DE102019208506A1 - Method for determining the direction of incidence of a special signal sound source - Google Patents

Abstract

Method for determining a direction of incidence of a special signal sound source in relation to a platform by means of at least two data signals that were generated by at least two respectively assigned sound transducers, the sound transducers each being assigned a main angle of incidence, with the following steps: processing the at least two data signals, wherein Signal components of the special signal sound source are highlighted; - Calculating a signal value from the processed at least two data signals, depending on a time range adapted to the special signal; - Determining the direction of incidence of the special signal sound source, the at least two signal values with their main angles of incidence and at least one sound pressure angle function to one another be adjusted.

Description

The invention relates to a method for determining a direction of incidence of a special signal sound source in relation to a platform by means of at least two sound transducers.

State of the art

So far there are no commercial systems with special signal location in use in road traffic, because due to the driving dynamics and the associated disruptions, a location system with z. B. classic "beamforming" process due to the strong driving dynamics not simply transferred to automotive technology.

Disclosure of the invention

For at least partially automated vehicles, the perception of the acoustic environment is also of great importance. In addition to the detection and classification of an acoustic signal, it is important to know the direction of a special signal sound source, such as from emergency vehicles, in order to assess the relevance of this acoustic signal for the behavior of an at least partially automated vehicle.
Especially in cities with frequently occurring special signals such. B. San Francisco, it is important to reduce the "idle time" of the at least partially automated vehicle in order to be able to initiate a corresponding, possibly automatic action and / or z. B. to choose a suitable route. Supporting the driver by specifying the direction of a special signal is also important for driver assistance systems, since it is difficult for a driver to locate these signals in the closed interior of a vehicle. A practical application of a location system requires a high sensitivity and sufficient accuracy with a negligibly small number of errors in order to avoid an unnecessary or possibly incorrect reaction of the at least partially automated vehicle in traffic.

The present invention discloses a method for determining a direction of incidence of a special signal sound source in relation to a platform, a device, a computer program and a machine-readable storage medium according to the features of the independent claims. Advantageous configurations are the subject matter of the dependent claims and the description below.

The invention is based on the knowledge that a group of sound transducers, which have different directional characteristics of their sound sensitivity for different solid angles of the sound field around a platform, can be particularly sensitive to direction detection.

The method according to the invention for determining a direction of incidence of a special signal sound source in relation to a platform uses at least two data signals that were generated by at least two respectively assigned sound transducers, the sound transducers each being assigned a main angle of incidence.
The at least two data signals are processed in one step of the method, signal components of the special signal sound source being highlighted.

In a further step, a signal value is calculated from the processed at least two data signals, depending on a time range adapted to the special signal.
The direction of incidence of the special signal sound source is calculated by adapting the at least two signal values with their main angles of incidence and at least one sound pressure angle function to one another.

In this entire description of the invention, the sequence of process steps is shown in such a way that the process is easy to understand. The person skilled in the art will recognize, however, that many of the method steps can also be carried out in a different order and lead to the same result. In this sense, the sequence of the process steps can be changed accordingly.

With the direction of incidence of a special signal sound source in relation to a platform, the angle can be specified that results between a straight-ahead direction of the platform and a bearing direction of the sound source from the perspective of this platform. In particular, this platform can be a mobile platform in which the straight-ahead direction can be derived from a main direction of movement. An example of such a mobile platform would in particular be an at least partially automated vehicle. Such a mobile platform can, however, also represent a vehicle that is equipped with a driver assistance system.

Acoustic signals, which are emitted by special signal sound sources and identify dangerous situations, for example, play a major role in road traffic, since every road user has to react to such a signal, depending on the situation in which he is in road traffic. An important example of such signals are the signals from emergency vehicles, which are typically designed as special sirens and generate different tone sequences. In addition, special signal sound sources can in particular be all sound sources that have a certain relevance for road users in connection with road traffic. Furthermore, a special signal or a special signal sound source can be understood to mean any signal or any signal sound source whose direction of incidence can be determined in relation to a system using the method described here.

