DE102020109405A1 - Method for determining direction information from target objects in a radar system for a vehicle - Google Patents

Abstract

The invention relates to a method for determining direction information (220) about directions of at least two different target objects (5) in a radar system (2) for a vehicle (1), the following steps being carried out: - Providing (101) detection information ( 210), which is specific for radar signals (230) reflected on the target objects (5) and received by at least one receiving antenna (20) of the radar system (2), - Carrying out an angle evaluation (102) on the detection information (210) in order to determine the directional information (220) as information about angles (240) of the radar signals (230), the angles (240) being specific for the directions of the target objects (5), performing a DML method (103) on the detection information (210 ) in order to use a result (225) of the DML method (103) to identify the directions of the target objects (5) in the directional information (220).

Description

The present invention relates to a method for determining directional information from target objects in a radar system for a vehicle. The invention also relates to the radar system.

It is known from the prior art that, in addition to determining a distance to a target object, radar systems can also determine further parameters of the target object. So it is z. B. possible that when using several transmitting and / or receiving antennas a directional information of the target object is determined. The different geometric arrangement of the transmitting and / or receiving antennas has an influence on the transit times of the radar signals transmitted by the transmitting antenna (s), which are reflected on the target object and received by the at least one receiving antenna. On the basis of a difference in the transit times of the various radar signals, information about the angle of incidence or angle of reflection at the antennas can therefore be obtained. The angle in turn depends on the direction of the target object.

From the DE10 2014 201 026 A1 a method for angle estimation is known which can thus also be used to determine directional information.

The problem is that when there are several target objects, it is necessary to differentiate between the angles for the radar signals reflected on the various target objects. Such a target separation therefore makes it possible to clearly determine the direction of the individual target objects. In the following, this is also referred to as an identification of the directions of the target objects in the directional information. Depending on the geometry of the antenna arrangement, ambiguities can also arise in the target separation. Target separation is technically problematic, especially when using a NULA (non-uniform linear array) antenna geometry.

It is therefore an object of the present invention to at least partially remedy the disadvantages described above. In particular, it is the object of the present invention to propose an improved solution for determining directional information from target objects in a radar system.

The above object is achieved by a method with the features of claim 1 and by a radar system with the features of claim 14. Further features and details of the invention emerge from the respective subclaims, the description and the drawings. Features and details that are described in connection with the method according to the invention naturally also apply in connection with the radar system according to the invention, and in each case vice versa, so that with regard to the disclosure of the individual aspects of the invention, reference is or can always be made to the individual aspects of the invention.

The object is achieved in particular by a method for determining directional information about directions of at least two different target objects in a radar system for a vehicle. The vehicle is, for example, a passenger vehicle or a truck which can have a driver assistance system. The direction of the target objects can be used to operate the driver assistance system.

• - Bereitstellen einer Erfassungsinformation, welche für an den Zielobjekten reflektierte und von wenigstens einer Empfangsantenne des Radarsystems empfangene Radarsignale spezifisch ist,
• - Durchführen einer Winkelauswertung bei der Erfassungsinformation, um die Richtungsinformation als eine Information über Winkel der Radarsignale zu erhalten, wobei die Winkel für die Richtungen der Zielobjekte (also die Richtungen zu den Zielobjekten relativ zum Radarsystem bzw. Fahrzeug) spezifisch sind.

In particular, it is provided that the following steps are carried out in the method according to the invention:

• - Provision of detection information which is specific for radar signals reflected on the target objects and received by at least one receiving antenna of the radar system,
• Carrying out an angle evaluation for the detection information in order to obtain the direction information as information about the angle of the radar signals, the angles being specific for the directions of the target objects (i.e. the directions to the target objects relative to the radar system or vehicle).

For example, the angles can be designed as angles of incidence and / or emergence of the radar signals which the radar signals have at the receiving and / or transmitting antennas of the radar system. The angles thus provide a conclusion about the directions of the target objects. The angles can be determined, for example, by examining a transit time difference between the radar signals. Specifically, radar signals that are reflected on the same target object but are sent and / or received by different antennas of the radar system can have the transit time difference depending on the arrangement of the antennas and the angle of the radar signals or the direction of the target object. In the case of the angle evaluation, the radar signals reflected at the same target objects can be examined accordingly with regard to the transit time difference in order to determine the directional information with the angles of the corresponding radar signals.

