WO2020249375A1

WO2020249375A1 - Method for classifying the height of objects in the area surrounding a vehicle, and driver assistance system

Info

Publication number: WO2020249375A1
Application number: PCT/EP2020/064300
Authority: WO
Inventors: Stevens WANG; Juergen Schmidt; Tom Reimann; Michael Schumann
Original assignee: Robert Bosch Gmbh
Priority date: 2019-06-12
Filing date: 2020-05-22
Publication date: 2020-12-17
Also published as: CN113950629A; DE102019208508A1

Abstract

The invention proposes a method for classifying the height of objects in the area surrounding a vehicle (1) using ultrasonic sensors (10). The ultrasonic sensors (10) emit ultrasonic pulses and receive ultrasound echoes reflected by objects. The height classification makes a distinction between short objects that can be driven over and tall objects that can't be driven over. It is proposed in the method that a height indicator H is determined for an emitted ultrasonic pulse and is given by the number of received ultrasound echoes, the amplitudes of the received ultrasound echoes, the signal correlations of the received ultrasound echoes, and distances assigned to the received ultrasound echoes. A bar chart with n classes K_i is also created for the values of the height indicator H, the classes K₁ to K_h representing values of the height indicator H which characterise a tall object, and the classes K(h+1) to Kn representing values of the height indicator H which characterise a short object, and an indicator Q being determined, which specifies a likelihood for the presence of a tall object. The indicator Q is a quotient of a sum, weighted by the weighting factor b_i, across i of 1 to h classes K_i which specify tall objects, and the sum, weighted by the weighting factor bi, across i of 1 to n of all classes K_i. A further aspect of the invention relates to a driver assistance system (100) which is designed to carry out the method.

Description

description

title

Method for the height classification of objects in the vicinity of a

Vehicle and driver assistance system

The invention relates to a method for the height classification of objects in the vicinity of a vehicle using ultrasonic sensors which emit ultrasonic pulses and receive ultrasonic echoes reflected by objects again, with the height classification distinguishing between low objects that can be driven over and high objects that cannot be driven over. Another aspect of the invention relates to a driver assistance system which is set up to carry out the method.

State of the art

Modern vehicles are equipped with a variety of driver assistance systems that the driver of the vehicle when running

support various driving maneuvers. Furthermore are

Driver assistance systems known, which warn the driver of dangers in the area. For their function, the driver assistance systems require precise data about the surroundings of the vehicle and in particular about objects that are in the surroundings of the vehicle.

Ultrasound-based object localization methods are often used, in which two or more ultrasonic sensors are used. The

Ultrasonic sensors each send out ultrasonic pulses and receive ultrasonic echoes reflected from objects in the vicinity. The distance between a reflecting object and the respective sensor can be determined from the transit time of the ultrasonic pulses until the corresponding ultrasonic echo is received and the known speed of sound. If an object is in the field of view of more than one ultrasonic sensor, i.e. the distance to the object can be determined by more than one ultrasonic sensor, the exact position of the reflecting object relative to the sensors or to the vehicle can also be determined using lateration algorithms. Due to the ever-increasing fields of vision and sensitivities of the sensors, objects on the floor such as curbs, thresholds or manhole covers can also be increasingly recognized. For the driver assistance systems to function correctly, it is important to be able to differentiate between collision-relevant objects, such as posts, walls or traffic signs, and objects that cannot be driven over, such as curbs, thresholds or manhole covers that are not relevant for a collision.

From DE 10 2009 046 158 A1 a method for recognizing objects with a low height is known. Provision is made here to continuously detect a distance to an object by means of distance sensors and to check whether the object continues to be detected by the distance sensors when the vehicle approaches if it falls below a predetermined distance or whether it disappears from the detection range of the distance sensors. If it is recognized that the object disappears from the detection range of the distance sensors when it approaches, the object is classified as an object with a low height.

