Ultrasonic ranging method

Abstract

The invention relates to a method (100) for distance measurement using an ultrasonic sensor unit (12), the method (100) comprising the following steps: a step (110) of emitting at least one ultrasonic pulse (14) from the ultrasonic sensor unit (12) towards a measurement object (16); a step (120) in which the ultrasonic sensor unit (12) receives at least one ultrasonic pulse (18) reflected from the measurement object (16); a step (130) of generating a measurement signal (20) on the basis of the received at least one ultrasonic pulse (18); a step (140) of evaluating the measurement signal (20) and determining the distance between the ultrasonic sensor unit (12) and the measurement object (16) using a classifier or a regressor trained by training data.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method for distance measurement using an ultrasonic sensor unit. The invention also relates to a corresponding ultrasonic sensor unit. BACKGROUND

[0002] Ultrasonic sensor units are widely known in the art and are used in various application scenarios. For example, many modern passenger cars are equipped with one or more ultrasonic sensor units for measuring distances to objects, in particular to pedestrians or vehicles. Such sensor units are used, inter alia, for parking assistance systems, which measure the distance to a preceding or following vehicle (or other objects) and issue a warning if the distance is below a prescribed minimum distance, thus simplifying the operation of parking into a parking space and driving out of a parking space.

[0003] Ultrasonic sensor units are known in various designs. They comprise a transmitting unit and a receiving unit, which can be designed independently of one another or in an integrated fashion as a transceiver. The transmitting unit is designed to emit one or more ultrasonic pulses toward a measurement object. The receiving unit acquires the ultrasonic pulses reflected by the measurement object and generates a corresponding measurement signal. The sensor unit further comprises an evaluation unit for analyzing the generated measurement signal. By evaluating the peaks in the measurement signal, the distance between the sensor unit and the measurement object can be determined on the basis of the speed of sound and the time required for the ultrasonic pulse to propagate from the sensor unit to the measurement object and back to the sensor unit.

[0004] Figure 1 An example of ultrasonic distance measurement in the prior art is schematically illustrated. A vehicle 10 has an ultrasonic sensor unit 12, which is designed to emit ultrasonic pulses 14. The ultrasonic pulses 14 propagate at the speed of sound toward a measurement object 16. The measurement object can be, for example, a vehicle (in particular a car or a bicycle), a pedestrian, or a wall. The emitted ultrasonic pulses 14 are reflected by the measurement object 16. The ultrasonic sensor unit 12 acquires the reflected ultrasonic pulses 18 and generates a measurement signal 20 from the intensity of the received ultrasonic pulses. The total propagation distance of the ultrasonic pulses is thus twice the distance between the ultrasonic sensor unit 12 (or the vehicle) and the measurement object 16. Figure 1(bottom) also shows a measurement signal 20 generated by the ultrasonic sensor unit 12. The measurement signal 20 describes the amplitude (A) of the reflected ultrasonic pulse 18 and contains a useful signal 21 with a peak 22, the position of which depends on the distance between the ultrasonic sensor unit 12 and the measurement object 16. The time at which the peak 22 is received, taking into account the speed of sound, corresponds to twice the distance between the ultrasonic sensor unit 12 and the measurement object 16. By evaluating the time at which the peak is received, the distance between the ultrasonic sensor unit 12 and the measurement object 16 can be determined. The height of the peak 20 depends on the intensity of the transmitted ultrasonic pulse, the material properties of the measurement object (reflectivity of the object) and the distance between the ultrasonic sensor unit 12 and the measurement object 16. The measurement signal 20 also contains a noise signal 24, which is caused by the inherent properties of the detection unit of the ultrasonic sensor unit 12 and by secondary reflections of the ultrasonic pulse on the ground and other various surfaces. Although Figure 1 The distance measurement described in the above-mentioned document is very reliable in the region > 50 cm, but it has been shown in practice that it is very difficult to measure a measurement object in the near range region (in particular distances < 50 cm, distances < 20 cm and distances < 10 cm) using the above-mentioned method. This is because the receiving unit of the ultrasonic sensor unit 12 enters a saturation state in the near range region 26 (also referred to as dead zone in the context of the present application), which includes, for example, distances of 0 to 20 cm between the ultrasonic sensor unit 12 and the measurement object 16. In this near range region, no reliable ultrasonic measurement can be made according to the method described in the above-mentioned document. Figure 1 The method described in the above-mentioned document cannot be used to determine the exact position of the measurement object in the scenario shown in

[0005] Figure 2 The problem described in connection with Figure 1 is described. In the scenario shown in Figure 2 the measurement object 16 is located in the near range region 26. The distance between the ultrasonic sensor unit 12 and the measurement object 16 is less than 50 cm (for example 10 to 20 cm). The measurement signal caused by the reflection of the ultrasonic pulse on the measurement object 16 is superimposed with a signal generated due to saturation of the detection unit of the ultrasonic sensor unit 12. As a result, the measurement signal 20 shown in Figure 1 and Figure 2 cannot be used to determine the exact position of the measurement object in the scenario shown in Figure 2 .

