Method and driver assistance system for height classification of objects in the vehicle surroundings

Abstract

A method for height classification of objects in the surroundings of a vehicle (1) in the case of using an ultrasonic sensor (10) that emits and receives ultrasonic echoes is proposed. Among them, a low object that can be driven over is distinguished from a high object that cannot be driven over. A height indicator H is determined. A histogram with n categories K of the values of the height indicator is also created i where the categories K1 to K n represent the values of the height indicator that characterize high objects, and the categories K(h + 1) to Kn represent the values of the height indicator that characterize low objects, and an indicator Q for indicating the probability of the presence of a high object is determined, which is the sum of the categories K i characterizing high objects weighted by a weighting coefficient b i for i from 1 to h divided by the sum of all categories K i weighted by a weighting coefficient bi for i from 1 to n. Another aspect of the present invention relates to a driver assistance system (100) configured to implement this method.

Description

Field of the Invention

[0001] The present invention relates to a method for height classification of objects in the surroundings of a vehicle in the case of using an ultrasonic sensor, which emits ultrasonic pulses and receives ultrasonic echoes reflected by the objects. In this height classification, a distinction is made between drivable low objects and non-drivable high objects. Another aspect of the present invention relates to a driver assistance system configured to implement this method. Background Art

[0002] Modern vehicles are equipped with a large number of driver assistance systems, which assist the driver of the vehicle in performing various driving operations. In addition, driver assistance systems for warning the driver of dangers in the surroundings are known. For a driver assistance system to function, accurate data on the surroundings of the vehicle, in particular on the objects in the surroundings of the vehicle, is required.

[0003] Ultrasound-based object localization methods in which two or more ultrasonic sensors are used are often employed. The ultrasonic sensors each emit ultrasonic pulses and receive the ultrasonic echoes reflected by the objects in the surroundings. From the propagation time of the ultrasonic pulse until the corresponding ultrasonic echo is received and the known speed of sound, the distance between the reflecting object and each sensor can be determined. If an object is within the field of view of more than one ultrasonic sensor, i.e., the distance to the object can be determined by multiple ultrasonic sensors, the exact position of the reflecting object relative to the sensors or relative to the vehicle can also be determined by a lateration algorithm.

[0004] Due to the continuously increasing field of view and sensitivity of the sensors, objects on the ground (such as curbstones, bumps (Schwellen) or manhole covers) are also increasingly recognized. Here, for the normal operation of a driver assistance system, it is important to be able to distinguish between objects that are important for collision (such as pillars, walls or traffic signs) and drivable objects that are not important for collision (such as curbstones, bumps or manhole covers).

[0005] A method for recognizing objects with a low height is known from DE 10 2009 046 158 A1. It is provided here that the distance to the object is continuously detected by means of a distance sensor, and it is checked whether the object is still detected by the distance sensor when the vehicle approaches and the distance is below a pre-given distance, or whether the object disappears from the detection area of the distance sensor. If it is recognized that the object disappears from the detection area of the distance sensor during approach, the object is classified as an object with a low height.

[0006] In addition, methods are known in the prior art that make use of the following fact: Tall and extended objects usually do not have a unique, clearly defined reflection point, and this can thus cause multiple reflections on a single ultrasonic pulse and thus multiple ultrasonic echoes that are temporally successive to one another. In the case of a tall object, the reflection extends, for example, directly horizontally, i.e., from the sensor to the object parallel to the ground and back to the tall object. Another reflection is reflected back from the inner corner (Kehle) between the ground and the tall object. This second ultrasonic echo arrives after the first ultrasonic echo in time because the path from the mounting position of the sensor to the contact point between the object and the ground must be longer compared to the directly horizontally extending path. It is also known that certain objects such as bushes or pedestrians and flat objects such as drainage grates or manhole covers cause a large number of reflections that are regarded as echoes as noise-like signals.

[0007] DE 10 2007 061 235 A1 describes a method for classifying the height of an object by making use of statistical dispersion, in particular caused by multiple reflections of the measurement signal.