In the prior art, methods are also known which make use of the fact that tall and extensive objects generally do not have a single, clearly defined reflection point and therefore focus on a single one

Ultrasonic pulse several reflections and thus several in time

can cause successive ultrasonic echoes. With a high object, for example, a reflex runs directly horizontally, i.e. parallel to the floor from the sensor to the object and back again. Another reflex is thrown back from the throat between the floor and the tall object. This second ultrasound echo arrives after the first ultrasound echo, since a longer path has to be covered from the installation position of the sensor to the transition between the object and the floor than the direct path running parallel to the floor. It is also known that certain objects such as bushes or pedestrians, but also flat objects such as drain grids or manhole covers, cause a large number of reflections which are noticeable as a noise-like signal as an echo.

DE 10 2007 061 235 A1 describes a method for classifying the height of objects using statistical scattering, which is caused in particular by multiple reflections of the measurement signal. The problem with the known methods for height classification is that the parameters used for classification, such as in particular the number of received reflections of an object, are heavily dependent on further conditions such as the current interference level.

Disclosure of the invention

A method is proposed for classifying the height of objects in the vicinity of a vehicle using ultrasonic sensors. The ultrasonic sensors send out ultrasonic pulses and receive ultrasonic echoes reflected from objects. The height classification distinguishes between low objects that can be driven over and high objects that cannot be driven over. In the method, it is provided according to the invention that an altitude indicator H is determined for an emitted ultrasonic pulse. A histogram with n classes K is also created for the values of the height indicator H, with the classes KL to K, representing values of the height indicator H that characterize a tall object, and the classes K (h + u to K _n values of the height indicator H. which characterize a low object, and an indicator Q is determined which indicates a probability of the presence of a high object The indicator Q is a quotient of a sum weighted with the weighting factor b, over i from 1 to h of the classes K, which characterize tall objects, and the sum weighted with the weighting factor b, over i from 1 to n of all classes K. Thus the indicator Q is given by the formula

The height indicator H is determined based on the number of received ultrasonic echoes, the amplitudes of the received ultrasonic echoes, the signal correlations of the received ultrasonic echoes, the distances assigned to the received ultrasonic echoes and combinations of at least two of these parameters. All the parameters mentioned are preferably included in the determination of the altitude indicator H. In the method, at least one of the ultrasonic sensors of a vehicle sends out an ultrasonic pulse, which is also referred to as a single “shot” of this ultrasonic sensor, and the ultrasonic sensors of the vehicle then receive reflections from objects in the vicinity

Ultrasonic echoes. Depending on its size, height and nature, an object can not only lead to a single ultrasonic echo, but several ultrasonic echoes also occur as a rule. For example, a tall object that cannot be driven over usually delivers at least two ultrasonic echoes in response to an emitted ultrasonic pulse. A first ultrasonic echo runs horizontally from the object back to the ultrasonic sensor

Ultrasonic sensor. A second ultrasonic echo is produced by the throat between the object and the floor and is based on the

Height difference a longer way back than the first ultrasonic echo.

The ultrasonic echoes can be received by one or more ultrasonic sensors of the vehicle.

The method provides for the multiple ultrasonic echoes originating from an object to be combined to form the height indicator. The height indicator is preferably assigned to the object from which the

Ultrasonic echoes originate. In addition to the number of ultrasonic echoes received in response to an emitted ultrasonic pulse, their amplitudes, their signal correlations and / or those relating to the

Ultrasonic echoes a certain distance. The signal correlation denotes the correlation of the corresponding ultrasonic echo with the

emitted ultrasonic pulse. The distance assigned to an ultrasonic echo is derived from the known speed of sound in air and the respective transit time between the transmission of the ultrasonic pulse and the reception of the

Ultrasonic echoes determined. In particular, the deviations in the distances between the individual ultrasonic echoes can be considered, for example the difference between the distance of the first ultrasonic echo and the distance of the last ultrasonic echo of a multiple reflection.

A grouping of several received ultrasonic echoes into several

Reflections of an object preferably take place on a lower signal level in which there is still no information about the object position. For this purpose, for example, a temporal capture window is specified, with echoes within this capture window being treated as multiple reflections of an object. To determine the height indicator, the selected parameters such as the number of received ultrasonic echoes, the amplitudes of the received ultrasonic echoes, the signal correlations of the received ultrasonic echoes, the distances assigned to the received ultrasonic echoes and combinations of at least two of these parameters are linked to the height indicator H.