[0006] To further illustrate the above-mentioned problem, Figure 3 measurement signals recorded using the method described in Figure 1 and Figure 2 are shown. Figure 3 Two measurement signals 20 are shown. Each measurement transmits an ultrasonic pulse train containing one or more pulses. These measurement signals are envelope signals.

[0007] In the first measurement shown in (a) of Fig. 1, Figure 3 a dead zone or near range region 26 (also referred to as blind zone) can be seen. Peaks falling into this dead zone cannot be evaluated using the measurement methods of the prior art, because these peaks fall into the saturation region of the detection unit and thus do not provide a detectable signal. During the first measurement, a multiple reflection region 30 (also referred to as multiple reflection region) can also be seen (see (a) of Fig. 1 ), in which region individual peaks caused by multiple reflections of the ultrasound pulses emitted by the ultrasound sensor unit are formed. These peaks cannot be evaluated using the methods of the prior art. Figure 3 Figure 3 The measurement signal 20 shown in (a) of Fig. 1 has a first peak 28, which can be evaluated using the conventional measurement methods, since it lies outside the dead zone. This peak is detected at a point in time corresponding to a distance of 180 mm between the ultrasound sensor unit and the measurement object. Since the first peak 28 occurs outside the dead zone, the distance between the ultrasound sensor unit and the measurement object can be correctly determined using the methods described in the prior art.

[0008] In contrast, (b) of Fig. 1 shows the measurement signal 20 of a second measurement, in which the first peak occurs within the blind zone. In this case, the useful signal disappears into the saturation region, so that the distance between the ultrasound sensor unit and the measurement object cannot be correctly determined. Figure 3 SUMMARY As described above, the disadvantage of the ultrasonic distance measurement methods of the prior art is that the distance of near range objects cannot be reliably measured. On the basis of this disadvantage, it is an object of the present application to provide an ultrasonic distance measurement method which also enables reliable distance measurement in the near range region, wherein the near range region preferably comprises a distance of 0 to 50 cm, 0 to 20 cm or 0 to 10 cm between the ultrasound sensor unit and the measurement object.

[0009] In order to achieve the above-mentioned object, the present application proposes a method for distance measurement using an ultrasound sensor unit, wherein the method comprises the following steps:

[0010] emitting at least one ultrasound pulse from the ultrasound sensor unit towards a measurement object;

[0011] receiving at least one ultrasound pulse reflected from the measurement object by the ultrasound sensor unit;

[0012] generating a measurement signal on the basis of the received at least one ultrasound pulse; and

[0013]

[0014] ​​The measurement signal is evaluated and a classifier or regressor trained with training data is used to determine the distance between the ultrasonic sensor unit and the measurement object, so that the distance can be estimated.

[0015] With the classifier or regressor trained with training data, the characteristic shape of the measurement signal can be evaluated, so that the distance between the ultrasonic sensor unit and the measurement object can also be determined even in the near distance region. The method according to the application thus provides an innovative ultrasonic measurement method, which avoids the problem of measurement signal saturation, so that the range of distance measurement is extended to the near distance region. Another advantage of the application is that a higher measurement data rate can be provided since the method according to the application does not require a specific short distance measurement.

[0016] In the method according to the application, the ultrasonic pulses can be emitted, for example, in the form of ultrasonic pulse trains, each pulse train containing 8, 16 or n individual pulses. The ultrasonic pulses reflected from the measurement object are received by the ultrasonic sensor unit, so that the ultrasonic sensor unit generates a (ultrasonic) measurement signal describing the amplitude of the received ultrasonic pulses. The generation of the measurement signal can include the generation of (unfiltered) raw data describing the intensity and phase of the received ultrasonic pulses. Alternatively, the generation of the measurement signal can include the generation of an envelope signal.

[0017] The ultrasonic sensor unit comprises a transmitting unit and a receiving unit, which can be designed separately or integrated as a transceiver. The transmitting unit is designed to emit ultrasonic pulses, while the receiving unit is designed to receive ultrasonic pulses reflected from the measurement object and to generate a measurement signal on the basis of the received ultrasonic pulses.