Such special acoustic signals can be generated by means of sound transducers such. B. microphones are converted into electrical signals, the transducer largely acoustically directly or indirectly with a sound environment to be monitored z. B. can be coupled to a road traffic situation. Sound transducers z. B. are arranged outside a vehicle cabin, have the advantage over an arrangement in the interior of the vehicle that the acoustic signals are not attenuated by the cabin boundaries and are thus more directly coupled to the sound environment. However, with a suitable design, a sound transducer that is arranged inside the cabin can be connected directly to the external sound environment, for example with an acoustic coupling.

Such sound transducers can be based on all principles of converting sound into electrical signals, such as in particular on electrodynamic, piezoelectric or capacitive principles. Furthermore, the transducers can be in different designs, such. B. be designed as a microphone, loudspeaker or as a motion-sensitive sensor, etc. The electrical signals that are generated by such a sound transducer can in particular be electrical signals, for example voltage signals, which have been converted by means of a transfer function proportional to the sound pressure at the input of the sound transducer.

Sound transducers that provide data signals, in particular electrical signals, for this method, which are then optionally converted from analog signals of the sound transducer into digital data that follow one another in time, i.e. data signals in the form of audio data, output either synchronous signals or audio data or the signals or In a preprocessing stage, audio data are synchronized in such a way that they are provided for the method, with regard to a resulting error, sufficiently synchronously with simultaneously incident sound signals. Thus, the data signals are time-synchronized with regard to the sound conversion.

The electrical signals from the sound transducers can, for example, be converted into successive digital audio data, e.g. B. can be converted by means of an electronic analog-to-digital converter circuit, the digital audio data then containing the corresponding signal content, which for example characterize dangerous situations, digitally encoded.

The signals from the sound transducers, which are arranged at different points on the platform, for example installed at different locations in a vehicle, can be set up in such a way that the installation location of the sound transducer itself no longer has any influence on the level of the detected amplitude, provided that the respective sound transducer a correspondingly identical sound source is impressed. I.e. the sound transducers at different installation locations of the platform or a vehicle are normalized with regard to sensitivity.

The analog-to-digital conversion of the electrical signals can in particular be carried out in such a way that acoustic signals, which characterize a dangerous situation, cover a relevant frequency range of e.g. B. 250 Hz to 8 kHz and thus the signal components are recorded by means of a sampling or sample rate twice the highest frequency to be recorded and converted into digital audio data. A higher sampling rate can increase the accuracy of the conversion. The data signals that were generated by the respective sound transducer are assigned to each of the at least two sound transducers.

A main angle of incidence is assigned to each of these at least two sound transducers. This main angle of incidence is derived from the different directional characteristics of the sound transducers in relation to the platform for different solid angles of a sound field around the platform. The at least two sound transducers are set up to have different effective directional characteristics for different solid angles of the sound field around the platform. Such an effective directional characteristic for a sensitivity of the sound transducers can result from a directional characteristic of the sound transducers themselves, that is to say of the sound transducer systems, together with a corresponding alignment of these sound transducers with respect to the platform. Furthermore, such an effective directional characteristic can be brought about by the installation location of the sound transducers on the mobile platform, for example by sound-shielding effects, such as boundary surfaces of the platform, or by a combination of these effects.

For example, a microphone can be arranged as centrally as possible on the front of the vehicle, a microphone can be arranged on the left or right side of the vehicle, for example in the B-pillar or on the sill, and a microphone can be arranged on the The rear of the vehicle should in particular be arranged as centrally as possible. Each of the microphones can be placed in an area of the vehicle in which dynamic pressure occurs while driving, in order to reduce background noise due to the airstream.

As an example, a vehicle can be cited that each has such a sound transducer perpendicular to the four side surfaces of the vehicle and both due to the inherent directional characteristics of the sound transducers with the corresponding alignment of the sound transducers due to the mounting on the four side surfaces, as well as the effect of the respective Vehicle area, a resulting directional characteristic for different solid angles of the sound field around the platform.

The method described processes each of the at least two data signals, in particular digital audio data, in such a way that signal components of the special signal sound source are emphasized.
In particular, these signal components can be signal components of the special signal sound source which correspond to signals of the special signal sound source with at least one characteristic frequency range.