The method is preferably carried out continuously, so that the height indicator H is determined continuously. A histogram of the values obtained for the altitude indicator H is created for further evaluation. The histogram has n classes. For example, n is chosen in the range from 4 to 100, particularly preferably in the range from 6 to 20. For example, 7 or 16 classes are chosen.

The height indicator H can be in the form of continuous values, with each class being assigned a predetermined value range of the height indicator. Alternatively, the height indicator can be in the form of discrete values or statements, with one or more discrete values or statements being able to be assigned to a class. A statement can, for example, be a classification to the effect that the linked parameters result in a height indicator that indicates a specific situation or a specific type of object. For example, a height indicator can state that the parameters considered typically represent a curb, wall, bush or pedestrian. Of the examples mentioned, only the curb can be driven over.

A height indicator that is characteristic of a curb is distinguished, for example, by one or two received ultrasonic echoes, the distance between which is small. The amplitude of the second ultrasonic echo is very low and the amplitude of the first echo depends on the distance.

A height indicator characteristic of a wall is distinguished, for example, by two ultrasonic echoes, the first ultrasonic echo having a very high amplitude and a high correlation and the second ultrasonic echo having a low amplitude and a low correlation. The the

Distances associated with ultrasonic echoes are larger than with a curb. A height indicator characteristic of a bush is emerging

for example from two to four ultrasonic echoes, all echoes having a very low amplitude and correlation.

A height indicator characteristic of a pedestrian is distinguished, for example, by two to three ultrasonic echoes with medium amplitudes and low correlations. Typically there are variations of the den

Distances associated with ultrasonic echoes are very large (up to 40cm between the first and last echo).

On the basis of the histogram, which is continuously updated, the indicator Q is then continuously determined, which is given from the weighted sum of all classes which indicate a high object divided by the weighted sum of all classes.

Whether a class indicates a tall object that cannot be driven over, or, conversely, indicates a low object that cannot be driven over, is preferably determined empirically on the basis of tests. Accordingly, the

Parameter h, which indicates which of the classes indicate a tall object, chosen empirically.

The weighting factors bi, which are used for weighting the sum, are also preferably determined empirically through tests. The weighting factors b can then be stored, for example, in a memory of a control device which implements the method.

For example, the weighting factors b can be stored in the form of a lookup table.

The empirically determined weighting factor b preferably indicates how meaningful the class K is for the height classification.

The value of the indicator Q obtained in this way and preferably continuously updated is then used as an indicator for the presence of a high object that cannot be driven over. In the method it is preferably provided that a detection threshold of the ultrasonic sensors is adapted to an instantaneous interference level cl such that a rate for incorrect classification of an ultrasonic echo as the echo of an object is constant.

The interference level cl is usually dominated by the so-called clutter, which is caused by ground echoes. The amplitude and number of ground echoes or of the clutter is strongly dependent on the nature of the ground. With a smooth surface there are fewer ground echoes than with an uneven surface such as gravel. In addition, other ambient noises or ultrasonic pulses from other ultrasonic sensors can generate an interference level. Accordingly, provision is preferably made for the detection threshold to be adapted to the currently prevailing environmental conditions, so that the detection threshold value is lowered in the event of low ambient noise or a low number of ground echoes and vice versa in a noisy environment with a lot of interference signals and large

Noise and / or a high number of ground echoes, for example through a rough surface such as gravel, the detection threshold is raised.

To adapt the detection threshold, for example, an algorithm can be used which regulates the detection threshold in such a way that a constant false alarm rate is achieved (constant false alarm rate, CAFR).

Weighting according to the relationship is preferred when creating the histogram

Ki = Ki + ai * H, where a, is an empirically determined weighting factor.

In this case, the empirically determined weighting factor a is preferably selected such that a normalization takes place through which the height indicator H is independent of a currently set sensitivity of the ultrasonic sensors.