[0018] The measurement signal can then be fed into a trained classifier or regressor, which has "learned" to recognize measurement signal characteristic curves for different distances in the training phase. The classifier or regressor thus uses the knowledge about the measurement signal characteristic shape acquired during the training phase in order to subsequently infer the distance between the ultrasonic sensor unit and the measurement object during the measurement phase. By using the classifier or regressor, the conventional measurement signal evaluation method, which often leads to difficulties, especially in the case of objects in the near distance measurement region (especially less than 50 cm or less than 20 cm), can be dispensed with.

[0019] In the method according to the application, the measurement signal can comprise an unfiltered amplitude signal, an envelope signal or an IQ signal. The amplitude signal can directly represent the intensity of the received ultrasound waves as a function of time. The amplitude signal can be generated directly by the ultrasound sensor, wherein the received signal is converted into a digital signal by means of an analog-digital converter. The envelope signal can be generated using methods known from the prior art. For example, for this purpose the amplitude signal can be used, wherein the maximum amplitudes (peaks) of the amplitude signal are connected to one another.

[0020] In some embodiments of the method according to the application, the classifier can be designed as an artificial neural network, a support vector machine or a decision tree. Furthermore, the classifier can be designed as a naive Bayes classifier or a k-nearest neighbor classifier. By using the above-mentioned classifiers, a conventional analysis of the measurement data can be dispensed with, so that the above-mentioned problems associated with the ranging in the near range are avoided.

[0021] In some embodiments of the method according to the application, a regressor designed as a linear regressor or a polynomial regressor can be used.

[0022] Furthermore, in some embodiments of the application, the trained classifier or regressor can be trained with training data, wherein the training data comprise a plurality of measurement signals and distance values associated with the plurality of measurement signals. The measurement signals can comprise amplitude signals, envelope signals or IQ signals (in-phase and quadrature signals), respectively. For example, a distance value of 5 cm can be associated with a first measurement signal, while a distance value of 8 cm can be associated with a second measurement signal, and so on. The training data can contain several thousand measurement signals each associated with a corresponding distance value, so that the classifier can be trained during the training phase and learn which distance values can be associated with which measurement signals or which characteristic shape of a measurement signal. In particular, the classifier can learn to evaluate the characteristics of a measurement signal caused by multiple reflections, so as to infer the distance between the ultrasound sensor unit and the measurement object.

[0023] Furthermore, in order to achieve the above-mentioned objects, the present application proposes a method for ranging with an ultrasound sensor unit, the method comprising the following steps:

[0024] training a classifier, wherein the classifier is preferably designed as an artificial neural network;

[0025] emitting a plurality of ultrasound pulses from the ultrasound sensor unit towards a measurement object;

[0026] receiving ultrasound pulses reflected from the measurement object by the ultrasound sensor unit;

[0027] generating a measurement signal on the basis of the received ultrasound pulses; and

[0028] The measurement signal is evaluated with the classifier trained by the training data and the distance between the ultrasonic sensor unit and the measured object is determined.

[0029] In particular, training the classifier or regressor can comprise the following steps:

[0030] Training the classifier or regressor with training data, wherein the training data comprises ultrasonic measurement data and distance values associated with the measurement data.

[0031] In some preferred embodiments of the method according to the present application, the training data can comprise ultrasonic measurement data describing the amplitude of the received ultrasonic pulse as a function of time. In this case, preferably, the ultrasonic measurement data can be extracted from the raw measurement data generated by the ultrasonic sensor unit. In particular, the raw measurement data can be described as a vector containing N measurement values, and the ultrasonic measurement data used for training the classifier or regressor can be extracted from the vector by disregarding the first M values of the vector. M can in particular be 10%, 15% or 20% of N. As explained in connection with Figures 1 to 3 The first values of the measurement signal do not contain any information relevant for the evaluation, as they only reflect the saturation state of the detection unit of the ultrasonic sensor unit. By extracting the ultrasonic measurement data from the raw measurement data and discarding the part of the measurement signal that is related to the blind zone, compact ultrasonic measurement data can be provided, which makes the overall training process more efficient. This is particularly advantageous when the training data contains thousands or tens of thousands of measurement signals. In this case, the required training time can be significantly reduced.

[0032] Furthermore, in the method according to the present application, the training data can comprise a plurality of measurement signals, each containing an unfiltered amplitude signal, an envelope signal or an IQ signal. While the present application is not limited to using the above-mentioned measurement signals for training the classifier or regressor, preliminary studies have shown that using unfiltered amplitude signals, envelope signals and IQ signals can allow for an accurate classification of the measurement data and thus for an accurate distance determination.

