RECOGNIZING A BANQUET RIDE OF A MOTOR VEHICLE

Description

[0001] The invention relates to a method for detecting a motor vehicle driving on the shoulder of a road, wherein at least one measured variable is acquired by means of at least one sensor unit of the motor vehicle. The invention further relates to a corresponding system for detecting a motor vehicle driving on the shoulder of a road and to a computer program.

[0002] When a vehicle leaves the roadway and enters a shoulder, an unpaved area to the side of the road, serious accidents can occur. Attempts to steer the vehicle back onto the roadway can exacerbate this critical situation, particularly through excessive steering or braking by the driver. To assist the driver in such situations, automated braking or steering interventions can be performed by the vehicle's driver assistance system. For such a system to react appropriately, it is essential that the vehicle can reliably detect when it is driving on the shoulder.

[0003] The document EP 15 38 019 A2 describes a driver assistance system that uses a stereo camera to monitor whether a vehicle leaves the roadway.

[0004] Document US 2015 / 0203109 A1 also uses a camera to determine whether a vehicle leaves the roadway. This involves comparing a predicted vehicle path with an actual vehicle path.

[0005] Monitoring the vehicle's surroundings, for example with a camera system, is not sufficient on its own to reliably detect driving on the shoulder. Camera systems can, for instance, determine a vehicle's position based on lane markings. If these are heavily soiled, obscured, or missing, a camera system cannot reliably detect driving on the shoulder.

[0006] Document DE 10 2009 000079 A1 describes a method for detecting a dangerous driving condition of a vehicle. For this purpose, a damper signal is recorded, which represents the time course of a vertical movement of a wheel of the vehicle. The compression velocity is measured using a pressure sensor. From this measurement, a mean value and a standard deviation are calculated and compared with respective threshold values. If both threshold values ​​are exceeded, a marker is set for the wheel in question. It is stated that if a marker is set for one wheel and shortly thereafter a marker is also set for the wheel behind it, it can be concluded with a high degree of certainty that the vehicle is leaving the paved roadway.

[0007] Against this background, it is an object of the present invention to provide an improved concept for detecting a motor vehicle driving on the shoulder, with which the presence of the shoulder driving can be determined with a higher reliability.

[0008] According to the invention, this problem is solved by a method, a system, and a computer program as defined in the independent claims. Advantageous further developments and embodiments are the subject of the dependent claims.

[0009] The improved concept is based on the idea of ​​a two-stage process: first, individual confidence values ​​for the occurrence of a banquet run are determined, with the individual confidence values ​​based on different types of measurements. In the second stage, an overall confidence value for the occurrence of the banquet run is calculated based on the individual confidence values.

[0010] According to a first independent aspect of the improved concept, a method for detecting a motor vehicle driving off the shoulder is described, wherein at least one measured variable is acquired, in particular measured, by means of at least one sensor unit of the motor vehicle. Based on the at least one measured variable, a first control value relating to the wheel movement of a wheel of the motor vehicle is determined by means of a processing unit of the motor vehicle. Furthermore, based on the at least one measured variable, a second control value relating to a vehicle state variable of the motor vehicle that depends on a speed of the motor vehicle is determined by means of the processing unit. By means of the processing unit, a first confidence value for the occurrence of driving off the shoulder is determined based on a comparison of the first control value with a first limit value.Using the processing unit, a second confidence value for the occurrence of the banquet run is determined based on a comparison of the second control value with a second limit value. Finally, using the processing unit, an overall confidence value for the occurrence of the banquet run is determined based on the first and second confidence values.

[0011] The term "driving on the shoulder" as used here and in the following refers to a situation where at least one wheel of the motor vehicle is located on or in a shoulder while the vehicle is traveling, while at least one other wheel of the motor vehicle, in particular a wheel on the opposite side of the vehicle, is not in the shoulder but rather on the roadway. Specifically, in the case of a four-wheeled motor vehicle, driving on the shoulder might involve, for example, one right or one left wheel being in the shoulder while all other wheels are on the roadway, or both right wheels or both left wheels being in the shoulder while the other two wheels are on the roadway.

[0012] In the following, a roadway can be understood as a paved area intended for regular motor vehicle traffic, i.e., in particular a paved road, for example an asphalt road, a concrete road or a cobblestone road.

[0013] In this and the following, a shoulder can be understood as an unpaved area next to the roadway, particularly to the side of the roadway. The shoulder may, for example, have a grass surface, a gravel surface, a gravel lawn, or another unpaved surface. In particular, the surface condition of the shoulder differs from the surface condition of the roadway.