It may be possible that several potentially correct angles are determined during the angle evaluation, which only partially correspond to the actual angles of the radar signals. The through the The angles provided for directional information are therefore ambiguous. It is therefore important to filter out the incorrect angles in order to determine the directions of the target objects only on the basis of the actual angles. In other words, the directions of the target objects should be identified in the directional information, since the angles ambiguously indicate several possible directions.

It can therefore be provided in the invention that the directions of the target objects are identified in the directional information. This can specifically be understood to mean that the (possibly ambiguous) angles provided in the direction information are assigned to the received radar signals of the individual target objects. In this sense, each angle provided can potentially represent a direction of a target object. The identification of the actual directions of the target objects thus serves to find out from the angles provided the correct angles which actually reflect a direction of a target object.

In order to carry out an identification of the directions of the target objects, a further analysis, e.g. B. a statistical analysis, and a comparison of the result of the analysis with the directional information. It is also possible for the angle evaluation to be carried out to determine the directional information in order to carry out a first limitation to a reduced number of angles, so that only this reduced number of angles has to be evaluated by the further analysis.

• - Durchführen einer ML- (Maximum-Likelihood) insbesondere DML-(Deterministic ML) Methode bei der Erfassungsinformation, um anhand eines Ergebnisses der ML bzw. DML-Methode die Richtungen der Zielobjekte bei der Richtungsinformation zu identifizieren.

The method according to the invention advantageously provides for the following step to be carried out in order to identify the directions of the target objects in the directional information:

• Carrying out an ML (Maximum Likelihood), in particular DML (Deterministic ML) method, for the detection information in order to identify the directions of the target objects in the directional information on the basis of a result of the ML or DML method.

In other words, by identifying the directions, the target objects are separated from one another in the direction information and / or the angles are assigned to the target objects. Conventionally, a problem with target separation is that several targets, which are recognizable as peak values in signal processing, can lie on top of one another. The unambiguous and correct determination of the directional information can thus be made more difficult. This in turn can lead to angles in the direction information that are only ambiguously determinable. The ML or DML method helps to resolve this ambiguity. The ML or DML method relates to a maximum likelihood method and thus a parametric estimation method.

• - Heidenreich, P.: Antenna Array Processing: Autocalibration and Fast High-Resolution Methods for Automotive Radar, Dissertation Technische Universität Darmstadt (2012); Abrufbar unter https://tuprints.ulb.tudarmstadt.de/id/eprint/3090; insbesondere Seite 49 ff. und Seite 83 ff, wobei insbesondere als Richtungsinformation eine Direction of Arrival (DOA) bestimmt wird, und die Richtungsinformation als „BF Spectrum“ bezeichnet wird.

Known variants for implementing the DML method are known, for example, from the literature mentioned below:

• - Heidenreich, P .: Antenna Array Processing: Autocalibration and Fast High-Resolution Methods for Automotive Radar, Dissertation Technische Universität Darmstadt (2012); Available at https://tuprints.ulb.tudarmstadt.de/id/eprint/3090; in particular page 49 ff. and page 83 ff, with a direction of arrival (DOA) being determined in particular as direction information, and the direction information being referred to as “BF Spectrum”.

• - Winner H. (2015) Radarsensorik. In: Winner H., Hakuli S., Lotz F., Singer C. (eds) Handbuch Fahrerassistenzsysteme. ATZ/MTZ-Fachbuch. Springer Vieweg, Wiesbaden
• - Heidenreich P., Zoubir A.M. (2014) Computational Aspects of Maximum Likelihood DOA Estimation of Two Targets with Applications to Automotive Radar. In: Schmidt G., Abut H., Takeda K., Hansen J. (eds) Smart Mobile In-Vehicle Systems. Springer, New York, NY

The following prior art also refers to the aforementioned literature:

• - Winner H. (2015) Radar Sensor Technology. In: Winner H., Hakuli S., Lotz F., Singer C. (eds) Handbook of Driver Assistance Systems. ATZ / MTZ reference book. Springer Vieweg, Wiesbaden
• - Heidenreich P., Zoubir AM (2014) Computational Aspects of Maximum Likelihood DOA Estimation of Two Targets with Applications to Automotive Radar. In: Schmidt G., Abut H., Takeda K., Hansen J. (eds) Smart Mobile In-Vehicle Systems. Springer, New York, NY

The content of the preceding literature, in particular the aforementioned definitions and implementations of a DML method, are included in this patent application.