[0033] In some embodiments of the method according to the present application, the training data can comprise a plurality of measurement signals and distance values associated with the plurality of measurement signals, wherein the measurement signals have a plurality of peaks representing multiple reflections of the ultrasonic pulse emitted by the ultrasonic sensor unit. Thus, the information of the multiple reflections can be utilized for determining the distance between the ultrasonic sensor unit and the measured object in the near range. If the classifier or regressor is trained using measurement signals having a plurality of peaks representing multiple reflections, the measurement signals fed into the classifier during the measurement phase should of course also comprise measurement signals having a plurality of peaks representing multiple reflections.

[0034] Furthermore, the method according to the application can be designed for determining the distance between an ultrasonic sensor unit and a measuring object in a near range, wherein the near range comprises a distance between the ultrasonic sensor unit and the measuring object in the range of 0 cm to 50 cm, preferably in the range of 0 cm to 20 cm, particularly preferably in the range of 0 cm to 10 cm. The methods known from the prior art have proven to lead to measurement errors, especially in the near range. By using the method according to the application, a more reliable near range measurement can be achieved.

[0035] In the method according to the application, in particular, the method can be used for determining the distance between a vehicle and a measuring object. The ultrasonic sensor unit can preferably be integrated into the vehicle. In particular, the ultrasonic sensor unit can be integrated into the bumper of the vehicle. In the method according to the application, the ultrasonic sensor unit can be implemented as a rear sensor, a front sensor or a side sensor.

[0036] Furthermore, in the method according to the application, the method can be implemented as a computer-implemented method or a machine-implemented method. In the computer-implemented way of the application, the above-mentioned method steps can be executed by a computing unit, wherein the implementation is executed by a computer program product. In the machine-implemented way of the application, the above-mentioned method steps can be implemented using hardware components, in particular logic gates.

[0037] Furthermore, in order to solve the above-mentioned problems, an ultrasonic sensor unit for measuring a distance to a measuring object is proposed, wherein the ultrasonic sensor unit comprises:

[0038] a transmitting unit for transmitting a plurality of ultrasonic pulses;

[0039] a receiving unit for receiving the ultrasonic pulses reflected from the measuring object; and

[0040] an evaluation unit designed to:

[0041] generate a measurement signal from the received ultrasonic pulses; and

[0042] evaluate the measurement signal using a classifier or regressor trained on training data and determine the distance to the measuring object.

[0043] In the ultrasonic sensor unit according to the application, the transmitting unit and the receiving unit can be designed as separate units or as an integrated transmitting and receiving unit (transceiver).

[0044] In the ultrasonic sensor unit according to the application, the evaluation unit can be designed to generate the measurement signal in the form of an unfiltered amplitude signal, an envelope signal or an IQ signal and to evaluate the measurement signal using a classifier and to determine the distance to the measuring object.

[0045] In some embodiments of the ultrasonic sensor unit, the classifier can also be designed as an artificial neural network, a support vector machine or a decision tree.

[0046] Furthermore, in the ultrasonic sensor unit according to the application, the classifier can be trained with training data, wherein the training data comprise a plurality of measurement signals and distance values associated with the plurality of measurement signals, and the measurement signals have a plurality of peaks representing a plurality of reflections of an ultrasonic pulse emitted by the ultrasonic sensor unit.

[0047] In the ultrasonic sensor unit according to the application, the evaluation unit can be designed to determine a distance to a measurement object in a near distance region, wherein the near distance region comprises distances between the ultrasonic sensor unit and the measurement object in the range of 0 cm to 50 cm, preferably in the range of 0 cm to 20 cm, particularly preferably in the range of 0 cm to 10 cm.

[0048] Furthermore, to achieve the above-mentioned object, a computer program product is proposed, comprising computer instructions executable by a computing unit, wherein the computer instructions cause the computing unit to perform the steps of any of the above-mentioned methods when the computer instructions are executed by the computing unit. BRIEF DESCRIPTION OF DRAWINGS

[0049] The application will be explained in more detail below with reference to the drawings.

[0050] Figure 1 An ultrasonic distance measurement method according to the prior art is shown.

[0051] Figure 2 A measurement signal recorded using the method described in Figure 1 is shown.

[0052] Figure 3 A measurement signal recorded using the method described in Figure 1 and Figure 2 is shown.

[0053] Figure 4 A schematic diagram of the method according to the application is shown.

[0054] Figure 5 A schematic diagram of the training process of the classifier using the method according to the application is shown.