[0008] The problem with known methods for height classification is that the parameters used for classification, for example, in particular the number of received reflections of the object, are strongly correlated with other conditions such as the current interference level. Summary of the Invention

[0009] A method for classifying the height of objects in the surroundings of a vehicle using an ultrasonic sensor is proposed. The ultrasonic sensor emits ultrasonic pulses and receives the ultrasonic echoes reflected by the objects. During height classification, a distinction is made between drivable low objects and non-drivable tall objects. In this method, according to the invention, a height indicator H is determined for the emitted ultrasonic pulses. A histogram with n classes K i of the values of the height indicator H is also created, where the classes K 1 to K n represent values of the height indicator H that characterize tall objects, the classes K (h+1) to K n represent values of the height indicator H that characterize low objects, and an indicator Q that describes the probability of the presence of a tall object is determined. This indicator Q is the quotient of the sum of the classes K i characterizing tall objects weighted with a weighting factor b i for i from 1 to h and the sum of all classes K i weighted with a weighting factor bi for i from 1 to n. Thus, by the formula

[0010]

[0011] Give the indicator Q.

[0012] Determine the height indicator H based on a combination of at least two of the number of received ultrasonic echoes, the amplitude of the received ultrasonic echoes, the signal correlation of the received ultrasonic echoes, the distance assigned to the received ultrasonic echoes, and the parameter. Preferably, all the mentioned parameters are incorporated into the determination of the height indicator H.

[0013] In this method, at least one ultrasonic sensor of the vehicle emits an ultrasonic pulse, which is also referred to as a single "shot (Schuss)" of the ultrasonic sensor. The ultrasonic sensor of the vehicle then receives the ultrasonic echoes reflected by an object in the surrounding environment. Here, depending on the size, height, and characteristics of the object, the object may not only cause a single ultrasonic echo but usually multiple ultrasonic echoes occur. For example, an impassable tall object usually provides at least two ultrasonic echoes for the emitted ultrasonic pulse. Here, the first ultrasonic echo returns horizontally from the object at the height of the ultrasonic sensor to the ultrasonic sensor. The second ultrasonic echo is caused by the inner angle between the object and the ground and here travels a longer path than the first ultrasonic echo due to the height difference. Here, the ultrasonic echoes can be received by one or more ultrasonic sensors of the vehicle respectively.

[0014] In this method, it is provided that multiple ultrasonic echoes from one object are combined into a height indicator. Here, the height indicator is preferably assigned to the object from which these ultrasonic echoes originate. In addition to incorporating the number of ultrasonic echoes received for the emitted ultrasonic pulse into the height indicator H, the amplitude of the ultrasonic echoes, the signal correlation of the ultrasonic echoes, and / or the determined distance for the ultrasonic echo are also incorporated. The signal correlation here represents the correlation of the corresponding ultrasonic echo with the emitted ultrasonic pulse. The distance assigned to the ultrasonic echo is determined by the known speed of sound in air and the corresponding propagation time between the emission of the ultrasonic pulse and the reception of the ultrasonic echo. Here, in particular, the deviation of the distance between individual ultrasonic echoes can be considered, that is, for example, the difference between the distance of the first ultrasonic echo and the distance of the last ultrasonic echo in multiple reflections.

[0015] Grouping multiple received ultrasonic echoes into multiple reflections of an object preferably occurs at a lower signal level where there is no information about the object's position yet. For this purpose, for example, a capture window (Fangfenster) in time is pre-given, and the echoes within the capture window are regarded as multiple reflections of the object.

[0016] To determine the height indicator, the selected parameters, such as the number of received ultrasonic echoes, the amplitude of the received ultrasonic echoes, the signal correlation of the received ultrasonic echoes, the distances assigned to the received ultrasonic echoes, and combinations of at least two of these parameters, are associated with the height indicator H.

[0017] The method preferably runs continuously such that the height indicator H is continuously determined accordingly. For further analysis, a histogram of the values obtained for the height indicator H is created. The histogram has n classes. n is selected, for example, in the range from 4 to 100, particularly preferably in the range from 6 to 20. For example, 7 or 16 classes are selected.