In order to obtain the detection information, the radar system can first transmit a transmission signal s (t) via at least one transmission antenna. The transmission signal includes, for. B. several sequentially output frequency ramps (hereinafter also referred to as chirps). The chirps can each be frequency-modulated and thus have a varying frequency. Here comes z. B. a linear frequency modulation is used, in which the frequency changes linearly within a predetermined bandwidth for a respective chirp. The transmitted signal s (t) reflected on at least one target object and delayed by a signal propagation time can then be detected as a received signal e (t) by at least one receiving antenna of the radar system. _{The baseband signal b (t) with the frequency f b} = f _s -f _e is obtained from the mixture of the transmitted signal s (t) and the received signal e (t). Here, f _{s is} the frequency of the transmitted signal s (t) and f _{e is} the frequency of the received signal e (t). The frequency f _b depends on the signal propagation time τ and thus on the distance R from the target object.

In order to carry out the detection of the at least one target object, the detection information can be determined on the basis of the received signal e (t), and in particular on the basis of the baseband signal b (t). For example. the detection information results from the digitized baseband signal b (t) or from at least one frequency analysis of the baseband signal b (t). Correspondingly, the detection information can be digital information, that is to say data values. If N chirps are output, then the duration of each chirp is T1 / N. After the period T1, the acquisition information can be processed within the period T2-T1. The entire measurement cycle thus has a duration T2, so that the transmission of the transmission signal s (t) can be repeated at intervals of T2. T2 thus defines a transmission signal distance. It is also possible for the time periods T1 and / or T2 to vary. Different transmission signals s1 (t) and s2 (t) can also be transmitted, which e.g. B. differ in terms of the distances between the chirps.

During the time period T1, the individual values of the received signal e (t) can be recorded so that the detection information is formed from the recorded values and, if necessary, preprocessing (such as down-mixing and / or analog-digital conversion and / or at least one Fourier transformation) will. The recorded values can be understood as a matrix in which the values are stored one after the other in an MxN matrix with M samples per chirp and N chirps in a two-dimensional manner until the end of the time period T1. The detection information can correspond to this two-dimensional configuration. A first of the dimensions can be specific for a distance to the target object, the other (second) of the dimensions for the Doppler frequency and thus for the relative speed of the target object. Using this matrix, at least one spectrum can then be determined by at least one Fourier transformation of the matrix, from which the relative speed and / or the distance of the at least one target object in the vicinity of the vehicle can be determined. Specifically, a spectrum can be determined from a (e.g. column-wise) first Fourier transformation of the matrix in the direction of the first dimension (column-wise), which is again composed as a two-dimensional matrix and from which the distance can be determined. The relative speed can then also be determined from a (line-by-line) second Fourier transformation in the direction of the second dimension of the spectrum. This second spectrum resulting from the second Fourier transformation can also correspond to the detection information. If several transmitting and / or receiving antennas are used, a third dimension can also be used for the receiving signals e (t) of the various receiving antennas. In this case there are several second spectra for the different antennas. A third Fourier transformation in the direction of this third dimension can be used to determine an angle and thus the direction of the target object. The result of this third Fourier transformation can correspond to the direction information. The direction information is, for example, a so-called beamforming (BF) spectrum. The third Fourier transformation is carried out, for example, at a point, that is to say in a bin, of the second spectra at which the target object is detected. For this purpose z. B. a peak value detection (peak detection) in the second spectra. Even if only one peak value is detected here, it is possible that several target objects are superimposed for this bin. This can be made clear in the direction information through several peak values. However, the separation ability can be problematic at such narrow frequencies. Using a DML method can still make it possible to reliably separate the target objects. However, the DML method can be computationally complex, so that appropriate solutions for reducing the effort are described below.

Furthermore, it can be provided within the scope of the invention that the result of the DML method is designed as a two-dimensional spectrum which indicates an angle combination of the angles for the directions of the at least two target objects.