[0055] Figure 6 A schematic diagram of the distance measurement using the trained classifier based on the method according to the application is shown.

[0056] Figure 7 A schematic diagram of the ultrasonic sensor unit according to the application is shown. DETAILED DESCRIPTION

[0057] Figures 1 to 3 The ultrasonic distance measuring method according to the prior art is described and explained in detail at the outset.

[0058] Figure 4 An embodiment of a method 100 for distance measurement using an ultrasonic sensor unit according to the application is schematically shown. In this embodiment, in a first method step 110, the ultrasonic sensor unit emits a plurality of ultrasonic pulses towards a measurement object. The emitted ultrasonic pulses are then reflected on the surface of the measurement object and acquired by the ultrasonic sensor unit in a second method step 120. In a third method step 130, the ultrasonic sensor unit generates a measurement signal on the basis of the received ultrasonic pulses. This measurement signal can be designed as an unfiltered amplitude signal. Alternatively, for example, the ultrasonic sensor unit can generate an envelope signal. In a fourth method step 140, the generated measurement signal is evaluated and the distance between the ultrasonic sensor unit and the measurement object is determined. A classifier trained with training data is used for the evaluation and the distance determination. The training data can contain a plurality of measurement signals and distance values associated with these measurement signals. Thereby, the classifier is able to "learn" in the training process which measurement signals are associated with which distance values. This process is also referred to as supervised learning. During the measurement phase, the measurement signal is fed into the trained classifier, wherein the classifier outputs a distance value that can be associated with the input measurement signal. Here, in particular, the classifier can utilize the characteristic curve of the measurement signal formed by the multiple reflections of the ultrasonic pulses. Thus, in the method according to the application, not only the first peak is detected and evaluated, but also all information contained in the measurement signal is evaluated. Thereby, a reliable distance measurement can be provided, which provides reliable measurement results even in close-range measurements.

[0059] Figure 5 A training process for training a classifier according to an embodiment of the method of the application is schematically shown. In this embodiment, in a first training step 210, a plurality of measurement signals is acquired. These measurement signals are acquired by the ultrasonic sensor unit in a plurality of measurement phases. In a second training step 220, a plurality of distance values is determined. These distance values are determined by the ultrasonic sensor unit in the measurement phases. In a third training step 230, a classifier is trained with the acquired measurement signals and the determined distance values. The classifier is trained in such a way that it is able to determine a distance value associated with a measurement signal. In this embodiment, the classifier is trained in such a way that it is able to determine a distance value associated with a measurement signal on the basis of the characteristic curve of the measurement signal. In a fourth training step 240, the trained classifier is stored in a memory of the ultrasonic sensor unit. Figure 5In the illustrated embodiment, an artificial neural network (ANN) is used. During training, training data 34 can be used to train the artificial neural network. Training data 34 may contain multiple measurement signals and distance values ​​associated with those measurement signals. For example, training data 34 may contain thousands or tens of thousands of measurement signals and their corresponding distance values, each distance value representing the distance between the measured object and the ultrasonic sensor unit during the corresponding measurement process. Generally, the richer the training data 34 used during training, the higher the accuracy of the classifier. Furthermore, it goes without saying that the type of measurement data (measurement signals) used during ranging should be the same as the type of measurement signals used during training. For example, if the classifier is trained using an envelope signal (and the distance values ​​associated with that envelope signal), then the envelope signal should also be used and fed into the classifier during the measurement process. According to some embodiments, training data 34, including the entire measurement signal, can be used to train the artificial neural network 32.

[0060] Figure 6 A schematic diagram illustrating distance measurement using a trained classifier based on an embodiment of the method according to the invention is shown. In this embodiment, as detailed above, the measurement signal 20 is generated by an ultrasonic sensor unit. Subsequently, the measurement signal 20 (designed in this case as an envelope signal) is fed into a pre-trained classifier, which... Figure 6 In the illustrated embodiment, it is implemented as an artificial neural network 32. (As in conjunction with...) Figure 5 As described, the classifier is pre-trained using training data and evaluates the input measurement signal 20 during the measurement phase. The classifier then outputs a distance value that can be associated with the pattern contained in the measurement signal 20 with the highest probability. Figure 6 The right figure in the diagram shows the distance determined using the method according to the invention. The figure compares the distance (d) between the measured object and the ultrasonic sensor unit with the actual distance (d). As can be seen from the figure, the method according to the invention can provide fairly reliable ranging in a short-range region, where the short-range region in this example includes a distance of approximately 0 cm to 40 cm between the ultrasonic sensors.

