DE102014223990A1 - Method for detecting at least two targets with a radar sensor and radar sensor - Google Patents

Abstract

The present invention discloses a method for detecting at least two targets with a radar sensor, comprising emitting at least one frequency-modulated transmit signal having at least two signal sequences each having successive ramps with gaps therebetween, the at least two signal sequences having a predetermined temporal offset are interleaved such that a first ramp of each of the at least two signal sequences is output before a second ramp of one of the at least two signal sequences is output, mixing the transmitted frequency modulated transmit signal with received signals to baseband signals, and performing a multi-target velocity estimate based on the baseband signals a multi-objective optimization. Further, the present invention discloses a corresponding radar sensor.

Description

The present invention relates to a method for detecting at least two targets with a radar sensor and a corresponding radar sensor.

State of the art

Also, when the present invention is described below in the context of vehicle radar systems, it is not limited thereto and may be used with any radar systems.

Modern vehicles use a variety of electronic systems, e.g. Serve to assist a driver in driving the vehicle. Brake assistants, for example, can detect drivers driving ahead and brake and accelerate the vehicle accordingly, so that a predetermined minimum distance to the preceding road users is always maintained. Such brake assistants may also initiate emergency braking if they recognize that the distance to the preceding vehicle is too low.

In order to be able to provide such assistance systems in a vehicle, it is necessary to acquire data about the surroundings of the respective vehicle. In the above example of a brake assist, it is e.g. necessary to detect the position or relative speed of a preceding road user or its relative speed relative to the radar sensor or the vehicle in which the radar sensor is installed to capture.

To capture the vehicle environment different sensor types can be used. For example, radar sensors, ultrasonic sensors, cameras or the like can be used.

As radar sensors FMCW radar sensors are usually used in vehicles for detecting the vehicle environment. Such a radar sensor shows, for example, the DE 10 2013 200 951 A1 ,

If an undersampling of the Doppler shift frequency takes place during the evaluation of the FMCW signals, this leads to ambiguities in the calculation of the relative speed. By using several spectra, such ambiguities for a single case can be resolved via the occurring phase differences.

Disclosure of the invention

The present invention discloses a method having the features of claim 1 and a radar sensor having the features of claim 12.

Accordingly, it is provided:

A method for detecting at least two targets with a radar sensor, comprising emitting at least one frequency-modulated transmit signal having at least two signal sequences each having successive ramps with gaps in between in the frequency response, wherein the at least two signal sequences are interleaved with a predetermined time offset in such a way in that each a first ramp of each of the at least two signal sequences before outputting a second ramp of one of the at least two signal sequences, mixing the transmitted frequency modulated transmit signal with received signals to baseband signals, and performing a multi-target velocity estimate based on the baseband signals based on multi-target optimization.

It is also provided:

A radar sensor for detecting at least two targets with at least one transmitting / receiving antenna, with a signal generator, which is designed to output at least one frequency-modulated transmission signal via the at least one transmitting / receiving antenna, which has at least two signal sequences, each successive ramps in the frequency curve having gaps in between, wherein the at least two signal sequences are interleaved with each other at a predetermined time offset such that a first ramp of each of the at least two signal sequences is output before a second ramp of one of the at least two signal sequences is output, and with an evaluation device, which is designed to carry out a multi-target speed estimation according to one of the preceding claims.

Advantages of the invention

The finding underlying the present invention is that if the Doppler shift frequency is underscanned, two targets at different speeds and different distances may occupy the same bin. This results in unequally strong signal components of the individual targets to a masking of a target by the other target. With approximately equally strong signal components of the individual targets, this makes the speed resolution more difficult. Furthermore, an angle estimation for the individual objects is made more difficult.

The idea underlying the present invention is now to take this knowledge into account and provide a way to resolve a superposition of two targets in a bin.

For this purpose, the present invention provides that a frequency-modulated transmission signal is transmitted. In this case, the frequency-modulated transmission signal has at least two signal sequences which each have successive ramps with gaps in between in the frequency response. The at least two signal sequences are interleaved with each other at a predetermined time offset such that in each case a first ramp of each of the at least two signal sequences is output before a second ramp of one of the at least two signal sequences is output.