The empirically determined weighting factor a is preferably a function ai (cl, d) dependent on the interference level cl and the distance d of the object. The weighting factors a and the function ai (cl, d) are preferably in the form of a table or in the form of a lookup table in a memory of a

Stored control device that implements the method.

A weighting factor for the formation of the histogram that is specified as a function ai (cl, d) creates a normalization factor that eliminates the influences of distance d and the interference level cl, in particular the clutter. It is advantageous when combined with an adaptive

Threshold value for the ultrasonic sensors makes the system more sensitive with a low interference level and more insensitive with a high interference level. With ultrasonic sensors set to be more sensitive, the values obtained for the height indicator H would increase for all objects without a corresponding normalization, that is to say for both low and high objects. Furthermore, when the interference level is low, certain value ranges of the altitude indicator are not meaningful, so that they may be masked out with a weighting factor of 0.

The determined distance d of the object also has an influence on the

Significance of the height indicator. In the close range up to a distance of about 50 cm, ultrasonic echoes, which originate from the throat between an object and the floor, are outside the field of view of the ultrasonic sensors. The ultrasonic sensors are only able to receive ultrasonic echoes from an object that is within their field of view. The interference level caused by the clutter reaches its maximum in the range between 80 cm and 120 cm. In the far range above 200 cm, the amplitudes of some ultrasonic echoes drop below the detection threshold of the ultrasonic sensors. In these problematic areas, the weighting factors a are correspondingly preferably chosen to be lower than outside these

Distance ranges.

In the case of multiple reflections, that is, when more than one ultrasonic echo is assigned to an object, the distance assigned to the first ultrasonic echo is preferably viewed as the distance d of the object.

A confidence value is preferably determined for the indicator Q, the confidence value being dependent on the number of entries in the histogram. With a small number of entries, i.e. with few received

Ultrasonic echoes, the informative value of the indicator Q is still low and growing only with an increasing number of received ultrasonic echoes and correspondingly with an increasing number of entries in the histogram.

In addition to the indicator Q, at least one further classification parameter is preferably taken into account when performing the height classification, the at least one further classification parameter being selected from the confidence value of the indicator Q, the amplitude of the ultrasonic echoes, a gradient of the amplitude of the ultrasonic echoes, a quotient

Distance difference and the difference between the distance traveled and combinations of several of these classification parameters.

The gradient of the amplitude of the ultrasonic echoes is the change in the amplitude of the ultrasonic echoes associated with this object observed when the distance between an object and the vehicle changes. It is provided, for example, to continuously detect a distance to an object by means of the ultrasonic sensors and to check whether the object continues to be detected by the distance sensors when the vehicle approaches if it falls below a predetermined distance or whether it disappears from the detection range of the distance sensors. If, for example, it is recognized that the object disappears from the detection range of the distance sensors when approaching, the object is identified as a low,

traversable object classified.

With the quotient of the distance difference and the difference from the driven

Distance, a change in the distance measured over the transit time is considered in relation to a change in the distance to the object due to a movement of the vehicle. For a side echo, which

is thrown back from the throat between the floor and a high object, for example, the Pythagorean theorem gives the distance D calculated from the travel time of the ultrasonic echo

D = V d ² + h ² where d is the distance between the object and the sensor and h is the installation height of the sensor above the ground. The distance d of the object to the sensor is always the shortest distance, which results from a direct, parallel to the ground running line results. Since the sensor height h is constant, the calculated distance D changes less than the actual distance d to the object.

In order to be able to assign a received echo to an object in the environment, it is preferred to create object hypotheses and the received ones

Assign ultrasonic echoes to an object hypothesis. An assignment of ultrasonic echoes to an object hypothesis can, for example, under

Use of lateration. Such a position determination by means of lateration is advantageous for the use of the data obtained in one

Driver assistance system. For the implementation of the invention

However, such an assignment is not required for the classification procedure.

An interpretation of multiple reflections as ultrasonic echoes of a single object is preferably carried out using a temporal capture window, with all ultrasonic echoes received within this capture window being viewed as multiple reflections of the same object.