[0061] Figure 7An embodiment of an ultrasonic sensor unit 12 according to the application is shown schematically. The ultrasonic sensor unit 12 has a control device CTR which is designed to receive data, programs and / or commands from other computers, usually superior computers, via a data interface IF. These superior computers can be, for example, control units of a motor vehicle for controlling and monitoring the ultrasonic sensor unit 12 shown. The communication with these computers can take place via the data interface IF. The ultrasonic sensor unit 12 also has a storage unit 36 which can have a plurality of RAM (random access memory) storage elements and NVM (non-volatile memory) storage elements. A classifier, in particular which can be designed as an artificial neural network, can be stored in particular in the NVM storage elements. The measurement signals to be evaluated can preferably be temporarily stored in the RAM. The measurement data or measurement signals can be evaluated, for example, by the control device CTR which has a computing unit. Alternatively, the measurement data can also be evaluated by the input circuit DSI which also has a computing unit. The ultrasonic sensor unit 12 can also be provided with a total of two computing units, wherein a first computing unit is included in the control device CTR and a second computing unit is provided in the input circuit DSI. The first computing unit can be provided for communication with other computers, while the second computing unit is used for analyzing and evaluating the measurement data.

[0062] The control device CTR can transmit determined measurement values, error messages and test results to the superior computers. The control device CTR controls the digital signal generation unit DSO via a control signal SO. In addition, the control device CTR configures all other configurable sub-devices of the ultrasonic sensor unit 12. For the sake of clarity, Figure 7 The respective control lines and signals are not shown in the figure.

[0063] The control signal SO is preferably a data bus consisting of a plurality of digital signals. From the history and the control signal SO, the digital signal generation unit DSO generates the excitation signals for the subsequent signal path chains starting from the sub-devices of the ultrasonic sensor unit 12 which are located downstream in the signal path, as well as the useful signals, measurement signals and test signals. It is therefore proposed to design the digital signal generation unit DSO both as a measurement signal generator and as a test signal generator and test pattern generator. The digital signal generation unit DSO generates the excitation signals, the useful signals, the measurement signals and the test signals for the subsequent signal path chains as first digital signals SI. The first digital signals SI are preferably a digital data bus. In certain permitted states of the ultrasonic sensor unit 12, certain lines of the first digital signals SI can be in an inactive state, but in other permitted states of the ultrasonic sensor unit 12, they are in an active state.

[0064] The drive stage DR generates a second analog signal S2 from the first digital signal S1. Thus, the drive stage DR converts the first digital signal S1 into the analog second signal S1. The second analog signal S2 can consist of a plurality of second analog sub-signals. Here, too, not all sub-signals of the second analog signal S2 are necessarily active in all states of the ultrasonic sensor unit 12.

[0065] This does not necessarily mean only a digital-to-analog conversion of the digital value transmitted to the drive stage DR by the first digital signal S1. Rather, the drive stage can also contain more complex circuits, possibly with feedback, which change their active topology depending on the state of the ultrasonic sensor unit 12 and, if necessary, depending on the current time in the transmit / receive sequence. For example, such a transmit / receive sequence of the ultrasonic sensor unit can be divided into three phases. In a first phase (also referred to as a transmit phase) of an exemplary transmit sequence (ultrasonic sequence), the exemplary ultrasonic transducer TR is excited to mechanical oscillation, thereby transmitting an ultrasonic pulse as an output signal MS into the exemplary ultrasonic measurement channel CN. In this first phase, i.e. the transmit phase, the drive stage DR applies a measurement excitation to the ultrasonic transducer TR corresponding to the ultrasonic transmission frequency. The drive stage DR then transfers energy into the ultrasonic transducer TR.

[0066] In a subsequent second phase (i.e. an attenuation phase), the drive stage DR applies a measurement excitation to the ultrasonic transducer TR which is opposite to the oscillation frequency of the still oscillating ultrasonic transducer TR. This attenuates the oscillation of the ultrasonic transducer TR. The drive stage DR then extracts energy from the ultrasonic transducer TR. In this phase, the ultrasonic transducer TR transmits the output signal MS into the ultrasonic measurement channel CN in an external region located outside the ultrasonic sensor unit 12, wherein the signal radiation amplitude decreases gradually. In the ultrasonic channel CN in the external region located outside the ultrasonic sensor unit 12, there is usually one or more objects which usually generate a severely attenuated, delayed and distorted ultrasonic measurement signal echo. This echo is referred to below as the ultrasonic reception signal ES. In a third phase (i.e. a receive phase), the ultrasonic transducer TR is not driven by the drive stage DR. The drive stage DR does not extract energy from the ultrasonic transducer TR nor does it transfer any energy into the ultrasonic transducer TR. In this phase (i.e. the receive phase), the ultrasonic transducer TR can well receive the ultrasonic echo as a reception signal (ultrasonic reception signal) ES. The ultrasonic transducer TR is excited by the ultrasonic reception signal ES to vibrate and, due to its piezoelectric properties, generates a third analog signal S3. The third analog signal S3 can consist of a plurality of analog sub-signals.