The DML method can be represented in the following form, for example: $${\theta }_{\mathrm{D.}\mathrm{M.}\mathrm{L.}}=\underset{\theta }{\text{argmax}}\left(Tr\left\{{\mathrm{P.}}_{\mathrm{A.}}\left(\theta \right)\widehat{\mathrm{R.}}\right\}\right).$$

Here, θ = ψ ₁ , ψ _{2 can represent} the electrical angles of the radar signal - and thus the possible angle combination of the angles in the spectrum. One of the combination of angles can, for. B. by determining a maximum in the spectrum as the angle combination of the angles for the directions of the at least two target objects.

die Kovarianzmatrix sein. Die Werte der Erfassungsinformation können dabei durch x(i) repräsentiert sein. Bei zwei Zielobjekten kann x(i) durch x(i) = s₁(i)α(ψ₁) + s₂(i)α(ψ₂) + e(i), i = 1, ..., N beschrieben werden. x(i) soll nachfolgend auch allgemein mit dem Signalvektor ν für die Messung bezeichnet werden. Damit ergibt sich für P_A(θ)R̂ = A(A^HA)^-1A^Hνν^H.Furthermore, according to the aforementioned formula for the DML P _A (ψ ₁ , ψ ₂ ) = A (A ^H A) ^-1 A ^H , A = [α (ψ ₁ ), α (ψ ₂ )] the projection matrix and $\widehat{\mathrm{R.}}=\frac{1}{N}{\displaystyle {\sum }_{i=1}^{N}x\left(i\right)x{\left(i\right)}^H}$

be the covariance matrix. The values of the acquisition information can be represented by x (i). With two target objects, x (i) can be given by x (i) = s ₁ (i) α (ψ ₁ ) + s ₂ (i) α (ψ ₂ ) + e (i), i = 1, ..., N to be discribed. In the following, x (i) will also be referred to in general with the signal vector ν for the measurement. This results in P _A (θ) R̂ = A (A ^H A) ^-1 A ^H νν ^H.

It can optionally be possible for a result of the angle evaluation to be reused when the DML method is carried out in order to replace at least one arithmetic operation for determining the result of the DML method. This reduces the work involved in calculating the DML. In an exemplary implementation, A and H can be precalculated and thus predefined. A ^H is a matrix which can also be viewed as a task of a Fourier transform. ν corresponds to the detection information which is Fourier transformed (for example by means of a Fast Fourier transformation) during the angle evaluation in order to obtain _{the direction information FFT ν as the result of the angle evaluation.} The result FFT _ν is therefore already known when the DML method is carried out. Therefore, A ^H ν can also be assumed to be _{the direction information FFT ν.} When calculating θ _DML , A ^H ν is thus exchanged for the direction information FFT _{ν in} order, as described above, to replace the arithmetic operation for determining the result of the DML method.

Another advantage can be achieved within the scope of the invention if the directions of the target objects in the directional information are identified by assigning the angles from the directional information to the target objects on the basis of the result of the DML method. By the DML method z. B. several angle combinations can be evaluated, which include the angles for two different target objects. These target objects can be superimposed in the directional information and thus lead to an ambiguous result. In other words, the angles for the radar signals that were reflected from the target objects cannot be clearly derived from the directional information. The direction information is, for example, a spectrum, in particular a BF spectrum, in which the amplitudes are specified for different angles. The directional information may only include information about those target objects that are detected at a certain distance and / or with a certain relative speed (e.g. by selecting the bin for the third Fourier transformation). Nevertheless, at least or exactly two target objects can be superimposed in the directional information. This can lead to several peaks in the directional information, only some of which indicate the correct angles for the radar signals. In order to determine the correct angle combination of these angles, the DML method can thus be carried out.

Optionally, it is conceivable that the directions of the target objects are identified in the directional information in that a peak value detection is carried out in the directional information and / or in the result of the DML method. The detection of the peak values in the direction information makes it possible to determine candidates for the correct angles which are specific for the correct direction of the target objects. The peak value recognition in the result of the DML method can serve to identify possible angle combinations of the angles in the direction information at the peak values which are specific for the correct direction.

• - Unterscheiden zwischen einem einzigen Zielobjekt in der Richtungsinformation und wenigstens oder genau zwei Zielobjekten durch einen Vergleich des Ergebnisses der DML-Methode mit der Richtungsinformation.