Preference is given to carrying out the lateration over at least two

Ultrasonic sensors with at least partially overlapping fields of view

Distances between the respective ultrasonic sensor and ultrasonic pulses

reflective objects in the area determined, with a

The position of the reflecting objects is determined by means of lateration.

In particular, measured values obtained consecutively in time, i.e. distance values determined in succession, can be assigned to one and the same object hypothesis if lateration shows that the position of the respective object reflecting the ultrasound corresponds to the position assigned to an object hypothesis or is close to it. By evaluating the total amount of measurements assigned to an object hypothesis or of the distances and positions determined with the ultrasonic sensors, conclusions can then also be drawn about the contour of the object. For example, if the vehicle moves evenly in one direction and all positions assigned to an object hypothesis are on a line or all positions of all ultrasonic sensors of a bumper assigned to an object hypothesis are on a line, it can be concluded that the object assigned to this object hypothesis is located is a large object such as a wall or other vehicle. On the other hand, if the position does not change approximately, it is probably a point Object before which, seen in the plane parallel to the ground, has only a small geometric extent. For example, it is a post, a traffic sign or a characteristic corner of another object, such as a vehicle corner or a corner of a house, or even one

Curb corner. Such a combination of individual measured distances to form extended objects is described in DE 10 2007 051 234 A1, for example.

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 set up to carry out one of the methods described herein.

Since the driver assistance system is designed and / or set up to carry out one of the methods, features described in the context of one of the methods apply correspondingly to the driver assistance system and vice versa, features described in the context of one of the driver assistance systems apply conversely to the methods.

The driver assistance system is preferably also set up to use at least two ultrasonic sensors to determine the position of objects in the vicinity of a vehicle and to determine the received ultrasonic echoes of one

Assign the object hypothesis that represents the object.

Advantages of the invention

By the proposed method and proposed

Driver assistance system enables a robust height classification of objects into low objects that can be driven over and high objects that cannot be driven over.

Especially in difficult borderline cases, such as one

Curb edging can be increased compared to known methods

Selectivity can be achieved.

Advantageously, no additional sensors are required, so that simple implementation with the ultrasonic sensors usually used in modern vehicles is possible. Brief description of the drawings

Embodiments of the invention are explained in more detail with reference to the drawings and the following description.

FIG. 1 shows a vehicle with a driver assistance system according to the invention in a view from the side.

Embodiments of the invention

The figure represents the subject matter of the invention only schematically.

FIG. 1 shows a vehicle 1, which is located on a road 22, in a view from the side. The vehicle 1 includes a driver assistance system 100 with an ultrasonic sensor 10 and a control unit 20. In the side view of FIG. 1, only one ultrasonic sensor 10 is visible, but the vehicle 1 includes several ultrasonic sensors 10. In the embodiment shown in FIG. 1, the driver assistance system 100 also has Via a display device 28 connected to the control unit 20. The control unit 20 is also set up to display a

Perform braking intervention. This is in the illustration of Figure 1 by a

Connection of the control device 20 to a pedal 29 is shown.

The ultrasonic sensor 10 visible in FIG. 1 is mounted on the rear of the vehicle 1. The ultrasonic sensor 10 has a field of view 30 within which the ultrasonic sensor 10 is able to recognize objects such as the traffic sign 26 or a threshold 24. The further threshold 24 'also shown in FIG. 1, which is closer to the vehicle 1 than the threshold 24, can no longer be recognized by the ultrasonic sensor 10 in the situation shown in FIG. 1, since this further threshold 24' is outside of the field of view 30 of the ultrasonic sensor 10 is located. A height classification of the threshold 24 can be recognized when the vehicle 1 approaches the threshold 24 by a change in the amplitude or a change in the detection behavior. If the vehicle 1 drives slowly backwards in the direction of the threshold 24, this will leave the field of view 30 of the ultrasonic sensor 10 at a certain point, which can be recognized by a sharp drop in the amplitude of a corresponding ultrasonic echo. The point in time or the distance of the threshold 24 to the vehicle 1 at the point in time at which this can no longer be recognized by the ultrasonic sensor 10, can then be used to draw conclusions about the height of the threshold 24. If the threshold 24 were to be a tall object, similar to that

Traffic sign 26, the field of view 30 of the ultrasonic sensor 10 cannot be left when approaching. Leaving the field of view 30 in this way when approaching is only possible for low objects that can generally be driven over.