[0018] The height indicator H can exist in the form of continuous values, where a predefined value range of the height indicator is assigned to each class. Alternatively, the height indicator can exist in the form of discrete values or statements, where one or more discrete values or statements can be assigned to a class. A statement can be, for example, the following classification: the associated parameters result in a height indicator indicating a specific condition or a specific type of the object. The height indicator can state, for example, that the considered parameters typically represent a curb, a wall, a shrub, or a pedestrian. In the mentioned example, only the curb is drivable.

[0019] The height indicator characterizing the curb is, for example, characterized by one to two ultrasonic echoes at small distances received. The amplitude of the second ultrasonic echo is very low, while the amplitude of the first echo is distance-dependent.

[0020] The height indicator characterizing the wall is, for example, characterized by two ultrasonic echoes, where the first ultrasonic echo has a very high amplitude and high correlation, and the second ultrasonic echo has a low amplitude and low correlation. Here, the distances assigned to these ultrasonic echoes are greater than in the case of the curb.

[0021] The height indicator characterizing the shrub is, for example, characterized by two to four ultrasonic echoes, where all echoes have a very low amplitude and correlation.

[0022] The height indicator characterizing the pedestrian is, for example, characterized by two to three ultrasonic echoes with medium amplitude and low correlation. The variation in the distances assigned to these ultrasonic echoes is typically very large (up to 40 cm between the first echo and the last echo).

[0023] Subsequently, based on the continuously updated histogram, the indicator Q is continuously determined, which is obtained by dividing the weighted sum of all classes indicating high objects by the weighted sum of all classes.

[0024] Preferably, a category is determined empirically based on tests to indicate whether it is a non-drivable high object or, conversely, a drivable low object. Correspondingly, the parameter h is selected empirically, and this parameter h indicates which category in the said categories indicates a high object.

[0025] The weighting factor b for weighting the sum i is also preferably determined empirically by tests. Then, the weighting factor b i can be stored, for example, in the memory of the control device implementing this method. The weighting factor b i can be stored, for example, in the form of a look-up table.

[0026] The empirically determined weighting factor b i preferably indicates i how persuasive the category K

[0027] Subsequently, the value of the indicator Q thus obtained and preferably continuously updated is used as an indicator of the presence of a non-drivable high object.

[0028] In this method, it is preferably set that the detection threshold of the ultrasonic sensor is matched to the current interference level cl such that the ratio of misclassifying ultrasonic echoes as object echoes is constant.

[0029] The interference level cl is usually dominated by the so-called clutter caused by ground echoes here. The amplitude and quantity of the ground echoes or the clutter are strongly related to the nature of the ground here. Fewer echoes occur in the case of a smooth ground surface compared to, for example, an uneven ground surface such as gravel. In addition, other ambient noise or ultrasonic pulses from other ultrasonic sensors may generate the interference level. Correspondingly, it is preferably set here that the detection threshold is respectively matched to the current existing ambient conditions, so as to reduce the detection threshold in the case of low ambient noise or low ground echo quantity, while conversely, to increase the detection threshold in a noisy ambient environment with, for example, multiple interference signals, a large amount of noise, and / or a large amount of ground echoes due to a rough ground surface such as gravel. To adapt the detection threshold, for example, an algorithm can be used, and this algorithm adjusts the detection threshold so that a constant false alarm rate (CAFR) is achieved.

[0030] When creating a histogram, it is preferably based on the relationship

[0031] K i = K i + a i * H

[0032] for weighting, where a iis a weighting factor determined empirically.

[0033] Here, it is preferable to select the weighting factor a determined empirically i : perform normalization, through which the height indicator H is made independent of the currently set sensitivity of the ultrasonic sensor.

[0034] The weighting factor a obtained empirically i is preferably a function a related to the interference level cl and the distance d of the object i (cl, d).

[0035] The weighting factor a i or rather the function a i (cl, d) is preferably stored in the memory of the control device implementing this method in the form of a table or rather in the form of a look-up table.

[0036] The normalization factor is achieved by the weighting factor for constructing the histogram pre-given as the function a i (cl, d), and this normalization factor eliminates the influence of the distance d and the interference level cl, especially clutter. In an advantageous manner, when combined with the adaptive threshold for the ultrasonic sensor, the system is more sensitive at low interference levels and less sensitive at high interference levels. In the case where the ultrasonic sensor is set to be more sensitive, without corresponding normalization, the values obtained for the height indicator H would increase for all objects, that is, not only for low objects but also for high objects. In addition, the range of values determined for the height indicator at low interference levels is not persuasive, so it is necessary to mask these value ranges with the weighting factor 0 if necessary.