Another advantage within the scope of the invention can be achieved if the identification of the directions of the different target objects comprises the following step:

• Distinguishing between a single target object in the directional information and at least or precisely two target objects by comparing the result of the DML method with the directional information.

For example. For this purpose, a peak value for an angle in the direction information and a peak value for the same angle in the result of the DML method can be compared. If the peak value in the direction information exceeds the peak value of the result of the DML method, then there is only a single target object, otherwise (at least or precisely) two target objects. Alternatively or additionally, it can be examined whether the maximum in the result of the DML method is smaller than in the direction information. In this case there is only one target object.

It can optionally be possible that when the DML method is carried out, different combinations in each case two different angles are evaluated in order to determine one of the combinations for the directions of the two target objects on the basis of the result of this evaluation. For example, a local maximum can be determined for each of these possible combinations, and these maxima can be compared with one another. On the basis of the comparison, the angle combination can be determined (for example at the greatest) which corresponds to the actual angle of the radar signals that were reflected by the two target objects. The direction of these target objects is thus also available.

A further advantage can be that the direction information provides ambiguously different possible directions of the target objects on the basis of the angles, and the directions for the target objects are clearly determined on the basis of the result of the DML method. For example, the direction information corresponds to a two-dimensional beamforming spectrum which specifies the different angles for different amplitudes. The result of the DML method can be structured in exactly the same way, but three-dimensionally in a first dimension the different angles and in a second dimension the different ones Specify angles, and thus angle combinations, for different amplitudes.

According to a further advantage, it can be provided that the result of the DML method provides different possible combinations of the angles for different directions of the target objects. The result of the DML method is, for example, a two-dimensional matrix which includes different angles for a first target object in a first dimension and different angles for the second target object in the second dimension, so that the peak values in this matrix represent the angle combinations.

It can be advantageous if, within the scope of the invention, at least one peak value is recognized in the result of the DML method and / or at least one maximum of the result of the DML method is determined in order to clearly determine the angles for the directions of the target objects. For example, the angle combination which is used to determine the directions of the target objects can be identified at the point of the maximum or peak value. Although the DML method could in principle also be carried out directly (without angle evaluation and determination of the directional information), the computational effort in the DML method can be reduced through the analysis according to the angle evaluation. For example, on the basis of recognized peak values in the direction information, start values for the DML method can be set in order to evaluate angle combinations according to the DML method only there. The directional information is used to limit the angles in question to a reduced number of angles, which are evaluated using the DML method.

• - Durchführen einer Erkennung von wenigstens oder genau zwei Spitzenwerten bei der Richtungsinformation, um bei den (erkannten) Spitzenwerten jeweilige Winkel zu ermitteln,
• - Setzen von Kombinationen der ermittelten Winkel als Startpunkte für die DML-Methode,
• - Durchführen der DML-Methode für die Startpunkte und deren Umgebung,
• - Finden eines lokalen Maximums in jeder der Umgebungen,
• - Bestimmen eines Maximums der lokalen Maxima, um die Winkel für die Richtungen der Zielobjekte eindeutig zu bestimmen.

Further, performing the DML method may include the following steps:

• - Carrying out a recognition of at least or exactly two peak values in the direction information in order to determine respective angles for the (recognized) peak values,
• - Setting of combinations of the determined angles as starting points for the DML method,
• - Performing the DML method for the starting points and their surroundings,
• - Finding a local maximum in each of the surroundings,
• Determination of a maximum of the local maxima in order to uniquely determine the angles for the directions of the target objects.

Furthermore, it is optionally possible within the scope of the invention for the angle evaluation to be carried out in the manner of a frequency analysis of a part of the detection information, this part being identified by means of a detection of at least one peak value in the detection information prior to the angle evaluation. At least one target object can thus be detected, with two target objects possibly also being superimposed in this part, in particular Bin. The detection information can be understood as a three-dimensional matrix, for example. The peak value can then be detected by means of peak value recognition in the two-dimensional spectrum of the first and second dimensions (in particular distance and relative speed), and the angle evaluation can be carried out there in the direction of the third dimension. The values of the detection information in the direction of this third dimension are also referred to in the context of the invention as a bin or as a signal vector ν, from which R̂ can be determined as the covariance matrix of ν. The angle evaluation can, for. B. with a ULA (uniform linear array) by means of zero padding and a subsequent Fast Fourier Transformation (FFT) and with a non-ULA by inserting zeros and then FFT.