[0067] The analog multiplexer AMX switches the third analog signal S3 to a fourth analog signal S4 in a predetermined state of the ultrasonic sensor unit 12. Which signal is switched through by the analog multiplexer AMX depends on the state of the ultrasonic sensor unit 12. In general, the analog multiplexer AMX is controlled by the control device CTR, which preferably controls and monitors the state and configuration of the ultrasonic sensor unit 12. Here too, not all sub-signals of the fourth analog signal S4 are necessarily valid in all states of the ultrasonic sensor unit 12. The analog input circuit AS receives the fourth analog signal S4. This reception can depend on the control signal S0 and the state of the ultrasonic sensor unit 12 and other factors. The specific way in which the analog input circuit AS receives the signal is preferably specified by the control device CTR using a corresponding control signal (not shown). The reception method in the analog input circuit AS preferably correlates with the excitation signal, measurement signal or test signal used (generated by the digital signal generation unit DSO and the drive stage DR) and the configuration of the ultrasonic sensor unit 12, which is preferably set by the control device CTR. For example, it is conceivable that depending on the specific case, analog filters, levels, gains, etc. are adjusted to accommodate the specified excitation signal, useful signal, measurement signal and test signal. This adjustment is preferably controlled by the control device CTR. The analog input circuit AS generates a fifth digital signal S5 from the fourth analog signal S4. The analog input circuit AS thus preferably also acts as an analog-digital converter ADC. The fifth digital signal S5 can contain a plurality of digital sub-signals. Here too, not all sub-signals of the fifth digital signal S5 are necessarily valid in all states of the ultrasonic sensor unit 12. The digital multiplexer DMX passes the fifth digital signal S5 as a sixth digital signal S6 in a predetermined state of the ultrasonic sensor unit 12. Which signal is passed by the digital multiplexer DMX as the sixth signal S6 can likewise depend on the state of the ultrasonic sensor unit 12. In general, the digital multiplexer DMX is controlled by the control device CTR, which preferably controls and monitors the state and configuration of the ultrasonic sensor unit 12. Here too, not all sub-signals of the sixth digital signal S6 are necessarily valid in all states of the ultrasonic sensor unit 12. The digital input circuit DSI receives the sixth digital signal S6 and processes it depending on the state of the ultrasonic sensor unit 12. The specific way in which the digital input circuit DSI receives the signal is preferably specified by the control device CTR using a corresponding control signal (not shown). The reception method in the digital input circuit DSI preferably correlates with the excitation signal, measurement signal or test signal used (generated by the digital signal generation unit DSO and the drive stage DR) and the selected reception method in the analog input circuit AS and the configuration of the ultrasonic sensor unit 12.For example, it is conceivable that the digital filter, in particular the matched filter, and the digital signal processing method can be adapted to the specified excitation signal, the useful signal, the measurement signal and the test signal and the selected reception method in the analog input circuit AS depending on the specific case. This adaptation is preferably controlled by the control device CTR. Thus, the digital input circuit DSI generates a seventh digital signal S7 from the sixth digital signal S6, wherein the seventh digital signal S7 should already contain the measurement result, the test result or other results. The other results can be, for example, error messages which are sent to the control device CTR. However, the control device CTR can also compare the measurement result and the test result with a target value or a tolerance interval of such target values and generate an error message if necessary. Thus, the digital input circuit DSI can not only have a signal conditioning function in the predetermined state of the measurement system, but can also be used as a test device for checking whether the signal processing chain of the ultrasonic sensor unit 12 is able to respond to the excitation from the digital signal generation unit DSO in such a way that the response corresponds to a predetermined value or a predetermined signal sequence in a known system configuration or deviates from this value or sequence by no more than a predetermined value. Here again, not all sub-signals of the seventh digital signal S7 necessarily have to be valid in all states of the measurement system.