A reliable classification of the traffic sign 26 as a tall object is due to the comparatively small area, which ultrasound of the

Ultrasonic sensor 10 can reflect, and thus due to the comparatively small amplitudes of the received ultrasonic echoes not only possible on the basis of the amplitude. Further criteria must therefore be used.

According to the invention, a height indicator H is determined for each emitted ultrasonic pulse, which is preferably given by the number of received

Ultrasonic echoes, the amplitudes of the received ultrasonic echoes, the

Signal correlations of the received ultrasonic echoes and the received ones

Distances associated with ultrasonic echoes d. Furthermore, a histogram of the values of the height indicator H is created with n classes K, the classes KL to K, representing values that characterize a tall object, and the classes K (h + u to K _n represent values that characterize a low object Finally, an indicator Q is determined for the histogram, which indicates a probability of the presence of a high object. The indicator Q is a quotient of one and the

Weighting factor b, weighted sum over i from 1 to h of the classes K, which indicate high objects, and the sum weighted with the weighting factor b, over i from 1 to n of all classes K.

The value of the indicator Q obtained in this way and preferably continuously updated is then used as an indicator of the presence of a high level which cannot be traversed

Object used.

The invention is not restricted to the exemplary embodiments described here and the aspects emphasized therein. Rather, within the range specified by the claims, a large number of modifications are possible that are within the scope of expert knowledge.

Claims

Expectations

1. Procedure for the height classification of objects in the vicinity

of a vehicle (1) using ultrasonic sensors (10) which emit ultrasonic pulses and receive ultrasonic echoes reflected by objects again, a distinction being made in the height classification between low objects that can be driven over and high objects that cannot be driven over, characterized in that for an transmitted ultrasonic pulse an altitude indicator H is determined, a histogram of the values of the altitude indicator H is created with n classes K, the classes K _L to K, values

represent that characterize a high object, and the classes K (h + i) to K _n represent values that represent a low object

mark, and an indicator Q is determined, the one

Indicates probability of the presence of a high object, where the indicator Q is given by

where bi is an empirically determined weighting factor for class K and the altitude indicator H is determined based on the number of received ultrasonic echoes, the amplitudes of the received ultrasonic echoes, the signal correlations of the received ones

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 a

The detection threshold of the ultrasonic sensors (10) is adapted to an instantaneous interference level cl such that a rate for incorrect classification of an ultrasonic echo as the echo of an object is constant.

3. The method according to claim 1 or 2, characterized in that when creating the histogram a weighting according to the relationship Ki = Ki + ai * H

takes place, where a, is an empirically determined weighting factor.

4. The method according to claim 3, characterized in that the

empirically determined weighting factor a is selected so that a normalization takes place, by means of which the height indicator H is independent of a currently set sensitivity of the ultrasonic sensors (10).

5. The method according to claim 3 or 4, characterized in that the empirically determined weighting factor a is a function a, (cl, d) which is dependent on the interference level cl and a distance d of the object.

6. The method according to any one of claims 1 to 5, characterized in that the empirically determined weighting factor b indicates how meaningful the class K is for the 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, the confidence value being dependent on 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 in addition to

Indicator Q at least one further classification parameter is taken into account, the at least one further

The classification parameter is selected from the confidence value of the indicator Q, the amplitude of the ultrasonic echoes, a gradient of the amplitude of the ultrasonic echoes, a quotient of the distance difference and the difference from the distance traveled and combinations of several of these classification parameters.

9. The method according to any one of claims 1 to 8, wherein at least two ultrasonic sensors (10) with at least partially overlapping fields of view (30) distances between the respective ultrasonic sensor (10) and ultrasonic pulses reflecting objects in the environment are determined and a position determination of the reflecting objects takes place by means of lateration.

10. 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 set up for the execution of one of the methods according to one of claims 1 to 9.