via steering interventions or lateral torque distribution actuators.

[0035] The Student distribution mentioned above can be used particularly when there are few measured values. Such a small number of measured values ​​would typically result in a low standard deviation. The Student distribution expands this again. It can be specified in tabular form. It can specify a factor with which the standard deviation is corrected to account for the fact that there are only a few measured values. For example, for an error of 33% this value can converge to 1 (1 o), for an error of 5% it can converge to 2 (2 o), and for an error of 1% it can converge to 3 (3 o). This can be understood in particular as convergence for an increasing number of measured values. For such a value the integral over the Gaussian distribution then typically remains outside the tolerance range and is smaller than the permissible error.

[0036] With such a tabulated Student distribution, a factor can be specified for each one to account for the inaccuracy of the estimation of the statistical parameters. For a small number of values, the standard deviation is less precisely known, so a larger factor is typically used for safety reasons. The Student distribution is typically implemented in a table, in which the error remains below a definable limit, regardless of the number of available samples. The selection of the parameters sets the sensitivity of the system, i.e. the boundary between false detection and false non-detection is drawn. The error defined in the Student distribution typically corresponds to "false detection", i.e. the detection of an event without there actually being one.

[0037] Further features and advantages will become apparent to those skilled in the art from the exemplary embodiment described below with reference to the accompanying drawings. These show:

[0038] Fig. 1 : a motor vehicle in a typical usage situation, and

[0039] Fig. 2: a schematic procedure for carrying out a method according to the invention.

[0040] Fig. 1 shows a first motor vehicle 10 and a second motor vehicle 15. They are traveling in the same direction on a road, with the first motor vehicle 10 traveling in a left lane and the second motor vehicle 15 traveling in a right lane. The first motor vehicle 10 is traveling faster and is about to overtake the second motor vehicle 15.

[0041] As shown, both motor vehicles 10, 15 are exposed to a strong crosswind from the right. In the situation depicted in Fig. 1, the first motor vehicle 10 is still largely in the slipstream of the second motor vehicle 15. However, if the first motor vehicle 10 continues to move forward, it will move out of the slipstream and thus be fully impacted by the crosswind. This can lead to a crosswind event, which can cause the first motor vehicle 10 to deviate from its lane and / or shift. This can be detected and compensated for using the procedure schematically depicted in Fig. 2.

[0042] The starting point is a sensor system, which in this case consists of a lateral acceleration sensor 100 and a yaw rate sensor 105. The two sensors 100, 105 continuously measure the lateral acceleration and the yaw rate at predetermined times or at predetermined intervals and report the measured values ​​to a first functional module 110 and a second functional module 115. The first functional module 110 evaluates the measured lateral acceleration and calculates a mean value and a standard deviation based on a predetermined number of previous measured values. The second functional module 115 does the same with the yaw rate. Depending on the number of available measured values, the two functional modules 110, 115 take a factor from a table 112. The calculated standard deviation is multiplied by this factor.A tolerance range is then created around the calculated mean, extending from a lower limit to an upper limit. The lower limit is calculated from the mean minus the product of the standard deviation and the factor, and the upper limit is calculated from the mean plus the product of the standard deviation and the factor. The distance between the lower limit and the upper limit from the mean is therefore the same in this case. Alternatively, different values ​​could be used, for example, by specifying separate upper and lower factors. These can also be stored in Table 112.

[0043] The measured values ​​of the lateral acceleration and yaw rate, as well as the calculated tolerance ranges, are then forwarded to an evaluation module 120. This module checks whether the lateral acceleration is outside its tolerance range. If this is not the case, no action is taken. If it is, a time window is defined within which the yaw rate is monitored. If the yaw rate leaves its tolerance range within this time window, a crosswind event is detected. This information is forwarded to a correction module 130, which implements appropriate steering and / or braking intervention to restabilize the motor vehicle. This enables reliable detection of a crosswind event and a response to it.

[0044] The steps mentioned in the method according to the invention can be carried out in the specified order. However, they can also be carried out in a different order, as long as this is technically reasonable. The method according to the invention can be carried out in one of its embodiments, for example, with a specific combination of steps, in such a way that no further steps are carried out. However, in principle, further steps can also be carried out, even those not mentioned.

[0045] It should be noted that features may be described in combination in the claims and the description, for example, to facilitate understanding, although they may also be used separately. Those skilled in the art will recognize that such features may also be combined independently with other features or combinations of features.

[0046] References in subclaims may indicate preferred combinations of the respective features, but do not exclude other combinations of features.

[0047] List of reference symbols

[0048] 10 first motor vehicle

[0049] 15 second motor vehicle 100 lateral acceleration sensor

[0050] 105 Yaw rate sensor

[0051] 110 first functional module

[0052] 112 Table

[0053] 115 second functional module 120 evaluation module

[0054] 130 Correction module

Claims

Patent claims 1. A method for detecting a crosswind event in a motor vehicle (10), the method comprising the following steps: measuring at least one operating parameter of the motor vehicle (10), calculating a mean value and a standard deviation of the operating parameter over a predetermined period of time and / or over a predetermined number of measured values, Calculating a tolerance range based on the mean and standard deviation, and Detecting a crosswind event based on the operating parameter being outside the tolerance range at at least one measurement point.

2. The method according to claim 1, wherein the steps are repeated continuously.

3. Method according to one of the preceding claims, wherein the or an operating parameter is a yaw rate of the motor vehicle.

4. Method according to one of the preceding claims, wherein the or an operating parameter is a lateral acceleration of the motor vehicle.

5. Method according to one of the preceding claims, wherein the tolerance range has an upper limit which is calculated by adding the standard deviation multiplied by an upper factor to the mean value.

6. Method according to one of the preceding claims, wherein the tolerance range has a lower limit which is calculated by subtracting the standard deviation multiplied by a lower factor from the mean value.

7. Method according to one of claims 5 or 6, wherein the lower factor and / or the upper factor are taken from a table which specifies the lower factor and / or the upper factor as a function of the number of measured values ​​which were used for the calculation of the standard deviation.

8. The method according to claim 7, wherein the table specifies the lower and / or upper factor such that the rate of false detection of a crosswind event is below a predetermined value.

9. Method according to one of claims 7 or 8, wherein the table indicates the lower and / or the upper factor according to a Student's t-distribution.

10. Method according to one of the preceding claims, wherein the crosswind event is only detected when the operating parameter is outside the tolerance range for at least a predetermined number of at least two measured values.

11. Method according to one of the preceding claims, wherein the crosswind event is only detected when a lateral acceleration initially lies outside a tolerance range and only in a subsequent predetermined time window does a yaw rate lie outside a tolerance range.

12. A method according to any one of the preceding claims, further comprising, in response to detecting a crosswind event, the following step: Correcting a vehicle's course so that the course corresponds to that before the crosswind event.

13. The method according to claim 12, wherein the correction is carried out by means of a braking intervention, a steering intervention and / or by means of an intervention of one or more transverse torque distributing actuators.

14. Control device configured to carry out a method according to any one of the preceding claims.

15. A computer program product which, when executed on a control device, carries out a method according to one of claims 1 to 13.

Images