Special signal sound sources from emergency vehicles in road traffic have a defined and specified frequency range from which a special signal is composed. The highlighting of those signal components of the audio data that contain this characteristic frequency range improves the signal-to-noise ratio of the relevant signal components for the further process steps.
This highlighting of the corresponding signal components can already be done by the sound transducers themselves, it can be done by coupling the sound transducers to the sound field, it can also be done by analog-digital conversion of the signals, or by analog or digital filtering.

The processing of the at least two data signals with respect to special signal components in order to in particular special signals, e.g. B. from emergency vehicles to protrude from the sound field, can also be carried out using other filter techniques. For example, frequency bands of a certain width around a fundamental frequency and the first three harmonics of the fundamental frequency can be filtered out of the audio signal. I.e. Frequency components outside these frequency bands are attenuated in order to filter out the frequency bands mentioned. I.e. the acoustic signals are filtered in order to reduce the signal to the frequency range of the signal to be located. Depending on the signal, additional filters can be used to e.g. to emphasize tonal components.

Signal values for each of the at least two data signals are calculated, this calculation being carried out as a function of a time range adapted to the special signal. This means that a specific signal can be extracted from the sound field to determine the direction of incidence. Because many of the relevant special signals have a specific time behavior, such as a periodicity. Such a signal value can thus take place by calculating an effective value of the processed data signals or digital audio signals for each of the at least two data signals. For the following signal from a German emergency vehicle, this time range can be 1 second, for example.

The signal value can, for example, be matched to the special signal to be detected by calculating the effective value of the data signals, in that the averaging associated with the calculation of the effective value is carried out over an amount of the digital audio data over a correspondingly selected time range.
For example, the rms value can be calculated using the square root of the integral or the sum over the squares of the individual chronologically successive data signals or digital audio data, all digital audio data being used for squaring over a time range of one second. The time interval for the averaging can thus coincide with a repetition frequency of the special signal or is at least of the same order of magnitude.
Such signal values can also be calculated using other mathematical methods if they lead to a particular special signal being highlighted from the set of signals in the sound field.

A sound pressure angle function can indicate how the sound pressure varies depending on a horizontal angle of a sound source to the platform when a sound source in the far field of the platform or a sound transducer assumes different angles relative to the platform or a sound transducer. In both cases, the angle is specified, ie the horizontal angle in relation to the platform. A sound pressure angle function therefore indicates a distribution of a sound pressure depending on a horizontal direction of incidence of a sound source in relation to the platform.
The respective main angle of incidence of a sound transducer results from the angle at which the respective sound transducer has its maximum sensitivity. In other words, the main angle of incidence of a sound transducer is the angle in relation to the platform at which a sound source, at which only the horizontal angle is varied in relation to the platform, generates the maximum output signal for this sound transducer.

Each of the at least two sound transducers is thus assigned a main angle of incidence which corresponds to the horizontal angle of a sound source at which the sound source produces a maximum output signal in the respective sound transducer.

The adaptation or comparison of the at least two signal values and the at least one sound pressure angle function takes place in such a way that either the at least two signal values with their associated main angles of incidence are adapted or compared to the at least one sound pressure level function or the at least one sound pressure angle function is adapted or compared to the at least two signal values to determine the direction of incidence of the sound source. A combination of these two different method steps can also be provided.

With the method described, an identification of the signal to be detected from the acoustic environment is achieved both by processing the at least two data signals and by calculating the signal values with a time range adapted to the special signal. This can result in a low error rate for determining the direction of incidence, since certain signal components of the special signal sound source are referenced. Using this method, a very simple approach for determining the direction of sound sources in relation to a platform is given. The specified method improves the indication of direction, which can be given solely on the basis of a volume comparison when, for example, four microphones are placed around a vehicle.
Compared to other methods such as the “beamforming” method, which evaluates a phase shift, or the “Time Difference of Arrival (TDOA) method, the computational effort to determine the direction of incidence is not very expensive.

According to one embodiment of the method, it is proposed that the sound pressure angle function be parameterized and adapted to the at least two signal values, the adaptation being carried out in such a way that a first standard from the at least two signal values with the main angles of incidence assigned to the generating sound transducers, and the sound pressure angle function is minimized.