[0037] The obtained distance d of the object also has an impact on the persuasiveness (Aussagekraft) of the height indicator. In the near region at a distance of up to about 50 cm, the ultrasonic echoes from the internal angle between the object and the ground are outside the field of view of the ultrasonic sensor. However, the ultrasonic sensor can only receive the ultrasonic echoes of the objects located within its field of view. The interference level caused by clutter reaches its maximum value within the range between 80 cm and 120 cm. In the far region above 200 cm, the amplitudes of some ultrasonic echoes drop below the detection threshold of the ultrasonic sensor. Compared with outside these distance ranges, a lower weighting factor a is preferably selected in these problematic regions i .

[0038] In the case of multiple reflections, that is, when more than one ultrasonic echo is assigned to an object, it is preferable to regard the distance assigned to the first ultrasonic echo as the distance d of the object.

[0039] Preferably, the indicator Q determines a confidence value, where the confidence value is related to the number of entries in the histogram. In the case of a small number of entries, i.e., with few received ultrasonic echoes, the persuasiveness of the indicator Q is still relatively low and only increases with the increase in the number of received ultrasonic echoes and correspondingly with the increase in the number of entries in the histogram.

[0040] When performing height classification, it is preferable to consider at least one additional classification parameter in addition to the indicator Q, where the at least one additional classification parameter is selected from the confidence value of the indicator Q, the amplitude of the ultrasonic echo, the gradient of the ultrasonic echo amplitude, the quotient of the distance difference and the difference formed by the traveled distance, and combinations of multiple of these classification parameters.

[0041] The gradient of the ultrasonic echo amplitude is the change in the amplitude of the ultrasonic echo assigned to the object observed when the distance between the object and the vehicle changes. Here, for example, it is set that the distance to the object is continuously detected by the ultrasonic sensor and it is checked whether the object is still detected by the distance sensor when the vehicle approaches and when the distance is below a pre-given distance or the object disappears from the detection area of the distance sensor. If, for example, it is recognized that the object disappears from the detection area of the distance sensor when approaching, the object is classified as a low object that can be driven over.

[0042] The quotient obtained from the distance difference and the difference formed by the traveled distance can be regarded as the ratio of the change in the distance measured in terms of the propagation time to the change in the distance to the object due to the vehicle movement. For the secondary echo (Nebenecho) caused by the interior angle between the ground and the high object, for example, according to the Pythagorean theorem

[0043]

[0044] the distance D calculated from the propagation time of the ultrasonic echo is obtained, where d is the distance between the object and the sensor and h is the mounting height of the sensor above the ground. The distance d between the object and the sensor is always the shortest distance obtained from a direct line extending parallel to the ground. Since the height h of the sensor is constant, the calculated distance D changes less compared to the actual distance d to the object.

[0045] In order to be able to assign the received echoes to objects in the surrounding environment, it is preferred to create an object hypothesis and assign the received ultrasonic echoes to the object hypothesis respectively. For example, the ultrasonic echo can be assigned to the object hypothesis when using lateration. This position determination with the help of lateration is advantageous for using the acquired data in a driver assistance system. However, such an assignment is not necessary for implementing the classification method according to the present invention. The interpretation of multiple reflections as ultrasonic echoes of a unique object is preferably carried out using a temporal capture window, wherein all ultrasonic echoes received within the capture window are regarded as multiple reflections of the same object.

[0046] Preferably, for edge measurement, the distance between the respective ultrasonic sensor and an object in the surroundings reflecting the ultrasonic pulse is ascertained by at least two ultrasonic sensors with at least partially overlapping fields of view, wherein the position of the reflecting object is determined by means of the edge measurement.