For evaluating the radar signals, the transmitted frequency-modulated transmission signals for each of the signal sequences are mixed with a received signal to obtain corresponding baseband signals.

The determination of the speeds for more than one target is then based on a multi-objective optimization.

Advantageous embodiments and further developments emerge from the dependent claims and from the description with reference to the figures.

In one embodiment, prior to performing the multi-target speed estimation, a single-speed estimate is performed, and the multi-speed estimate is performed only if the single-speed estimate does not produce a unique result. The single-speed estimation can be performed with less computational effort than the multi-target speed estimation. For this reason, in the usual case where there is no overlapping of two targets in a bin, the computational effort can be minimized when first performing a single-speed estimation.

In one embodiment, in the multi-target speed estimation, the maximum of a function indicative of a relative speed spectrum is calculated for possible speed values of the at least two destinations. This allows the superposition of two targets in a bin to be resolved by a simple search for a maximum.

wobei v₁ und v₂ für in die Funktion eingesetzte Geschwindigkeiten des ersten Ziels und des zweiten Ziels stehen, arg max für die argumentum maximi Funktion steht, P_A(v₁, v₂) für eine Projektionsmatrix steht, die sich aus einer Steuerungsmatrix A ergibt. Dies ermöglicht es, eine Art zweidimensionale Suche nach dem Maximum über alle möglichen Relativgeschwindigkeiten durchzuführen.In one embodiment, the function is defined for two destinations

where v ₁ and v _{2 represent} speeds of the first target and the second target used in the function, arg max stands for the argumentum maximi function, P _A (v ₁ , v ₂ ) stands for a projection matrix consisting of a control matrix A results. This makes it possible to perform a kind of two-dimensional search for the maximum over all possible relative velocities.

In one embodiment, the control matrix A is defined as A = [a (v ₁ ), a (v ₂ )]

definiert, wobei I die Anzahl der Signalsequenzen kennzeichnet, c die Lichtgeschwindigkeit ist und T_1i für die jeweilige Signalsequenz den vorgegebenen zeitlichen Abstand zur ersten Signalsequenz darstellt. Durch die Steuerungsvektoren werden die Rampen der einzelnen Signalsequenzen auf die erste der Signalsequenzen abgebildet, bzw. phasenverschoben. Dabei entspricht der Steuerungsvektor einer idealen Messung für ein Ziel mit einer gegebenen Geschwindigkeit v. Further, a (v ₁ ) and a (v ₂ ) are each as control vectors of the type

where I denotes the number of signal sequences, c is the speed of light and T _{1i represents} the predetermined time interval for the respective signal sequence for the first signal sequence. The control vectors mimic or phase-shift the ramps of the individual signal sequences to the first of the signal sequences. The control vector corresponds to an ideal measurement for a target with a given velocity v.

In one embodiment, the projection matrix P _A (v ₁ , v ₂ ) is defined as P _A (v ₁ , v ₂ ) = A (AA ^H ) ^-1 A ^H = AA ⁺ where A ⁺ stands for the Moore-Penrose pseudoinverse of A. This allows the pre-calculation of the Moore-Penrose pseudoinverse for later processing.

In one embodiment, the method comprises estimating angles, e.g. from an azimuth angle or elevation angle, between the radar sensor and the at least two targets, especially after the multi-target velocity estimation. This allows the exact location of the detected targets to be determined.

In one embodiment, the estimation of angles is based on the Moore-Penrose pseudoinverse A ⁺ , with measurement signals contained in a measurement vector a _m (n) producing individual measurements for each of the targets based on the Moore Penrose pseudoinverse A ⁺ which are fed to an angle estimator. This allows the separation of the individual measured values from the received signal and thus a separate processing in the angle estimation.

In one embodiment, the individual measurements for two targets are calculated according to [y _{m, 1} (n) y _{m, 2} (n)] = A ⁺ a _m (n).

In one embodiment, the individual measured values are weighted on the basis of their amplitude difference and, in the case of an amplitude difference which lies above a predetermined first limit value, only the target, which corresponds to the greater amplitude, is recognized as the target. This allows targets, e.g. due to disturbances were detected, to eliminate and to continue working only with real goals.