The main angles of incidence that belong to the sound transducers that generated the corresponding data signals for the respective signal values are therefore assigned to the respective signal values. The signal values for the adaptation to the sound pressure angle function thus correspond to an amplitude value with an abscissa which are determined by the associated main angles of incidence.

For the adaptation, parameters of the parameterized sound pressure angle function can be varied in such a way that a sum of distances between points, which are defined by the at least two signal values and their respectively assigned main angle of incidence of the sound transducers, is minimal from the parameterized sound pressure angle function.

The direction of the special signal sound source is determined by the abscissa value of a maximum value of the correspondingly adapted sound pressure angle function.

The described method for adapting the sound pressure angle function is associated with little computational effort.

According to a further embodiment of the method, it is proposed that the parameterized sound pressure angle function be described with a trigonometric function or in particular with a linearized trigonometric function with a periodicity of 360 degrees of the angle of the direction of incidence and that the parameterized sound pressure curve be characterized by an amplitude and a phase of the trigonometric functions will.

The periodicity of 360 degrees can be traced back to the definition given above of the angle of incidence in relation to the platform together with the angular symmetry of the arrangement of the sound source and platform. Such a trigonometric function can be adapted to the at least two signal values with the associated main angles of incidence by varying the amplitude and the phase. The trigonometric function is in particular a sine function or a cosine function or a combination of both. A freely selectable standard, for example a distance standard, can be used for the adaptation.
The relationship between the quantities and parameters of such a sound pressure angle function can be described with the following formula (1), where α indicates the horizontal angle of incidence, A the maximum amplitude, φ and B the fit parameters. $$f\left(\alpha \right)=\mathrm{A.}\cdot \text{sin}\left(\frac{2\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\alpha +\phi \right)+\mathrm{B.}$$

The variance that can be calculated from such an adaptation can be used as a measure of the accuracy of the direction determination.

In the linearized form of formula 1, the fit parameters a, b and c result in addition to the horizontal angle of incidence α. $$\text{f}\left(\alpha \right)=\text{a}\cdot \text{sin}\left(\frac{2\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\cdot \alpha \right)+\text{b}\cdot \text{cos}\left(\frac{2\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\cdot \alpha \right)+c$$

According to a further embodiment of the method, it is proposed that the parameterized sound pressure angle function in the form of a trigonometric function has a phase that is dependent on the angle of the direction of incidence. The parameterized sound pressure curve is characterized by the amplitude and a phase dependent on the angle of the direction of incidence.
Due to the geometric relationships of a platform, for example in the form of the rectangular design of a vehicle, with a simple trigonometric function, due to the shading by the boundary surfaces, the sound pressure curve can possibly be better adapted with such a function than with a trigonometric function, has solid phase.

The relationship between the quantities and parameters of such a sound pressure angle function, in which the phase also depends on the angle of incidence, can be described with the following formula (2), where φ (α) gives the phase dependent on the angle of incidence. $$f\left(\alpha \right)=\mathrm{A.}\cdot \text{sin}\left(\frac{2\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\alpha +\phi \left(\alpha \right)\right)+\mathrm{B.}$$

geschrieben werden kann, in der sie performanter berechnet werden kann.Whereby the formula (2) also in the form: $$\text{f}\left(\alpha \right)=\text{a}\cdot \text{sin}\left(\frac{\text{2}\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\cdot \alpha \right)+\text{b}\cdot \text{cos}\left(\frac{2\mathrm{<figure-callout\; id="360"\; label="\pi "\; filenames="DE102019208506A1\_0005.png"\; state="\{\{state\}\}">\pi </figure-callout>}}{360}\cdot \alpha \right)+\text{c}$$

can be written in which it can be calculated more efficiently.

According to a further embodiment of the method, it is proposed that the sound pressure angle function be determined by measurement technology, in which a sum signal of the entirety of all sound transducers is determined by means of a far-field measurement or a near-field measurement. A parameterized form of this sound pressure angle function is then adapted to the signal values as described. The abscissa value of such a sound pressure angle function then determines the angle of incidence. Depending on the amplitude of the signal, one of the two sound pressure angle functions can then be selected as a fit function. I.e. Different sound pressure angle functions are used for different distances between the sound source and the sound transducers, and the corresponding sound pressure angle function is selected according to the sound amplitude. The sound pressure angle function determined in a far-field measurement can be used for low amplitudes; the other sound pressure angle function can be used for high amplitudes.