[0014] In particular, the coefficients of friction of a vehicle's wheel differ between the road surface and the shoulder. The coefficient of friction on the road surface is generally higher than the coefficient of friction on the shoulder. Driving on the shoulder therefore results in a so-called µ-split situation.

[0015] The control values ​​can each be a value of at least one measured quantity, or an average of one or more of the at least one measured quantity, whereby the averaging is carried out over a specified time interval, or a quantity derived from one or more of the at least one measured quantity or its time average in the specified time interval.

[0016] The wheel's movement can, for example, include a vertical movement of the wheel, i.e., a movement essentially parallel to a vertical axis of the vehicle and / or a normal axis of the surface on which the vehicle is located, such as a normal axis of the shoulder and / or the roadway, particularly relative to the vehicle's body. For example, the vertical wheel movement can be characterized by the suspension travel of the respective wheel or the damper acceleration of the respective wheel's damper. The first control value can therefore be determined based on the suspension travel, the damper acceleration, or another measured quantity that characterizes the vertical movement of the wheel relative to the body.

[0017] Wheel movement can also include horizontal movement, meaning movement of the wheel in a plane perpendicular to the vehicle's vertical axis, i.e., in a plane defined by the vehicle's longitudinal and transverse axes, particularly relative to the surface, such as the roadway or shoulder. Wheel slip, for example, can characterize horizontal wheel movement. The first control value can therefore also be determined based on wheel slip, especially measured slip or slip derived from a measured quantity, or on any other measured quantity characterizing the wheel's horizontal movement.

[0018] The first control value can also be determined based on several measurements that characterize the vertical and / or horizontal wheel movement.

[0019] In particular, wheel motion does not directly describe the motion, especially the speed, of the vehicle body. This means that knowing a measurement that describes wheel motion is not sufficient, for example, to uniquely determine the vehicle body speed. However, a measurement can certainly be used both to calculate a quantity characterizing wheel motion, such as wheel slip, and to determine the vehicle body speed or acceleration. For example, wheel rotational speed can be used to calculate wheel slip and, if the wheel diameter is known, also to calculate the vehicle body speed.

[0020] The vehicle state variable that depends on the vehicle's speed is, in particular, the longitudinal or lateral speed of the vehicle body, or a quantity derived from the longitudinal or lateral speed, such as the longitudinal or lateral acceleration of the vehicle body. The vehicle state variable can also be a rotational speed or a quantity derived from the rotational speed, such as the rotation rate or rotational acceleration. For example, it could be the yaw rate or yaw velocity of the vehicle body.

[0021] The confidence value for the occurrence of a banquet run can be understood here and in the following as an estimated probability or a measure of the likelihood that a banquet run actually occurred. The overall confidence value can therefore be considered the result of the procedure for detecting the banquet run. In particular, the banquet run can be defined as detected if the overall confidence value is greater than a predefined minimum value for the overall confidence.

[0022] The first and second confidence values ​​are weighted using different weighting factors by the calculation unit. The overall confidence value is determined based on the weighted first and second confidence values.

[0023] Due to the different origins of the individual confidence values, particularly due to differing calculation or measurement methods or sensor types, the individual control values ​​can generate confidence values ​​with varying degrees of reliability. This can be taken into account through weighting, by assigning higher weights to more reliable confidence values ​​when determining the overall confidence value than to less reliable ones. This improves the overall reliability of determining whether the vehicle was traveling at a roadblock.

[0024] The at least one measured variable can include, for example, wheel speed, wheel height, acceleration (e.g., longitudinal acceleration, lateral acceleration, or vertical acceleration of the vehicle body). The at least one measured variable can also include, for example, rotational acceleration (in particular about the vehicle's normal axis as the axis of rotation), steering torque, wheel suspension travel, damper acceleration, damper travel, damper position, or damper velocity of a damper associated with the wheel, or other measured variables. The at least one sensor unit contains a corresponding sensor or sensor system for measuring or determining the respective measured variable.

[0025] Using a method based on the improved concept, the first and second confidence values ​​are initially determined in a first stage. These values ​​already estimate the probability of the presence of a banquet run. However, the two confidence values ​​are further processed, particularly weighted, to determine the overall confidence value in a second stage of the procedure. This increases the reliability of banquet run detection. In particular, it reduces the probability of false positive detections or false negative non-detections of the banquet run.

[0026] The improved reliability stems primarily from the use of different types of control values ​​or measurements. Therefore, influences on one measurement that distort the first confidence value do not necessarily also distort the second confidence value, or vice versa. The at least partial independence of the two confidence values ​​from each other, achieved through the use of different measurements and / or control values ​​for their determination, increases the reliability of the overall confidence value.