In one embodiment, the amplitudes of the individual measured values are each set in relation to a total signal power of the received signal or the signal-to-noise ratio, wherein the target corresponding to the respective individual measured value is only detected if the corresponding individual measured value is above a second limit value. This also enables targets, e.g. due to disturbances were detected, to eliminate and to continue working only with real goals.

In one embodiment, the amplitudes of the individual measured values are checked for plausibility and no target is recognized if the individual measured values are classified as implausible. This can prevent misrecognition of targets.

The above embodiments and developments can, if appropriate, combine with each other as desired. Further possible refinements, developments and implementations of the invention also include combinations of features of the invention which have not been explicitly mentioned above or described below with regard to the exemplary embodiments. In particular, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the present invention.

Brief description of the drawings

The present invention will be explained in more detail with reference to the exemplary embodiments indicated in the schematic figures of the drawings. It shows:

1 a flowchart of an embodiment of the method according to the invention; and

2 a block diagram of an embodiment of the radar sensor according to the invention.

In all figures, the same or functionally identical elements and devices - unless otherwise stated - have been given the same reference numerals.

Embodiments of the invention

1 shows a flowchart of an embodiment of the method according to the invention.

The method sees the transmission S1 of at least one frequency-modulated transmission signal 3 in front. The transmission signal 3 has at least two signal sequences 4-1 . 4-2 on, which in the frequency course each successive ramps 5-1 - 5-6 with gaps in between 6-1 - 6-6 exhibit. In this case, the at least two signal sequences 4-1 . 4-2 with a given time offset 7 so nested in each other, that in each case a first ramp 5-1 - 5-6 each of the at least two signal sequences 4-1 . 4-2 is spent before a second ramp 5-1 - 5-6 one of the at least two signal sequences 4-1 . 4-2 is issued. The temporal "nesting" of the two signal sequences 4-1 . 4-2 is related to 2 explained in more detail.

The method also looks for each of the signal sequences 4-1 . 4-2 the mixing S2 of the transmitted frequency-modulated transmission signal 3 with a received signal 16 to one baseband signal each 8-1 . 8-2 in front.

Finally, a multi-target velocity estimate is determined based on the baseband signals 8-1 . 8-2 based on multi-objective optimization, S3. In one embodiment, the multi-target speed estimation is performed only when a single-speed estimate previously performed 9 no clear result.

Multi-objective optimization is based on the search for a maximum 10 a given function 11 representing a relative velocity spectrum for the baseband signals 8-1 . 8-2 represents.

For two goals 2-1 . 2-2 is the function 11 is defined as

stellen die gesuchten Relativgeschwindigkeiten der Ziele 2-1, 2-2 dar. Der Index m steht für Messwerte bzw. aus den Basisbandsignalen gewonnene Messwertvektoren. Die Komponenten eines Messvektors a_m(n) geben Spektralwerte von für die Basisbandsignale 8-1, 8-2 der einzelnen Signalsequenzen 4-1, 4-2 berechneten Spektren an einer Position eines Maximums in dem jeweiligen berechneten Spektrum an. Die Spektren können insbesondere als zweidimensionale Spektren berechnet werden, bei welchen in der ersten Dimension jeweils Rampe 5-1–5-6 für Rampe 5-1–5-6 einer Signalsequenz 4-1, 4-2 Fourier-transformiert wird und in der zweiten Dimension über den Rampenindex transformiert wird.Where v1 and v2 are the parameters of the function 11 which indicates possible speeds of the first target 2-1 and the second goal 2-2 stand. P _A (v ₁ , v ₂ ) stands for a projection matrix resulting from a control matrix A = [a (v ₁ ), a (v ₂ )]. "Arg max" also denotes the well-known "argumentum maximi" function.

set the sought relative speeds of the goals 2-1 . 2-2 The index m stands for measured values or measured value vectors obtained from the baseband signals. The components of a measurement vector a _m (n) give spectral values of for the baseband signals 8-1 . 8-2 the individual signal sequences 4-1 . 4-2 calculated spectra at a position of a maximum in the respective calculated spectrum. The spectra can be calculated, in particular, as two-dimensional spectra, in which ramps in each case in the first dimension 5-1 - 5-6 for ramp 5-1 - 5-6 a signal sequence 4-1 . 4-2 Fourier transformed and transformed in the second dimension via the ramp index.