According to a further embodiment of the method, it is proposed that each of the at least two sound transducers is assigned a sound pressure angle function which has a maximum at the respective main angles of incidence. The at least two signal values and the at least two sound pressure angle functions are adapted to one another in that a parameter for the adaptation is precisely one angle assigned to the signal values at which a second standard of the at least two signal values and the associated sound pressure angle functions is minimized.

The sound pressure angle function assigned to the respective sound transducers always relates to an angle relative to the platform when referring to the angle.
Exactly one angle is therefore sought for each of the signal values for which a second standard, which is based in each case on a distance standard for the respective signal values from their associated sound pressure angle functions, and minimizes the sum of these distance standards. The direction of the sound source is the angle for which the total of all measured sound pressures can best be fitted into the total of the sound pressure angle functions.

To record this sound pressure angle function for each sound transducer, an acoustic sound source can be moved circularly around the vehicle. The sound pressure angle function for each sound transducer can then be specified from the result of such a measurement.

In the far field, the sound pressure angle function can be determined for each microphone. The measured amplitudes of the four microphones can then be fitted into these four created curves while driving.
The measured function is the actual, real sound pressure angle function, while the trigonometric function is just a simple approximation of it.

According to a further embodiment of the method, it is proposed that the second standard be calculated from a sum of distances between the respective signal values and the assigned sound pressure angle functions.

According to a further embodiment of the method, it is proposed that the minimization of the first or the second standard takes place under the additional condition that at least one parameter for adapting at least one corresponding parameter from a previous one Determination of the direction of incidence may not deviate by more than a predefined amount.

It can thus be achieved that outliers when determining the direction of incidence are filtered out for a method that follows the determination of the direction, and thus the fit parameters, but in particular the phase shift, depend on the temporal history of the fit parameters. If further actions are initiated on the basis of the direction determination, the method of direction determination becomes more robust through this embodiment of the method.

According to a further embodiment of the method, it is proposed that the minimization of the first or the second standard take place under the additional condition that at least one of the parameters for adaptation is calculated using a Kalman filter and at least one parameter of a previous determination of a direction of incidence. The Kalman filter is simply connected downstream. It is a temporal filter that filters out outliers.

According to a further embodiment of the method, it is proposed that the sound pressure angle function of each of the at least two sound transducers is determined by means of a far field measurement and / or a near field measurement.

This increases the accuracy of the determination of a sound source that is typically located in the far field of a sound transducer or a platform. If the sound pressure angle function was determined both in the far field and in the near field, the method can provide that either one or the other sound pressure angle function is used depending on the strength of the signal from the sound source at the sound transducer. This means that if the signal from the sound source is measured particularly intensively at the sound transducer, the sound pressure curve that was determined for the near field is used for the adaptation and if the signal is detected weaker, the sound pressure curve that was determined for the far field can be used for the adjustment can be used.

According to a further embodiment of the method, it is proposed that the directional characteristic of the respective sound transducers be set up by means of an individual directional characteristic of the respective sound transducers themselves and / or an arrangement of the respective sound transducers in relation to the platform.

This results in the possibility of achieving the necessary directional characteristic both via an individual directional characteristic of the respective sound transducer and also by finding suitable positions in relation to the platform, such as a vehicle, in order to provide or optimize the directional characteristic. In particular, the boundary surfaces of the platform, for example the chassis of a vehicle, can be used to provide a corresponding directional characteristic via effects such as sound insulation by means of the side walls.

According to a further refinement of the method, it is proposed that, depending on the determined direction of incidence, a control signal for controlling an at least partially automated vehicle and / or a warning signal for warning a vehicle occupant is transmitted.

If, for example, the deployment signal of a special vehicle is determined directly in front of an at least partially automated vehicle, then, for example, a route or a speed of the at least partially automated vehicle can be adjusted accordingly. Correspondingly, for example, when a deployment signal is detected behind the at least partially automated vehicle, a free path can be created in which the at least partially automated vehicle executes appropriate steering movements or, in the case of a driver assistance system, a vehicle occupant is informed or warned accordingly.