[0027] Another advantage of the improved concept is that the sensors used to determine the measured values ​​for the control parameters generally do not need to be specifically designed for the shoulder detection procedure. Sensors like those described above are often already present in modern vehicles and are used for other purposes. Numerous driver assistance systems utilize such sensors. This allows for a synergy effect, as the use of sensors already present in the vehicle enables particularly reliable shoulder detection.For example, wheel motion sensors, i.e. sensors which record measured quantities used to determine wheel movement, can be used in Dynamic Chassis Control systems (DCC), or wheel speed sensors can be used, for example, for vehicle dynamics control systems (ESC - Electronic Stability Control).

[0028] According to at least one embodiment of the method for detecting driving on the shoulder of a road, a warning signal, particularly for the driver of the motor vehicle, is generated and issued by the processing unit based on the overall confidence value. Alternatively or additionally, a partially or fully automatic corrective measure can be initiated by the processing unit to end the driving on the shoulder, for example by automatically intervening in the steering of the motor vehicle or by applying individual wheel brakes.

[0029] According to at least one embodiment, one or more additional first control values ​​are determined based on the at least one measured quantity by means of the computing unit, each of which relates to a further wheel movement of the wheel.

[0030] According to at least one embodiment, the computing unit determines one or more additional second control values ​​based on the at least one measured quantity, each relating to a speed quantity, for example another speed quantity, of the motor vehicle.

[0031] For each of the additional first and / or additional second control values, a respective first and / or second additional confidence value for the occurrence of the banquet run is determined based on a comparison with a corresponding first and / or second additional limit value. The overall confidence value is then determined based on the first and second confidence values ​​as well as on the additional first and second confidence values.

[0032] The reliability of the method can be further increased by using multiple control values ​​of the same type, i.e., those relating to a wheel movement or a vehicle condition variable. All determined control values ​​can optionally be weighted accordingly to determine the overall confidence score.

[0033] The specified set of weighting factors can, for example, be based on empirically gained experience and / or determined based on machine learning methods.

[0034] According to at least one embodiment of the method, a road characteristic value, for example a road surface characteristic value, is generated by means of the at least one sensor unit and / or an environmental sensor signal is generated. By means of the processing unit, particularly before determining the overall confidence value, for example before determining the first and second confidence values ​​and / or before determining the first and second control values, it is determined, based on the road characteristic value and / or the environmental sensor signal, whether the presence of driving on the shoulder can be ruled out.

[0035] The fact that the presence of the shoulder of the road can be ruled out can be understood to mean that a sufficiently high, i.e. predefined, probability can be determined that the shoulder of the road is not present, in particular based on quantities or measured values ​​or parameters that are not used to determine the first or second control value.

[0036] The environmental sensor signal is, in particular, a signal from at least one sensor unit, which is generated to capture an image of the vehicle's surroundings, i.e., the vehicle's environment. This can be, for example, a signal from a camera or another optical sensor system, such as a lidar system, a radar system, or an ultrasonic sensor system of the at least one sensor unit.

[0037] For example, the existence of driving on the shoulder can be ruled out if, based on the environmental sensor signal, it is determined that lane marking lines are visible on both sides of the vehicle, i.e., to the left and right of the vehicle, and that the vehicle is driving between these lane marking lines, and / or that road edges, which in particular separate the roadway from the shoulder, are visible on both sides of the vehicle, i.e., to the left and right of the vehicle, and that the vehicle is driving between these road edges.

[0038] The road surface characteristic can be, in particular, a characteristic value or measurement for a surface condition of the substrate on which the vehicle travels. For example, the road surface characteristic can be a characteristic value or measurement for the surface roughness of the substrate. For example, depending on the road surface characteristic, the presence of a shoulder can be ruled out if the area on which the vehicle travels can be detected with a required probability as a roadway and, in particular, not as a shoulder.

[0039] In such configurations, a preliminary analysis can be performed based on the environmental sensor signal or the road surface characteristic to avoid false positive detections of driving on the shoulder by evaluating the first and second control values. This further increases the reliability of the procedure.

[0040] In various embodiments, the overall confidence value is only determined or used further if the presence of the shoulder cannot be ruled out based on the environmental sensor signal and / or the road surface characteristics. Optionally, the steps of determining the control values ​​and / or the first and second confidence values ​​can also be omitted in such a case.

[0041] In such embodiments, for example, the preliminary analysis—that is, determining whether the occurrence of the shoulder of the road can be ruled out—can be carried out repeatedly or continuously. The steps of determining the control values ​​and the confidence values ​​are only performed if, and in particular precisely if, the occurrence of the shoulder of the road cannot be ruled out based on the preliminary analysis, or not with sufficient probability.