Furthermore, it can be provided within the scope of the invention that the angle evaluation includes at least one Fourier transformation for frequency analysis. This can in particular be carried out in the manner of an FFT.

The invention also relates to a radar system for a vehicle, having an (electronic) processing device such as a digital signal processor, which is designed to carry out the steps of a method according to the invention. For this purpose, the processing device can, for. B. execute a computer program which is not stored in a volatile memory in the radar system. The radar system according to the invention thus has the same advantages as are described in detail with reference to a method according to the invention. It is optionally conceivable that the radar system is designed as a continuous wave radar, in particular frequency-modulated continuous wave radar (FMCW radar).

• 1 eine schematische Darstellung eines Fahrzeuges mit einem erfindungsgemäßen Radarsystem in einer Seitenansicht des Fahrzeuges,
• 2 eine schematische Darstellung von Teilen eines erfindungsgemäßen Radarsystems mit einer Sendeantenne,
• 3 eine schematische Darstellung von Teilen eines erfindungsgemäßen Radarsystems mit einer Empfangsantenne,
• 4 eine schematische Darstellung von Teilen eines erfindungsgemäßen Radarsystems mit mehreren Sende- und Empfangsantennen,
• 5 eine schematische Darstellung von Teilen eines erfindungsgemäßen Radarsystems mit mehreren Empfangsantennen,
• 6 eine schematische Darstellung zur Visualisierung eines erfindungsgemäßen Verfahrens,
• 7 eine weitere schematische Darstellung zur Visualisierung eines erfindungsgemäßen Verfahrens,
• 8 eine weitere schematische Darstellung zur Visualisierung eines erfindungsgemäßen Verfahrens,
• 9 eine weitere schematische Darstellung zur Visualisierung eines erfindungsgemäßen Verfahrens,
• 10 eine weitere schematische Darstellung zur Visualisierung eines erfindungsgemäßen Verfahrens.

Further advantages, features and details of the invention emerge from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. The features mentioned in the claims and in the description can each be essential to the invention individually or in any combination. Show it:

• 1 a schematic representation of a vehicle with a radar system according to the invention in a side view of the vehicle,
• 2 a schematic representation of parts of a radar system according to the invention with a transmitting antenna,
• 3 a schematic representation of parts of a radar system according to the invention with a receiving antenna,
• 4th a schematic representation of parts of a radar system according to the invention with several transmitting and receiving antennas,
• 5 a schematic representation of parts of a radar system according to the invention with several receiving antennas,
• 6th a schematic representation for the visualization of a method according to the invention,
• 7th a further schematic representation to visualize a method according to the invention,
• 8th a further schematic representation to visualize a method according to the invention,
• 9 a further schematic representation to visualize a method according to the invention,
• 10 a further schematic representation for the visualization of a method according to the invention.

In the following figures, the same reference numerals are used for the same technical features from different exemplary embodiments.

In 1 is schematically a vehicle 1 with a radar system according to the invention 2 shown. The radar system 2 can at least one receiving antenna 20th have, and a processing device 3 to use at least one through the receiving antenna 20th received radar signal 230 to carry out a method according to the invention.

Specifically, as in the 2 until 5 is shown several transmitting antennas 21 and receiving antennas 20th be provided. You can also have multiple target objects 5 can be detected in different directions. To detect the directions, z. B. a MIMO (Multiple Input Multiple Output) or SIMO (Single Input Multiple Output) transmission scheme can be used.

In 2 is just a transmitting antenna 21 provided, but at least two receiving antennas 20th . This enables an angle 240 in the form of an angle of incidence on a target object 5 reflected radar signals 230 based on a transit time difference to determine which between the radar signals 230 present from different receiving antennas 20th be received. This procedure can be used for the radar signals 230 be repeated which of other target objects 5 to be reflected to also for this the angle 240 to investigate.

In 3 is just a receiving antenna 20th provided, but at least two transmitting antennas 21 . This enables an angle 240 in the form of an angle of reflection that of a target object 5 reflected radar signals 230 based on a transit time difference to determine which between the radar signals 230 present from different transmitting antennas 21 be sent out. This procedure can be used for the radar signals 230 be repeated which of other target objects 5 to be reflected to also for this the angle 240 to investigate.