[0047] Here, if the measurement results that the position of each object reflecting ultrasound is consistent with or near the position assigned to an object hypothesis, then in particular, the measured values ​​obtained in succession in time, that is, the distance values ​​determined in sequence in time, can be assigned to the same object hypothesis. By analyzing the total amount of measurements assigned to an object hypothesis or by analyzing the distance and position determined by means of an ultrasonic sensor, the contour of the object can also be inferred. If, for example, the vehicle moves forward uniformly in one direction, and all the positions assigned to an object hypothesis are located on a line, or all the positions of all ultrasonic sensors of a bumper assigned to an object hypothesis are located on a line, it can be inferred that the object assigned to the object hypothesis is an extended object, such as a wall or another vehicle. On the contrary, if the position does not change approximately, there may be a point-like object, which has only a small geometric extension scale when viewed on a plane parallel to the ground. It is, for example, a column, a traffic sign, or a landmark corner of another object, such as a vehicle corner or a house corner, or even a curb edge corner. Such a joining of individually measured distances to form an extended object is described, for example, in DE 10 2007 051 234 A1.

[0048] Another aspect of the invention relates to a driver assistance system comprising at least one ultrasonic sensor and a control device. The driver assistance system is designed and / or arranged to carry out one of the methods described herein.

[0049] Since the driver assistance system is configured and / or set up to carry out any of the methods described above, the features described in the context of any of these methods accordingly apply to the driver assistance system, and vice versa, the features described in the context of any of the driver assistance systems apply to the methods.

[0050] The driver assistance system is preferably additionally set up to determine the position of objects in the vehicle's surroundings using at least two ultrasonic sensors and to assign the received ultrasonic echoes to object hypotheses representing the objects.

[0051] The proposed method and the proposed driver assistance system enable a robust high-level classification of objects into low objects that can be driven over and high objects that cannot be driven over. In particular, in difficult boundary situations, such as the edge of a curb, a higher discrimination accuracy can be achieved compared to known methods.

[0052] Advantageously, no additional sensors are required, so that it can be simply implemented using ultrasonic sensors commonly used in modern vehicles. Description of the Drawings

[0053] Embodiments of the invention are explained in more detail on the basis of the drawings and the following description.

[0054] Figure 1 : A vehicle with a driver assistance system according to the invention is shown in a side view. Detailed Description

[0055] The drawings only schematically show the subject matter of the invention.

[0056] Figure 1 A vehicle 1 on a road 22 is shown in a side view. The vehicle 1 includes a driver assistance system 100, which has ultrasonic sensors 10 and a controller 20. In Figure 1 the side view, only one ultrasonic sensor 10 is visible, but the vehicle 1 includes a plurality of ultrasonic sensors 10. In Figure 1 the illustrated embodiment, the driver assistance system 100 also has a display device 28 connected to the control unit 20. The control unit 20 is furthermore set up to carry out braking interventions. This is shown in Figure 1 the illustration by the connection of the control unit 20 to the pedal 29.

[0057] In Figure 1 the ultrasonic sensor 10 that is visible is mounted at the rear of the vehicle 1. The ultrasonic sensor 10 has a field of view 30 within which the ultrasonic sensor 10 can identify objects such as traffic signs 26 or bumps 24. In Figure 1Another elevation 24' (which is closer to the vehicle 1 than elevation 24) shown therein can no longer be recognized by the ultrasonic sensor 10 in the Figure 1 shown situation, because this other elevation 24' is outside the field of view 30 of the ultrasonic sensor 10. When the vehicle 1 approaches the elevation 24, the height classification of the elevation 24 can be recognized by a change in amplitude or a change in the detection behavior. If the vehicle 1 drives slowly backwards in the direction of the elevation 24, the elevation leaves the field of view 30 of the ultrasonic sensor 10 at a specific point, which can be recognized by a strong decrease in the amplitude of the corresponding ultrasonic echo. The moment when the ultrasonic sensor 10 can no longer recognize the elevation 24 or the distance between the elevation 24 and the vehicle 1 at this moment can be used to infer the height of the elevation 24. If the elevation 24 is a high object, similar to the traffic sign 26, it does not leave the field of view 30 of the ultrasonic sensor 10 when approaching. This situation of leaving the field of view 30 when approaching is only possible for low, usually drivable objects.

[0058] However, since the area that can reflect the ultrasonic waves of the ultrasonic sensor 10 is relatively small, and thus due to the relatively small amplitude of the received ultrasonic echo, the traffic sign 26 cannot be reliably classified as a high object based only on the amplitude. Therefore, other criteria must be considered for use.