[0042] According to at least one embodiment, at least one further environmental sensor signal is generated by means of the at least one sensor unit, and a third control value is determined by the processing unit based on this at least one further environmental sensor signal. Based on a comparison of the third control value with a third limit value, a third confidence value for the occurrence of the shoulder driving is determined by the processing unit. The overall confidence value is determined based on the first, second, and third confidence values.

[0043] By using a third type of control value, i.e., in addition to the first and second control values, a third control value derived from monitoring or mapping the vehicle's surroundings, the number of at least partially independent input variables for determining the overall confidence value, and thus the reliability of the entire procedure for detecting or determining whether the shoulder driving has occurred, can be further increased.

[0044] In various embodiments, one or more additional third control values ​​are determined by the processing unit, and based on comparisons of these additional third control values ​​with corresponding additional third limit values, additional third confidence values ​​for the occurrence of the banquet run are determined. These additional third confidence values ​​are also taken into account when determining the overall confidence value.

[0045] According to at least one embodiment, a further first control value is determined by the processing unit based on the at least one measured variable. This further control value relates to the wheel movement of another wheel of the motor vehicle, wherein the wheel and the further wheel are arranged on opposite sides of the motor vehicle. Based on a comparison of the first control value with the further first control value, the processing unit identifies one of the sides of the vehicle as the side of the shoulder drive.

[0046] The vehicle sides refer in particular to the left and right sides of the motor vehicle.

[0047] The term "side of the shoulder" here and in the following refers to the side of the vehicle on which at least one wheel of the motor vehicle is located in the shoulder.

[0048] The first and the subsequent first control values ​​are determined in the same way and from the same respective measured quantities, only for different wheels.

[0049] As described for the first control value and the first additional control value, an additional first control value relating to a wheel movement of the further wheel can be determined analogously for each additional first control value, and based on a comparison of the respective first additional control values ​​with the first additional control values, one of the vehicle sides can be identified as the side of the shoulder journey.

[0050] What has been explained here for the first and first additional control values ​​can also be applied analogously to the second and second additional control values ​​as well as to the third and third additional control values, provided that the respective control values ​​can be recorded page-specifically.

[0051] In various embodiments, one side of the vehicle is identified as the shoulder side based on different control values ​​and other control values. For a final identification of the shoulder side, several of the individual results can be weighted, for example.

[0052] In various embodiments, the side of the shoulder is only determined if the overall confidence value for the occurrence of the shoulder is greater than a predetermined overall limit.

[0053] By identifying the side of the shoulder, an improved automated response to the presence of the shoulder, i.e., a better adapted intervention in the vehicle control, can be carried out automatically.

[0054] According to at least one embodiment, at least one measured variable is the wheel speed of the wheel and / or the wheel speed of the other wheel.

[0055] According to at least one embodiment, the at least one measured quantity includes an acceleration of the motor vehicle, in particular of the body, especially a longitudinal, lateral and / or vertical acceleration of the body.

[0056] According to at least one embodiment, the at least one measured quantity includes an angular acceleration of the motor vehicle, in particular of the body, and in particular a rate of rotation, for example about the vertical axis of the motor vehicle. The angular acceleration corresponds, for example, to a yaw rate of the motor vehicle.

[0057] According to at least one embodiment, the at least one measured quantity includes a suspension travel of the wheel, i.e., in particular of a spring device of the motor vehicle which is assigned to the wheel, and / or of the further wheel.

[0058] According to at least one embodiment, the at least one measured quantity includes a damper acceleration of the wheel, i.e., a damper acceleration of a damper device of the motor vehicle which is assigned to the wheel, and / or a damper acceleration of the other wheel.

[0059] According to at least one embodiment, the first control value is determined depending on a first mean value and / or a first standard deviation and / or a first variance of the at least one measured quantity.

[0060] According to at least one embodiment, the second control value is determined depending on a second mean value and / or a second standard deviation and / or a second variance of the at least one measured quantity.

[0061] The mean, standard deviation, and variance refer to a predefined, in particular moving, time interval. Accordingly, the first and second mean, the first and second standard deviation, and the first and second variance can each be moving means, standard deviations (moving standard deviations), or variances, respectively. This applies analogously to all additional and further first, second, and third control values.

[0062] The use of time-based means or time-based standard deviations or variances further improves the robustness of the procedure and ultimately the reliability of determining the banquet run.

[0063] According to at least one embodiment, the first, second, and third confidence values ​​are weighted by the processing unit using different weighting factors. The overall confidence value is determined based on the weighted first and second confidence values ​​and the weighted third confidence value.

[0064] According to at least one embodiment, the at least one measured variable includes the wheel rotational speed. The first control value is determined as a function of the wheel rotational speed in such a way that it relates to a vertical wheel movement.