A device is specified which is set up to carry out a method as described above. With such a device, the method can easily be integrated into different systems.

A computer program is specified which comprises instructions which, when the program is executed by a computer, cause the computer to execute one of the methods described above. Such a computer program enables the described method to be used in different systems.

A machine-readable storage medium is specified on which the computer program described above is stored.

Figure list

• 1 eine Bestimmung einer Einfallsrichtung einer Schallquelle mit einer Schalldruckwinkelfunktion;
• 2 eine Bestimmung einer Einfallsrichtung einer Schallquelle mit vier Schalldruckwinkelfunktionen;
• 3 die Schritte des Verfahrens zur Bestimmung einer Einfallsrichtung.

Embodiments of the invention will be described with reference to 1 to 3 and explained in more detail below. Show it:

• 1 a determination of a direction of incidence of a sound source with a sound pressure angle function;
• 2 a determination of a direction of incidence of a sound source with four sound pressure angle functions;
• 3 the steps of the method for determining a direction of incidence.

To determine a direction of incidence of a special signal sound source in relation to a platform, four data signals are used in a first embodiment that were generated by four respectively assigned sound transducers, each of the four sound transducers being assigned a main angle of incidence.

In this exemplary embodiment, four sound transducers are assumed which are arranged symmetrically on the four sides of a platform. This results in symmetrical signals in relation to the platform. However, starting with a minimum number of two sound transducers, the method can be expanded to any number of sound transducers and can also be carried out with asymmetrically arranged sound transducers as long as the sound transducers have different main angles of incidence.

Corresponding 3 the four data signals are processed S1 in one step of the method, signal components of the special signal sound source being highlighted. As shown above, this can be done by filtering the data signals with respect to a frequency band. The data signals can be present in the form of digital audio signals or have been converted into these from analog signals.

In a further step, the processed four data signals are converted into four signal values, depending on a time range adapted to the special signal 11a , 11b , 11c , 11d calculates S2. Each of the four signal values 11a , 11b , 11c , 11d associated with the main angle of incidence that belongs to the sound transducer that generated the corresponding data signals.

In the 1 are exemplary the four signal values 11a , 11b , 11c , 11d with the associated main angles of incidence as the abscissa value in a diagram that plots an amplitude of the signal values over a horizontal angle. The main angles of incidence of the four signal values 11a , 11b , 11c , 11d can therefore be seen on the abscissa of this diagram 1 read off.

According to the method, the direction of incidence of the sound source is determined using the four signal values 11a , 11b , 11c , 11d with their main angles of incidence and at least one sound pressure angle function 12 be adapted to each other S3 . In this embodiment, the sound pressure angle function 12 to the four signal values 11a , 11b , 11c , 11d customized.

In the 1 is a sound pressure angle function 12 drawn with a periodicity of 360 degrees with a minimum and a maximum. This sound pressure angle function 12 can, as shown above, be a parameterized trigonometric function with a fixed phase or an angle-dependent phase or a function determined by measurement. This sound pressure angle function 12 becomes so to the four signal values 11a , 11b , 11c , 11d adjusted so that a norm is minimal. For example, this sound pressure angle function 12 so to the four signal values 11a , 11b , 11c , 11d be fitted that a sum of the individual distances between the signal values 11a , 11b , 11c , 11d of the sound pressure angle function 12 becomes minimal. In this exemplary embodiment, the respective main angles of incidence are assigned to the signal values.
By determining the position of the maximum of the sound pressure angle function 12 With reference to the abscissa value, the direction of incidence of the sound source is determined S3.

The 2 shows in an exemplary embodiment how the direction of incidence of the special signal sound source is determined S3 if each of four transducers has a sound pressure angle function 21a , 21b , 21c , 21d assigned. The respective sound pressure angle function 21a , 21b , 21c , 21d at the respective main angles of incidence of the corresponding sound transducers to a maximum. In this exemplary embodiment, the signal values are therefore assigned to the respective sound pressure angle function.