[0059] According to the invention, a height indicator H is determined for each of the emitted ultrasonic pulses, which is preferably given by the number of received ultrasonic echoes, the amplitude of the received ultrasonic echoes, the signal correlation of the received ultrasonic echoes, and the distance d assigned to the received ultrasonic echoes. In addition, a histogram with n classes K i is created for the values of the height indicator H, where the classes K 1 to K h represent values characterizing high objects, and the classes K (h+1) to K n represent values characterizing low objects. Finally, an indicator Q is determined for the histogram, which indicates the probability of the presence of a high object. The indicator Q is the quotient of the sum of the classes K i characterizing high objects weighted by the weighting factor b i from i = 1 to h and the sum of all classes K i weighted by the weighting factor bi from i = 1 to n.

[0060] Subsequently, the value of the indicator Q thus obtained and preferably continuously updated is used as an indicator indicating the presence of a non-drivable high object.

[0061] The invention is not limited to the embodiments described herein and the aspects emphasized therein. On the contrary, within the scope defined by the claims, a large number of modifications can be made within the scope of professional operations in this field.

Claims

1. A method for height classification of an object in the surroundings of a vehicle (1) in the case of using an ultrasonic sensor (10), which ultrasonic sensor emits ultrasonic pulses and receives ultrasonic echoes reflected by the object, wherein, During the height classification, a distinction is made between drivable low objects and non-drivable high objects, characterized in that a height indicator H is determined for the emitted ultrasonic pulses, and a histogram with n classes K i of the values of the height indicator H is created, where the classes K 1 to K h represent values characterizing high objects, the classes K (h+1) to K n represent values characterizing low objects, and an indicator Q for determining the probability of the presence of high objects is determined, and the indicator Q is given by the following formula: where b i is a weighting factor determined empirically for the category K i and the height indicator H is determined based on the number of received ultrasonic echoes, the amplitude of the received ultrasonic echoes, the signal correlation of the received ultrasonic echoes, the distances assigned to the received ultrasonic echoes, and combinations of at least two of these parameters.

2. The method according to claim 1, characterized in that, the detection threshold of the ultrasonic sensor (10) is matched to the current interference level cl such that the ratio of misclassifying ultrasonic echoes as object echoes is constant.

3. The method according to claim 1 or 2, characterized in that, when creating the histogram, according to the relationship K i = K i + a i * H Perform weighting, where a i is a weighting factor determined based on experience.

4. The method according to claim 3, characterized in that, The weighting factor a determined empirically is selected in such a way i : that normalization is carried out, by means of which the height indicator H is made independent of the currently set sensitivity of the ultrasonic sensor (10).

5. The method according to claim 3 or 4, characterized in that, The weighted factor a obtained according to experience i is a function a related to the interference level cl and the distance d of the object i (cl, d).

6. The method according to any one of claims 1 to 5, characterized in that, The weighting factor b determined empirically i Describe the said category K i How persuasive it is for height classification 7. The method according to any one of claims 1 to 6, characterized in that, a confidence value is determined for the indicator Q, wherein the confidence value is related to the number of entries in the histogram.

8. The method according to any one of claims 1 to 7, characterized in that, when performing the height classification, at least one further classification parameter is taken into account in addition to the indicator Q, wherein the at least one further classification parameter is selected from the confidence value of the indicator Q, the amplitude of the ultrasonic echo, the gradient of the amplitude of the ultrasonic echo, the quotient of the distance difference and the difference formed by the traveled distance, and combinations of a plurality of these classification parameters.

9. The method according to any one of claims 1 to 8, wherein, the distances between each ultrasonic sensor (10) and the object reflecting the ultrasonic pulses in the surroundings are determined by means of at least two ultrasonic sensors (10) having at least partially overlapping fields of view (30), and the position of the reflecting object is determined by means of triangulation.

10. A driver assistance system (100) comprising at least one ultrasonic sensor (10) and a control device (20), characterized in that, the driver assistance system (100) is configured to carry out any one of the methods according to any one of claims 1 to 9.

Images