[0065] In particular, the wheel speed is measured continuously or repeatedly during a specified time interval. Discontinuities in the wheel speed value then indicate, for example, vertical wheel movements. These so-called wheel ticks can thus be used to infer vertical wheel movement from the wheel speed; that is, movement of the wheel relative to the vehicle body along or parallel to the vehicle's vertical axis.

[0066] Such embodiments have the particular advantage that no specific ride height sensors or damper acceleration sensors are required to determine the vertical wheel movement. Such ride height sensors or damper acceleration sensors are not standard equipment in many motor vehicles. In contrast, wheel speed sensors are installed in a very large number of motor vehicles, especially those equipped with an ESC system. This allows for a further synergy effect, as existing sensors can also be used for the inventive detection of shoulder driving.

[0067] According to a further independent aspect of the improved concept, a system for detecting a motor vehicle driving off the shoulder is described. The system comprises at least one sensor unit configured to acquire at least one measured variable of the motor vehicle, and a processing unit. The processing unit is configured to determine, based on the at least one measured variable, a first control value relating to the wheel movement of a wheel of the motor vehicle, and a second control value relating to a vehicle state variable of the motor vehicle that depends on the vehicle's speed. The processing unit is also configured to compare the first control value with a first limit value and, based on the comparison, in particular based on the result of the comparison of the first control value with the first limit value, to determine a first confidence value for the occurrence of driving off the shoulder.The computing unit is set up to compare the second control value with a second limit value and, based on the comparison, in particular a result of the comparison between the second control value and the second limit value, to determine a second confidence value for the occurrence of the banquet run and, based on the first and second confidence values, to determine an overall confidence value for the occurrence of the banquet run.

[0068] The computing unit is set up to weight the first and second confidence values ​​with different weighting factors and to determine an overall confidence value for the occurrence of the banquet cruise based on the weighted first and the weighted second confidence values.

[0069] Further embodiments of the system for detecting the shoulder of the road follow directly from the various configurations of the method for detecting the shoulder of the road according to the improved concept, and vice versa. A system according to the improved concept is, in particular, set up or programmed to carry out a method according to the improved concept, or it carries out a method according to the improved concept.

[0070] According to another independent aspect of the improved concept, a computer program is specified with instructions which, when the computer program is executed by a computing unit of a system according to the improved concept, cause the system to carry out a procedure according to the improved concept.

[0071] According to another independent aspect of the improved concept, a computer-readable storage medium is specified on which a computer program is stored according to the improved concept.

[0072] According to another independent aspect of the improved concept, a motor vehicle is specified with a system for detecting a shoulder drive according to the improved concept.

[0073] The invention also includes further developments of the method according to the invention, which have features already described in connection with the further developments of the system according to the invention. For this reason, the corresponding further developments of the method according to the invention are not necessarily described again here.

[0074] The invention also includes combinations of the features of the described embodiments.

[0075] The following are exemplary embodiments of the invention described. The following are shown: Fig. 1 a flowchart of a method for ending a shoulder journey of a motor vehicle with an exemplary embodiment of a method for detecting a shoulder journey according to the improved concept; Fig. 2 a motor vehicle with an exemplary embodiment of a system for detecting a shoulder journey according to the improved concept; and Fig. 3 a further exemplary embodiment of a method for detecting a shoulder journey according to the improved concept.

[0076] The exemplary embodiments described below are preferred embodiments of the invention. In these exemplary embodiments, the described components each represent individual features of the invention that can be considered independently of one another. Each of these features further develops the invention independently and can therefore be considered part of the invention individually or in a combination other than that shown. Furthermore, the described embodiment can also be supplemented by other features of the invention already described.

[0077] In the figures, functionally identical elements are each provided with the same reference symbols.

[0078] Fig. 1 shows a flowchart of an exemplary embodiment of a method for ending a banquet journey.

[0079] In step 1 of the procedure, a method for detecting a motor vehicle's shoulder drive is carried out according to the improved concept. As a result, for example, an overall confidence value for the occurrence of the shoulder drive is determined. For further details, please refer to the description of... Fig. 3 referred.

[0080] In step 2 of the procedure, for example, the overall confidence value is compared with a minimum value. If the overall confidence value is less than the minimum value, the procedure continues with step 8, whereby, in particular, no action is taken to end the banquet cruise, since the overall confidence value indicates that a banquet cruise is not taking place.

[0081] If, on the other hand, the overall confidence value is greater than or equal to the minimum value, step 3 of the procedure determines, for example, the reaction of the driver of vehicle 9 to the presence of the shoulder. In step 4 of the procedure, a measure can be initiated, for example, by means of a system for leaving or ending the shoulder, depending on the determined driver reaction. The measure can, for example, be a warning, stabilization, or agility enhancement, in particular automatic stabilization or agility enhancement, by intervening in the vehicle control system, for example, by intervening in the steering of the vehicle. In step 5, the measure can be continued.