According to 4th a combination of the two aforementioned variants can be provided.

In 5 it is made clear that for several target objects 5 different angles 240 ' and 240 " can be determined which respectively for the direction of the target object 5 relative to the vehicle 1 are specific.

In 6th A method according to the invention for determining directional information is shown 220 about the directions of at least two different target objects 5 in a radar system 2 for a vehicle 1 visualized schematically. According to a first method step, a provision can be made 101 a detection information 210 take place which for at the target objects 5 reflected and from at least one receiving antenna 20th of the radar system 2 received radar signals 230 is specific. This acquisition information 210 is z. B. a two-dimensional spectrum from which the distance and the relative speed of the target objects 5 can be determined. An angle evaluation can then be carried out according to a second method step 102 in the acquisition information 210 done to the direction information 220 as information about angles 240 the radar signals 230 to get taking the angles 240 for the directions of the target objects 5 are specific. Carrying out a DML method can also be used as a third method step 103 in the acquisition information 210 be provided to be based on a result 225 the DML method 103 the directions of the target objects 5 in the direction information 220 to identify. It is possible that then the directions of the target objects 5 in the direction information 220 be identified by having a peak value detection 104 in the direction information 220 and / or the result 225 the DML method 103 is carried out. Leave that way z. B. the in 7th shown peak values 250 in the direction information 220 identify. Every peak 250 can through a local maxima of the amplitude A. and a possible angle 240 for a target object 5 and thus the direction for the target object 5 indicate. If at the direction information 220 e.g. with only two superimposed target objects 5 more than two peaks 250 are recognized, then the identification of the directions based on the angles 240 not clearly possible. So is in 7th recognizable that several peaks 250 and thus several possible angles 240 occur, but only the first and second angles 240 ' , 240 " the actual angles of the two targets 5 represent.

• - Durchführen einer Erkennung 104 von wenigstens zwei Spitzenwerten 250 bei der Richtungsinformation 220, um bei den Spitzenwerten 250 jeweilige Winkel 240 zu ermitteln,
• - Setzen von Kombinationen der ermittelten Winkel 240 als Startpunkte 260 für die DML-Methode 103,
• - Durchführen der DML-Methode 103 für die Startpunkte 260 und deren Umgebung 261,
• - Finden eines lokalen Maximums in jeder der Umgebungen 261,
• - Bestimmen eines Maximums der lokalen Maxima, um die Winkel 240 für die Richtungen der Zielobjekte 5 eindeutig zu bestimmen.

Therefore, the DML method 103 can be used to determine each of the angle combinations (according to the detected peak values 250 ) to check. A result 225 the DML method 103 is exemplary in 8th shown. Performing the DML Method 103 can include the following steps:

• - Perform a discovery 104 of at least two peaks 250 in the direction information 220 to get at the peaks 250 respective angles 240 to investigate,
• - Setting combinations of the determined angles 240 as starting points 260 for the DML method 103 ,
• - Performing the DML method 103 for the starting points 260 and their surroundings 261 ,
• - Finding a local maximum in each of the neighborhoods 261 ,
• - Determining a maximum of the local maxima around the angles 240 for the directions of the target objects 5 to be clearly determined.

The angles 240 for the actual directions of the target objects 5 can be determined according to the angle combination at which the maximum is present. The first and second angles are present 240 ' , 240 "

• - Unterscheiden zwischen einem einzigen Zielobjekt 5 in der Richtungsinformation 220 und wenigstens oder genau zwei Zielobjekten 5 durch einen Vergleich des Ergebnisses 225 der DML-Methode 103 mit der Richtungsinformation 220.

In 9 an example is shown in which only one target object 5 is present (and no overlay of several target objects 5 in the direction information 220 ). Therefore, according to 10 identifying the directions of the different target objects 5 include the following step:

• - Distinguish between a single target object 5 in the direction information 220 and at least or exactly two target objects 5 by comparing the result 225 the DML method 103 with the direction information 220 .

The comparison can e.g. B. be done by the fact that the maximum of the result 225 with the maximum of the direction information 220 is compared.

The above explanation of the embodiments describes the present invention exclusively in the context of examples. Of course, individual features of the embodiments can be freely combined with one another, provided that they are technically sensible, without departing from the scope of the present invention.