The two control vectors a (v ₁ ), a (v ₂ ) of the control matrix are respectively based on that for the individual signal sequences 4-1 . 4-2 predetermined time offset 7 educated. In particular, the respective, for the respective signal sequence 4-2 corresponding predetermined time offset 7 , expected phase difference to the first signal sequence 4-1 recorded in the control vectors a (v ₁ ), a (v ₂ ).

an, wobei I die Anzahl der Signalsequenzen 4-1, 4-2 kennzeichnet, c die Lichtgeschwindigkeit ist und T_1i für die jeweilige Signalsequenz 4-2 den vorgegebenen zeitlichen Abstand zur ersten Signalsequenz 4-1 darstellt.Thus, the control vectors a (v ₁ ), a (v ₂ ) take the form

where I is the number of signal sequences 4-1 . 4-2 indicates, c is the speed of light and T _1i for the respective signal sequence 4-2 the predetermined time interval to the first signal sequence 4-1 represents.

From the control vectors a (v ₁ ), a (v ₂ ) thus obtained, the projection matrix P _A (v ₁ , v ₂ ) is calculated as P _A (v ₁ , v ₂ ) = A (AA ^H ) ^-1 A ^H. Here, (AA ^H ) ^-1 A ^{H can} also be regarded as the Moore-Penrose pseudoinverse A ⁺ of A. P _A represents, so to speak, a "velocity beamformer", through which the measurement times between the individual signal sequences 4-1 . 4-2 or their phase offset can be compensated.

In a search of the maximum 10 the function 11 can now have all possible combinations of possible speeds v ₁ , v _{2 of} the goals 2-1 . 2-2 be iterated and the maximum 10 be searched.

eine Winkelschätzung anschließen, die für die zwei Ziele 2-1, 2-2 die Winkel, z.B. die Azimutwinkel oder Elevationswinkel, gegenüber dem Radarsensor 1 berechnet.In one embodiment of the method, the calculation of the relative velocities can take place

connect an angle estimate that for the two goals 2-1 . 2-2 the angles, eg the azimuth angle or elevation angle, relative to the radar sensor 1 calculated.

Based on the already calculated Moore-Penrose pseudo inverse A ⁺ can in a very simple way by multiplying the measurement vector a _m (n) by the Moore-Penrose pseudo inverse A ⁺ the separated individual measured values (y _{1, m} (v), y _{2, m} (v)). For two goals 2-1 . 2-2 The individual measured values (y _{1, m} (v), y _{2, m} (v)) result in: [y _{m, 1} (n) y _{m, 2} (n)] = A ⁺ a _m (n).

Before or after the calculation of the angles for w ₁ , w ₂ for the calculation results can be made plausible.

For example, the individual measurements (y _{1, m} (v), y _{2, m} (v)) may be weighted based on their amplitude difference, and only the destination for an amplitude difference greater than a predetermined first limit 2-1 . 2-2 , which corresponds to the larger amplitude, as a target 2-1 . 2-2 be recognized.

Additionally or alternatively, the amplitudes of the individual measurements (y _{1, m} (v), y _{2, m} (v)) may each be related to a total signal power of the received signal 16 or the signal-to-noise ratio of the received signal 16 are set, the target corresponding to the respective individual measured value (y _{1, m} (v), y _{2, m} (v)) 2-1 . 2-2 is only detected if the corresponding individual measured value (y _{1, m} (v), y _{2, m} (v)) is above a second limit.

2 shows a block diagram of an embodiment of the radar sensor according to the invention 1 ,

The radar sensor 1 has a signal generator 14 on, consisting of two signal sequences 4-1 . 4-2 a frequency modulated transmit signal 3 generated and via a transmitting / receiving antenna 13 outputs.

Each of the two signal sequences 4-1 . 4-2 has a variety of ramps in the frequency response 5-1 - 5-6 on, between which gaps 6-1 - 6-6 are arranged.