[0082] In step 6, for example, it can be determined whether the vehicle 9 has left the shoulder 15. For this purpose, a shoulder driving detection procedure can be performed again according to the improved concept, and the resulting overall confidence value can be compared with the minimum value. If the overall confidence value is still greater than or equal to the minimum value, the process continues with step 5. If step 6 reveals that the overall confidence value is now less than the minimum value, the procedure is terminated in step 7, and normal driving resumes in step 8.

[0083] In Fig. 2 Figure 9 schematically depicts a motor vehicle, which includes an exemplary embodiment of a system 30 for detecting a shoulder drive of the motor vehicle 9. (Left in) Fig. 2 The motor vehicle 9 is shown as it is located on a roadway 12, specifically between a lane marking line 13 and a roadway edge 14. In the middle of Fig. 2 The motor vehicle 9 is partially located in a shoulder 15. In particular, the right wheels of the motor vehicle 9 are located in the shoulder 15 and the left wheels of the motor vehicle 9 are located on the roadway 12.

[0084] Right in Fig. 2 The motor vehicle 9 is shown after it has, for example using a procedure for ending the shoulder journey, as described in relation to Fig. 1 As described, it is now fully back on carriageway 12.

[0085] The system 30 for ending the shoulder drive includes, for example, respective measuring units 10 for determining wheel movements of the right wheels of the motor vehicle 9 and corresponding measuring units 10' for determining the wheel movements of the left wheels of the motor vehicle 9. In particular, the damper travel and / or wheel speed of the respective wheels can be determined using the measuring units 10, 10'.

[0086] The system 30 can also include one or more measuring units 11 for determining one or more vehicle state variables, for example a body speed, a yaw rate or a longitudinal, lateral or vertical acceleration of the motor vehicle 9, in particular a body of the motor vehicle 9.

[0087] Optionally, the system 30 can include an environment sensor unit 16, for example a camera, a lidar system, a radar system or an ultrasonic sensor system, to determine infrastructure features, for example lane marking lines 13, road edges 14, free spaces on the roadway 12, obstacles on the roadway 12 and / or a type of surface of the roadway 12 or the shoulder 15.

[0088] The system 30 also includes a computing unit 17. The computing unit 17 is configured to determine, based on measured quantities acquired by means of the measuring units 10, 10', 11 and, if applicable, the environmental sensor unit 16, a first control value relating to the wheel movement of the respective wheel of the motor vehicle, and a second control value relating to the speed of the motor vehicle, and, if applicable, to determine a third control value based on the environmental sensor signals of the environmental sensor unit 16.

[0089] The computing unit 17 is set up to compare each of the control values ​​with a respective limit value and, based on this, to determine a respective confidence value for the occurrence of the banquet run and, based on the respective confidence values, to determine an overall confidence value for the occurrence of the banquet run.

[0090] For further details, please refer to the explanations regarding Fig. 3 referred.

[0091] In Fig. 3 An exemplary embodiment of a method for recognizing a banquet drive is shown as a corresponding flowchart.

[0092] In step 18, for example, one or more measured variables, such as vehicle state variables like vehicle speed, yaw rate, or accelerations, are acquired using at least one of the measuring units 10, 10'. In step 19, for example, another measured variable for detecting wheel movement of the vehicle 9 is acquired using one of the measuring units 11. Optionally, in step 20, an environmental sensor signal is generated using the environmental sensor unit 16.

[0093] In the optional step 21, a preliminary analysis is performed, primarily based on the environmental sensor signals. This preliminary analysis can, for example, determine whether the vehicle 9 is clearly and unambiguously located on lane 12. This can be done, for instance, by detecting that lane markings 13 and / or road edges 14 are visible to the left and right of the vehicle 9 and that the vehicle 9 is positioned between them. Alternatively or additionally, a classification can be performed based on the environmental sensor signal to identify the road surface and clearly and reliably classify the area on which the vehicle 9 is traveling as lane 12.If, in step 21, it is evident and unambiguous that the vehicle is traveling on carriageway 12, and thus driving on the shoulder can be ruled out with sufficient probability, the procedure is terminated in step 29 by not recognizing the current situation as driving on the shoulder and setting the overall confidence value, for example, to zero.

[0094] However, if step 21 reveals that the banquet run cannot be ruled out based on the preliminary analysis, the process continues with steps 22 and 23, and optionally with step 24.