List of reference symbols

1

vehicle

2

Radar system

3

Processing device

5

Target object

20th

Receiving antenna

21

Transmitting antenna

101

Provide

102

Angle evaluation

103

DML method

104

Detection, peak value detection

210

Acquisition information

220

Directional information, beamforming spectrum

225

two-dimensional spectrum, result of the DML method

230

Radar signal

240

angle

250

Peak value, peak

260

Starting point

261

Surroundings

240 '

first angle

240 "

second angle

A.

amplitude

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102014201026 A1 [0003]

Claims (15)

Method for determining direction information (220) about directions of at least two different target objects (5) in a radar system (2) for a vehicle (1), the following steps being carried out: - Provision (101) of detection information (210) which is specific for radar signals (230) which are reflected on the target objects (5) and received by at least one receiving antenna (20) of the radar system (2), - Carrying out an angle evaluation (102) on the detection information (210) in order to obtain the direction information (220) as information about angles (240) of the radar signals (230), the angles (240) for the directions of the target objects (5) are specific - Carrying out a DML method (103) on the detection information (210) in order to identify the directions of the target objects (5) in the direction information (220) on the basis of a result (225) of the DML method (103). Procedure according to Claim 1 , characterized in that the directions of the target objects (5) in the directional information (220) are identified by using the result (225) of the DML method (103) to assign the angles (240) from the directional information (220) to the target objects ( 5) can be assigned. Method according to one of the preceding claims, characterized in that the directions of the target objects (5) in the direction information (220) are identified by a peak value detection (104) in the direction information (220) and / or in the result (225) of the DML method (103) is carried out. Method according to one of the preceding claims, characterized in that a result of the angle evaluation (102) is reused when the DML method (103) is carried out in order to replace at least one arithmetic operation for determining the result of the DML method (103). Method according to one of the preceding claims, characterized in that the result (225) of the DML method (103) is designed as a two-dimensional spectrum (225) which is an angle combination of the angles (240) for the directions of the at least two target objects (5 ) indexed. Method according to one of the preceding claims, characterized in that the identification of the directions of the different target objects (5) comprises the following step: - Distinguishing between a single target object (5) in the direction information (220) and at least or exactly two target objects (5) by comparing the result (225) of the DML method (103) with the direction information (220). Method according to one of the preceding claims, characterized in that when the DML method (103) is carried out, different combinations of two different angles (240) are evaluated in order to use the result of this evaluation to determine one of the combinations for the directions of the two target objects (5 ) to investigate. Method according to one of the preceding claims, characterized in that the direction information (220) provides ambiguously different possible directions of the target objects (5) on the basis of the angles (240), and on the basis of the result (225) of the DML method (103) the directions for the target objects (5) are clearly determined. Method according to one of the preceding claims, characterized in that the result of the DML method (103) provides different possible combinations of the angles (240) for different directions of the target objects (5). Procedure according to Claim 9 , characterized in that at least one peak value in the result (225) of the DML method (103) is recognized and / or at least one maximum of the result (225) of the DML method (103) is determined to the angle (240) for to clearly determine the directions of the target objects (5). Method according to one of the preceding claims, characterized in that the implementation of the DML method (103) comprises the following steps: - implementation of a detection (104) of at least two peak values (250) in the direction information (220) to in the peak values (250) to determine respective angles (240), - setting combinations of the determined angles (240) as starting points for the DML method (103), - performing the DML method (103) for the starting points (260) and their surroundings (261), - Finding a local maximum in each of the surroundings (261), - Determining a maximum of the local maxima in order to uniquely determine the angles (240) for the directions of the target objects (5). Method according to one of the preceding claims, characterized in that the angle evaluation (102) is carried out in the manner of a frequency analysis of part of the detection information (210), with prior to the angle evaluation (102) this part is identified by means of a recognition (104) of at least one peak value (250) in the detection information (210). Method according to one of the preceding claims, characterized in that the angle evaluation (102) comprises at least one Fourier transformation for frequency analysis. Radar system (2) for a vehicle (1), having a processing device (3) which is designed to carry out the steps of a method according to one of the preceding claims. Radar system (2) Claim 14 , characterized in that the radar system (2) is designed as a continuous wave radar.

Images