To the transmission signal 3 to generate, the signal generator interleaves 14 the two signal sequences 4-1 . 4-2 such that each of the ramps 5-3 - 5-6 the second signal sequence 4-2 in the gaps 5-1 - 5-3 the first signal sequence 4-1 lie and vice versa. Here is the time offset 7 given, with which the ramps 5-1 - 5-6 into the respective gaps 6-1 - 6-6 be inserted. In 2 eg a ramp starts 5-3 - 5-6 the second signal sequence 4-2 right after the end of a ramp 5-1 - 5-3 the first signal sequence 4-1 , The temporal offset 7 but may be larger or smaller in other embodiments.

für die zwei Ziele 2-1, 2-2 separiert.The transmission signal 3 is next to the transmit / receive antennas 13 also a mixer 19 additionally provided by the transmit / receive antenna 13 received signals 16 receives and this with the respective transmission signal 3 mixed. The result of this mixture is for each of the signal sequences 4-1 . 4-2 a baseband signal 8-1 . 8-2 , In 2 become the baseband signals 8-1 . 8-2 first a single speed estimate 9 in which is attempted a relative velocity v for a single target 2-1 . 2-2 to calculate. If this fails or only with a very low quality, the baseband signals become 8-1 . 8-2 an evaluation device 15 fed, which is formed, a method according to the invention, such as 1 to perform described and a maximum 10 the function 11 to determine. From the maximum 10 , which may be a vector with two components, for example, then become the relative speeds

for the two goals 2-1 . 2-2 separated.

As in the calculation of the function 10 already calculated the Moore Penrose pseudoinverse A ⁺ , this can be a computing device 18 from which the separated individual measured values (y _{1, m} (v), y _{2, m} (v)) are calculated and an angle estimator 17 that provides for the two goals 2-1 . 2-2 the angles w ₁ , w ₂ , eg elevation angle w ₁ , w ₂ or azimuth angle w ₁ , w ₂ , calculated.

In 2 For reasons of clarity, not shown are further possible devices that can perform a plausibility check of the results of the multi-target speed estimation and the multi-target angle estimation.

For example, the individual measurements (y _{1, m} (v), y _{2, m} (v)) may be weighted based on their amplitude difference, and only the destination for an amplitude difference greater than a predetermined first limit 2-1 . 2-2 , which corresponds to the larger amplitude, as a target 2-1 . 2-2 be recognized.

Additionally or alternatively, the amplitudes of the individual measurements (y _{1, m} (v), y _{2, m} (v)) may each be related to a total signal power of the received signal 16 or the signal-to-noise ratio of the received signal 16 are set, the target corresponding to the respective individual measured value (y _{1, m} (v), y _{2, m} (v)) 2-1 . 2-2 is only detected if the corresponding individual measured value (y _{1, m} (v), y _{2, m} (v)) is above a second limit.

Although the present invention has been described above with reference to preferred embodiments, it is not limited thereto, but modifiable in a variety of ways. In particular, the invention can be varied or modified in many ways without deviating from the gist of the invention.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been 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.

Cited patent literature

• DE 102013200951 A1 [0006]

Claims (12)

Method for detecting at least two destinations ( 2-1 . 2-2 ) with a radar sensor ( 1 ), comprising: transmitting (S1) at least one frequency-modulated transmit signal ( 3 ) containing at least two signal sequences ( 4-1 . 4-2 ), which in the frequency course each successive ramps ( 5-1 - 5-6 ) with gaps in between ( 6-1 - 6-6 ), wherein the at least two signal sequences ( 4-1 . 4-2 ) with a predetermined time offset ( 7 ) are interleaved in such a way that in each case a first ramp ( 5-1 - 5-6 ) each of the at least two signal sequences ( 4-1 . 4-2 ) is output before a second ramp ( 5-1 - 5-6 ) one of the at least two signal sequences ( 4-1 . 4-2 ) is output; Mixing (S2) the emitted frequency-modulated transmission signal (S2) 3 ) with received signals ( 16 ) to baseband signals ( 8-1 . 8-2 ); and performing (S3) a multi-target velocity estimation based on the baseband signals ( 8-1 . 8-2 ) based on multi-objective optimization. The method of claim 1, wherein prior to performing the multi-target speed estimation, a single-speed estimate ( 9 ), and the multi-target speed estimation is performed only when the single-speed estimation ( 9 ) gives no clear result. Method according to one of the preceding claims, wherein in the multi-target speed estimation the maximum ( 10 ) of a function ( 11 ), which indicates a relative velocity spectrum, for possible velocity values (v1, v2) of the at least two targets ( 2-1 . 2-2 ) is calculated. The method of claim 3, wherein the function ( 11 ) for two goals ( 2-1 . 2-2 ) is defined as