[0068] List of reference signs

[0069] 10 vehicle

[0070] 12 ultrasonic sensor unit

[0071] 14 emitted ultrasonic pulse

[0072] 16 measurement object

[0073] 18 reflected ultrasonic pulse

[0074] 20 measurement signal

[0075] 21 useful signal

[0076] 22 peak

[0077] 24 noise signal

[0078] 26 near range region

[0079] 28 first peak

[0080] 30 multiple reflection region

[0081] 32 artificial neural network

[0082] 34 training data

[0083] 36 storage unit

[0084] 100 method

[0085] 110 first method step

[0086] 120 second method step

[0087] 130 third method step

[0088] 140 fourth method step

Claims

1. A method (100) of distance measurement using an ultrasonic sensor unit (12), wherein, The method (100) comprises: a step (110) of emitting at least one ultrasonic pulse (14) from the ultrasonic sensor unit (12) towards a measurement object (16); a step (120) of receiving at least one ultrasonic pulse (18) reflected from the measurement object (16) by the ultrasonic sensor unit (12); a step (130) of generating a measurement signal (20) based on the received at least one ultrasonic pulse (18); a step (140) of evaluating the measurement signal (20) and determining a distance between the ultrasonic sensor unit (12) and the measurement object (16) using a classifier or regressor trained by training data.

2. The method (100) according to claim 1, characterized in that The measurement signal (20) comprises an unfiltered amplitude signal, an envelope signal and / or an IQ signal.

3. The method (100) according to claim 1 or 2, characterized in that, The classifier is designed as an artificial neural network (32), a support vector machine or a decision tree.

4. The method (100) according to any one of claims 1 to 3, characterized in that, The trained classifier or regressor is trained by training data (34), wherein the training data (34) comprises a plurality of measurement signals (20) and distance values associated with the plurality of measurement signals (20).

5. The method (100) according to claim 4, characterized in that The training data (34) comprises a plurality of measurement signals (20) comprising an unfiltered amplitude signal, an envelope signal or an IQ signal, respectively.

6. The method (100) according to claim 4 or 5, characterized by, The training data (34) comprises a plurality of measurement signals (20) and distance values associated with the plurality of measurement signals (20), wherein the plurality of measurement signals (20) have a plurality of peaks (22) representing multiple reflections of the ultrasonic pulse (14) emitted by the ultrasonic sensor unit (12).

7. The method (100) according to any one of claims 1 to 6, characterized in that, The method (100) is for determining a distance between an ultrasonic sensor unit (12) and a measurement object (16) within a near distance range (26), wherein the near distance range (26) comprises a distance between the ultrasonic sensor unit (12) and the measurement object (16) in a range of 0 cm to 50 cm, preferably in a range of 0 cm to 20 cm, particularly preferably in a range of 0 cm to 10 cm.

8. The method (100) according to any one of claims 1 to 7, characterized by, The method (100) is for determining a distance between a vehicle (10) and a measurement object (16).

9. The method (100) according to any one of claims 1 to 8, characterized in that, The method (100) is implemented as a computer-implemented method or a machine-implemented method (100).

10. An ultrasonic sensor unit (12) for measuring a distance to a measurement object (16), comprising: an emission unit for emitting at least one ultrasonic pulse (14); a reception unit for receiving at least one ultrasonic pulse (18) reflected from the measurement object (16); and an evaluation unit designed to: generate a measurement signal (20) based on the received at least one ultrasonic pulse (18); and evaluate the measurement signal (20) using a classifier or regressor trained by training data (34) and determine a distance to the measurement object (16).

11. The ultrasonic sensor unit (12) according to claim 10, characterized by The measurement signal (20) comprises an unfiltered amplitude signal, an envelope signal or an IQ signal.

12. The ultrasonic sensor unit (12) according to claim 10 or 11, characterized by The classifier is designed as an artificial neural network (32), a support vector machine or a decision tree.

13. The ultrasonic sensor unit (12) according to any one of claims 10 to 12, characterized in that The classifier or the regressor is trained by training data (34), wherein the training data (34) comprises a plurality of measurement signals (20) and distance values associated with the plurality of measurement signals (20), and the plurality of measurement signals (20) has a plurality of peaks (22) representing a plurality of reflections of the ultrasonic pulse (14) emitted by the ultrasonic sensor unit (12).

14. The ultrasonic sensor unit (12) according to any one of claims 10 to 13, characterized in that The evaluation unit is designed for determining a distance to a measurement object (16) in a near distance region, wherein the near distance region comprises a distance between the ultrasonic sensor unit (12) and the measurement object (16) in a range of 0 cm to 50 cm, preferably in a range of 0 cm to 20 cm, particularly preferably in a range of 0 cm to 10 cm.

15. A computer program product comprising computer instructions executable by a computing unit, which, when executed by the computing unit, cause the computing unit to perform the steps of any of the methods (100) according to any of claims 1 to 9.

Images