The four signal values 22a , 22b , 22c , 22d were calculated from the four data signals of the four sound transducers and are the respective sound pressure angle function 21a , 21b , 21c , 21d assigned accordingly. The four signals 22a , 22b , 22c , 22d become the four sound pressure angle functions 21a , 21b , 21c , 21d adjusted to each other by using exactly one of the four signal values as the parameter for the adjustment 22a , 22b , 22c , 22d associated angle is determined at which a second norm of the four signal values 22a , 22b , 22c , 22d and the associated sound pressure angle functions 21a , 21b , 21c , 21d is minimized.

This standard can, for example, be a sum of the distances between the respective signal values 22a , 22b , 22c , 22d of the assigned sound pressure angle functions 21a , 21b , 21c , 21d which is then minimized in order to determine the direction of incidence of the special signal sound source, which is then given by this angle. The respective sound pressure angle function 21a , 21b , 21c , 21d measured for example by a far-field measurement and for determining the direction of incidence S3 of the special signal sound source.

Claims (14)

Method for determining a direction of incidence of a special signal sound source in relation to a platform by means of at least two data signals that were generated by at least two respectively assigned sound transducers, the sound transducers each being assigned a main angle of incidence, with the following steps: - Processing of the at least two data signals (11a, 11b) (S1), signal components of the special signal sound source being highlighted; - Calculating a respective signal value (S2) from the processed at least two data signals, depending on a time range adapted to the special signal; - Determining the direction of incidence of the special signal sound source (S3), the at least two signal values (11a, 11b) with their main angles of incidence and at least one sound pressure angle function (12) being adapted to one another. Procedure according to Claim 1 , the sound pressure angle function (12) being parameterized and adapted to the at least two signal values (11a, 11b), the adaptation taking place in such a way that a first standard from the at least two signal values (11a, 11b) with, corresponding to the generating sound transducers, assigned main angles of incidence, and the sound pressure angle function (12) is minimized. Method according to one of the preceding claims, wherein the parameterized sound pressure angle function (12) is described with a trigonometric function or a linearized trigonometric function, with a periodicity of 360 degrees of the angle of the direction of incidence and the parameterized sound pressure curve (12) is described by an amplitude and a phase of the trigonometric functions are characterized. Procedure according to Claim 3 , the trigonometric function having a phase dependent on the angle of the direction of incidence and the parameterized sound pressure curve (12) being characterized by the amplitude and a phase dependent on the angle of the direction of incidence. Procedure according to Claim 1 , each of the at least two sound transducers being assigned a sound pressure angle function (21a, 21b) which has a maximum at the respective main angles of incidence, and the at least two signal values (22a, 22b) and the at least two sound pressure angle functions (21a, 21b) are adapted to one another in that a parameter for the adaptation is exactly one angle assigned to the signal values (22a, 22b) at which a second standard of the at least two signal values (22a, 22b) and the associated sound pressure angle functions are minimized. Procedure according to Claim 5 , the second standard being calculated from a sum of distances between the respective signal values (22a, 22b) and the assigned sound pressure angle functions (21a, 21b). Procedure according to Claim 2 to 4th or 6th , wherein the minimization of the first or the second norm takes place under the additional condition that at least one parameter for adapting at least one corresponding parameter from a previous determination of the direction of incidence may not differ by more than a predefined amount. Procedure according to Claim 2 to 4th or 6th , wherein at least one of the parameters for adaptation is calculated using a Kalman filter and at least one parameter of a previous determination of a direction of incidence. Method according to one of the Claims 5 to 8th , wherein the sound pressure angle function of each of the at least two sound transducers is determined by means of a far-field measurement. Method according to one of the preceding claims, wherein a directional characteristic of the respective sound transducers is set up by means of an individual directional characteristic of the respective sound transducers themselves and / or an arrangement of the respective sound transducers in relation to the platform. Method according to one of the preceding claims, wherein, depending on the determined direction of incidence, a control signal for controlling an at least partially automated vehicle and / or a warning signal for warning a vehicle occupant is transmitted. Device that is set up, a method according to one of the Claims 1 to 11 perform. Computer program, comprising instructions which, when the program is executed by a computer, cause the computer to execute the program, the method according to one of the Claims 1 to 11 execute. Machine-readable storage medium on which the computer program is based Claim 13 is stored.

Images