where v ₁ and v ₂ for the function ( 11 ) speeds of the first target ( 2-1 ) and the second objective ( 2-2 ), arg max stands for the argumentum maximi function, and P _A (v ₁ , v ₂ ) stands for a projection matrix resulting from a control matrix A. The method of claim 4, wherein the control matrix A is defined as: A = [a (v ₁ ), a (v ₂ )] and a (v ₁ ) and a (v ₂ ) each as control vectors of the type

where I is the number of signal sequences ( 4-1 . 4-2 ), c is the speed of light and T _1i for the respective signal sequence ( 4-1 . 4-2 ) the predetermined time interval to the first signal sequence ( 4-1 . 4-2 ). The method of claim 5, wherein the projection matrix P _A (v ₁ , v ₂ ) is defined as P _A (v ₁ , v ₂ ) = A (AA ^H ) ^-1 A ^H = AA ⁺ where A ⁺ stands for the Moore-Penrose pseudoinverse of A. Method according to one of the preceding claims, comprising: estimating angles between the radar sensor ( 1 ) and the at least two objectives ( 2-1 . 2-2 ), especially after the multi-target speed estimation. The method of claims 6 and 7, wherein estimating angles is done by estimating angles based on Moore Penrose pseudoinverse A ⁺ ; wherein from measurement signals contained in a measurement vector a _m (n), based on the Moore-Penrose pseudoinverse A ^+, individual measured values (y _{1, m} (v), y _{2, m} (v)) for each of the targets ( 2-1 . 2-2 ) which are fed to an angle estimator. Method according to claim 8, wherein the individual measured values (y _{1, m} (v), y _{2, m} (v)) for two targets ( 2-1 . 2-2 ) are calculated according to [y _{m, 1} (n) y _{m, 2} (n)] = A ⁺ a _m (n). Method according to one of claims 8 and 9, wherein the individual measured values (y _{1, m} (v), y _{2, m} (v)) are weighted on the basis of their amplitude difference and only the target if the amplitude difference exceeds a predetermined first limit value ( 2-1 . 2-2 ), which corresponds to the greater amplitude, as the target ( 2-1 . 2-2 ) is recognized. Method according to one of claims 8 to 10, wherein the amplitudes of the individual measured values (y _{1, m} (v), y _{2, m} (v)) in each case in relation to a total signal power of the received signal ( 16 ) or the signal-to-noise ratio, the target corresponding to the individual measured value (y _{1, m} (v), y _{2, m} (v)) ( 2-1 . 2-2 ) is only detected if the corresponding individual measured value (y _{1, m} (v), y _{2, m} (v)) is above a second limit value. Radar sensor ( 1 ) for recording at least two destinations ( 2-1 . 2-2 ), with at least one transmit / receive antenna ( 13 ); with a signal generator ( 14 ), which is formed, at least one frequency-modulated transmit signal ( 3 ) via the at least one transmitting / receiving antenna, which at least two signal sequences ( 4-1 . 4-2 ), which in the frequency course each successive ramps ( 5-1 - 5-6 ) with gaps in between ( 6-1 - 6-6 ), wherein the at least two signal sequences ( 4-1 . 4-2 ) with a predetermined time offset ( 7 ) are interleaved in such a way that in each case a first ramp ( 5-1 - 5-6 ) each of the at least two signal sequences ( 4-1 . 4-2 ) is output before a second ramp ( 5-1 - 5-6 ) one of the at least two signal sequences ( 4-1 . 4-2 ) is output; and with an evaluation device ( 15 ) configured to perform a multi-target velocity estimation according to one of the preceding claims.

Images