[0095] In step 22, an initial control value is determined based on the measurements recorded in step 18, specifically by averaging the respective measurements over a predefined time interval. If the respective averaged measurement is less than a first threshold value, this could, for example, indicate that the respective wheel is traveling on lane 12. If the averaged measurement is greater than or equal to the first threshold value, this could indicate that the respective wheel is traveling on the shoulder 15.

[0096] Similarly, a corresponding evaluation is performed in step 23 for the measured values ​​recorded in step 19. Optionally, a corresponding evaluation can be performed in step 24 for the environmental sensor signals recorded in the optional step 20.

[0097] Each comparison of one of the averaged measurements with the one corresponding limit value then yields a single confidence value for the occurrence of the banquet run.

[0098] The comparisons performed in steps 22 to 24 may, for example, include one or more of the following comparisons: Is the moving standard deviation (MSD) of a given wheel speed greater than the corresponding threshold? Is the MSD of a given vertical acceleration greater than the corresponding threshold? Is the MSD for a given wheel slip greater than the corresponding threshold? Is the MSD of a yaw rate greater than the corresponding threshold? Is the MSD of a lateral acceleration greater than the corresponding threshold? Is the MSD of a steering torque less than the corresponding threshold? Is a lateral offset of the vehicle 9 to the relevant lane marking line 13 greater than the corresponding threshold? Is a lateral offset of the vehicle to the relevant road edge 14 greater than the corresponding threshold? Is the lateral offset of each wheel of the vehicle 9 to the relevant lane marking line 13,Is the lateral offset of each wheel of the vehicle to the relevant lane marking 14, calculated, for example, based on the lateral offset of the vehicle to the relevant lane marking 13 and a relative angle between the direction of travel of the vehicle 9 and the relevant lane marking 13, greater than the corresponding threshold value? Is the lateral offset of each wheel of the vehicle to the relevant road edge 14, calculated, for example, based on the lateral offset of the vehicle 9 to the relevant road edge 14 and the relative angle between the direction of travel of the vehicle 9 and the relevant road edge 14, greater than the corresponding threshold value?

[0099] The more clearly a given control value exceeds its individual threshold or limit, the higher the probability of a road closure that can be derived from the respective measurement signal. In step 25, an overall confidence value, which describes the overall probability of the road closure being correct, is calculated from the weighted individual confidence values ​​or probabilities. The more measured variables exceed their individual thresholds, thus indicating a road closure, the higher the overall confidence value. The overall confidence value can be determined, in particular, from those control values ​​that have exceeded their individual thresholds.

[0100] In steps 26 to 28, for those measured variables for which a side-specific statement can be made, a classification or determination of the side of the shoulder can be carried out. This determines whether the vehicle 9 is driving with its right or left wheels in the shoulder 15. For each measured variable for which side-specific counterparts exist, the respective control value is compared with a corresponding control value of the measured variable belonging to the opposite side of the vehicle 9. If the respective control value is higher on one side than on the other, this indicates, for example, that the vehicle 9 is driving with the wheels of the side on which the stronger measurement signal was detected in the shoulder 15.

[0101] In particular, in steps 26 to 28, wheel speeds of the left and right wheels of the front axle and / or of the left and right wheels of the rear axle can be compared, and / or vertical acceleration at the left and right wheels of the front or rear axle and / or wheel slip at the left and right wheels of the front or rear axle. Alternatively or additionally, respective values ​​for a lateral offset of the vehicle 9 with respect to the left and right lane marking line 13 can be compared, and / or values ​​for a lateral offset of the vehicle 9 with respect to the left and right road edge 14 can be compared.

[0102] According to the improved concept, reliable detection of whether the motor vehicle 9 is on the roadway 12 or in the shoulder 15 is enabled. Optionally, it can be classified whether the right or left wheels of the motor vehicle 9 are in the shoulder 15.

[0103] According to the improved concept, control values ​​for different measured variables, particularly different types or classes of measured variables, are determined and initially evaluated independently of one another with regard to the presence of a shoulder drive. The individual control values ​​yield individual confidence values ​​for the presence of a shoulder drive. The individual confidence values ​​are processed into an overall confidence value, which enables reliable detection of the shoulder drive.

[0104] For example, measuring units can be used to detect wheel movement in a vertical direction, such as suspension travel or damper acceleration, and / or in a horizontal direction, such as wheel speed and the resulting wheel slip. DCC ride height sensors or wheel speed sensors can be used for this purpose.

[0105] Furthermore, measuring units can be used to detect vehicle state variables, such as speed, yaw rate, or longitudinal, lateral, or vertical acceleration of the vehicle. Wheel speed, acceleration, or yaw rate sensors can be used for this purpose.

[0106] Optionally, additional environmental sensor units can be used to detect infrastructure features such as lane markings, road edges, clearances, obstacles, or to classify the road surface. This can be done, for example, using a camera. The surface classification can be performed in different image segments, in the simplest case as "roadway" and "non-roadway." Image processing algorithms for surface classification can be trained using machine learning methods (artificial neural networks, machine learning, deep learning) and image training data. Alternatively or additionally, a high-precision map combined with high-precision localization can be used.

[0107] The improved concept allows for a simple and robust method for evaluating sensor information to detect and classify road shoulders. In particular, the method used is less complex and more transparent than methods based on evaluating information from various sensors with different output variables, such as Bayesian networks, artificial neural networks, or methods based on the Dempster-Shafer evidence theory.

Claims

1. Method for detecting a motor vehicle (9) driving on a verge, at least one measured variable being recorded by means of at least one sensor unit (10, 10', 11, 16) of the motor vehicle (9), and by means of a computing unit (17) of the motor vehicle (9): - a first control value relating to a wheel movement of a wheel of the motor vehicle (9) being determined based on the at least one measured variable; - a second control value relating to a vehicle state variable dependent on a speed of the motor vehicle (9) being determined based on the at least one measured variable; - a first confidence value for the verge being driven on being determined based on a comparison of the first control value with a first limit value; - a second confidence value for the verge being driven on being determined based on a comparison of the second control value with a second limit value; characterized in that - the first confidence value and the second confidence value are weighted, using different weighting factors, by means of the computing unit (17); and - a total confidence value for the verge being driven on is determined based on the weighted first and the weighted second confidence value.

2. Method according to claim 1, <b>characterized in that: - a road characteristic value is generated and / or an environment sensor signal is generated by means of the at least one sensor unit (10, 10', 11, 16); - it is determined, by means of the computing unit (17), based on the road characteristic value and / or the environment sensor signal, whether the verge being driven on can be ruled out.

3. Method according to either claim 1 or claim 2, characterized in that - at least one additional environment sensor signal is generated by means of the at least one sensor unit (10, 10', 11, 16); - a third control value is determined, by means of the computing unit (17), based on the at least one additional environmental sensor signal; - a third confidence value for the verge being driven on is determined, by means of the computing unit (17), based on a comparison of the third control value with a third limit value; and - the total confidence value is determined based on the first, the second and the third confidence value.

4. Method according to any of claims 1 to 3, characterized in that - an additional first control value relating to a wheel movement of an additional wheel of the motor vehicle (9) is determined, by means of the computing unit (17), based on the at least one measured variable, the wheel and the additional wheel being arranged on opposite sides of the vehicle; and - one of the vehicle sides is identified as the side on which the verge is being driven on, by means of the computing unit, based on a comparison of the first control value with the additional first control value.

5. Method according to any of claims 1 to 4, characterized in that for recording the at least one measured variable - a wheel speed of the wheel; and / or - an acceleration of the motor vehicle (9); and / or - an angular acceleration of the motor vehicle (9); and / or - a compression travel of the wheel; and / or - a damper acceleration of the wheel; is recorded.

6. Method according to any of claims 1 to 5, characterized in that - the first control value is determined as a function of a first mean value and / or of a first standard deviation and / or of a first variance of the at least one measured variable; and / or - the second control value is determined as a function of a second mean value and / or of a second standard deviation and / or of a second variance of the at least one measured variable.

7. Method according to any of claims 1 to 6, <b>characterized in that: - the at least one measured variable includes a wheel speed of the wheel; and - the first control value is determined as a function of the wheel speed such that the first control value relates to a vertical wheel movement of the wheel; - the wheel speed of the wheel is measured continuously or repeatedly during a predetermined time interval and discontinuities in the measured wheel speed value are used to infer the vertical wheel movement from the wheel speed.

8. System for detecting a motor vehicle (9) driving on a verge, the system (30) having at least one sensor unit (10, 10', 11, 16) which is designed to record at least one measured variable of the motor vehicle (9), and a computing unit (17) which is designed to: - determine a first control value relating to a wheel movement of a wheel of the motor vehicle (9) based on the at least one measured variable; - determine a second control value relating to a vehicle state variable dependent on a speed of the motor vehicle (9); - compare the first control value with a first limit value and, based on the comparison, determine a first confidence value for the verge being driven on; - compare the second control value with a second limit value and, based on the comparison, determine a second confidence value for the verge being driven on; characterized in that the computing unit is further designed to: - weight, using different weighting factors, the first confidence value and the second confidence value; and - determine a total confidence value for the verge being driven on based on the weighted first and the weighted second confidence value.

9. Computer program comprising instructions which, when the computer program is run by a computing unit (17) of a system (30) according to claim 8, cause the system (30) to carry out a method according to any of claims 1 to 7.