CN117970265A

CN117970265A - Target recognition method, device and storage medium

Info

Publication number: CN117970265A
Application number: CN202410054826.XA
Authority: CN
Inventors: 何小静
Original assignee: Shenzhen Saifang Technology Co ltd
Current assignee: Shenzhen Saifang Technology Co ltd
Priority date: 2024-01-15
Filing date: 2024-01-15
Publication date: 2024-05-03

Abstract

The application relates to the technical field of radars, and discloses a target identification method which comprises the following steps: acquiring a first Digital Beam Forming (DBF) directional diagram of a target to be identified, wherein the first DBF directional diagram is obtained by performing DBF angle measurement on radar target echo signals; determining a plurality of first amplitude maximum points; acquiring i angle values corresponding to the first i first amplitude maximum value points according to the sequence from large to small; selecting an angle value A _k of a current target to be identified from the i angle values, and filtering a signal corresponding to A _k from radar target echo signals to obtain a residual signal component; determining a second DBF pattern corresponding to the residual signal component; in the second DBF direction diagram, if A _k belongs to the angle value corresponding to the amplitude minimum point, and i-1 angle values are not the largest angle value corresponding to i-1 second amplitude maximum points except A _k in the i angle values, determining that the current target to be identified is a real target. By the mode, the method and the device realize accurate identification of the real target according to the radar target echo signal.

Description

Target recognition method, device and storage medium

Technical Field

The embodiment of the application relates to the technical field of radars, in particular to a target identification method, target identification equipment and a storage medium.

Background

Radar (e.g., a vehicle-mounted millimeter wave radar) is widely used in the field of automatic driving due to its advantages such as high stability and no adverse weather effect. As an important component of an unmanned automobile, there is an increasing demand in the market for a radar that is inexpensive and excellent in performance, that is, high accuracy of radar detection of targets, wide application, and the like.

To ensure low price of the radar, the cost of the radar is generally reduced, so that the antenna units and related components required in the radar are correspondingly reduced, but performance of the radar is reduced after the antenna units and related components are reduced, for example, for scenes with more clutter, noise, interference and the like (such as complex urban traffic environment), the accuracy of detecting and identifying targets by the radar is reduced. Therefore, for the radar with a simplified structure, how to improve the accuracy of identifying the target is a problem to be solved at present.

Disclosure of Invention

In view of the above problems, embodiments of the present application provide a target recognition method, apparatus, and storage medium, which are used to solve the problem in the prior art that the accuracy of recognizing a target by a radar with a simplified structure is not high.

According to an aspect of an embodiment of the present application, there is provided a target recognition method, the method including: acquiring a first Digital Beam Forming (DBF) directional diagram of a target to be identified, wherein the first DBF directional diagram is obtained by DBF angle measurement of radar target echo signals; determining a plurality of first amplitude maximum points according to all amplitude points of the first DBF directional diagram, wherein each first amplitude maximum point is an amplitude point with amplitude values larger than adjacent left and right amplitude values in all amplitude points of the first DBF directional diagram; according to the sequence from large to small of the first amplitude maximum points in the plurality of first amplitude maximum points, i angle values corresponding to the first i first amplitude maximum points are obtained, wherein the first i first amplitude maximum points are points corresponding to the target to be identified, i is a positive integer, and i is larger than 1; selecting an angle value A _k from the i angle values, wherein the target corresponding to the A _k is a current target to be identified, and filtering a signal corresponding to the A _k from the radar target echo signal to obtain a residual signal component, wherein k is a positive integer, and k is less than or equal to i; determining a second DBF pattern corresponding to the residual signal component; in the second DBF pattern, if the a _k belongs to an angle value corresponding to an amplitude minimum value point, and the remaining i-1 angle values except the a _k in the i angle values are not angle values corresponding to the largest i-1 second amplitude maximum value points, determining that the current target to be identified is a real target, where the amplitude minimum value point is an amplitude point in the second DBF pattern, where the amplitude value is smaller than adjacent left and right amplitude values, and the second amplitude maximum value point is an amplitude point in the second DBF pattern, where the amplitude value is greater than adjacent left and right amplitude values.

In an optional manner, the radar target echo signal is a signal reflected by an object in the detection area after the radar array transmits electromagnetic waves to the detection area, and the filtering the signal corresponding to the a _k from the radar target echo signal to obtain a residual signal component includes: determining a theoretical signal Y _k of the current target to be identified under the detection of the radar array according to the A _k; determining a signal component estimated value S _k of the Y _k in the first DBF directional diagram according to the value corresponding to the A _k in the first DBF directional diagram and the Y _k, wherein the value is a complex-form value; and filtering the S _k from the radar target echo signal to obtain the residual signal component.

In an optional manner, if the a _k belongs to an angle value corresponding to a minimum value point of amplitude, and the remaining i-1 angle values except the a _k are not angle values corresponding to the maximum i-1 second maximum value points of amplitude, determining that the current target to be identified is a real target includes: determining a plurality of second amplitude maximum points and a plurality of amplitude minimum points according to all the amplitudes of the second DBF directional diagram; sequencing the second amplitude maximum points from large to small to obtain first i-1 second amplitude maximum points; if the a _k belongs to an angle value corresponding to one amplitude minimum point of the amplitude minimum points, and the remaining i-1 angle values except the a _k in the i angle values do not belong to an angle value corresponding to the first i-1 second amplitude maximum points, determining that the current target to be identified is the real target.

In an alternative, i is 3.

In an alternative, the method further comprises: in the second DBF pattern, if the a _k does not belong to an angle value corresponding to a minimum value point of amplitude, or if the remaining i-1 angle values other than the a _k are not the angle values corresponding to the maximum i-1 second maximum value points of amplitude, selecting an unselected angle value a _k from the i angle values; and turning to the step of filtering the signal corresponding to the A _k from the radar target echo signal to obtain a residual signal component.

In an alternative manner, before the obtaining the first DBF pattern of the object to be identified, the method further includes: acquiring the radar target echo signal; performing Fourier transform of distance dimension and speed dimension on the radar target echo signal to obtain distance-Doppler RD spectrum data; obtaining a target array original signal S _origin of a target distance and a target speed in the RD spectrum data; the obtaining the first DBF pattern of the object to be identified includes: and performing DBF angle measurement on the S _origin to obtain the first DBF directional diagram.

In an alternative manner, the determining, according to the a _k, the theoretical signal Y _k of the target to be identified under the radar array detection includes: calculating an index value according to a first preset parameter, the A _k, the radar array arrangement position, the pitching angle value of the real target and the wavelength of the radar target echo signal; and calculating the theoretical signal Y _k by taking a preset numerical value as a base number and the index value as an index.

In an alternative manner, the determining, according to the value corresponding to the a _k in the first DBF pattern and the Y _k, the signal component estimated value S _k of the Y _k in the first DBF pattern includes: the step S _k) is calculated by the formula S _k＝Y_k*S_map(A_k), wherein S _map(A_k) is the numerical value.

According to another aspect of an embodiment of the present application, there is provided an object recognition apparatus including: the system comprises a first acquisition module, a second acquisition module and a first Digital Beam Forming (DBF) directional diagram, wherein the first DBF directional diagram is obtained by DBF angle measurement of radar target echo signals; the first determining module is used for determining a plurality of first amplitude maximum points according to all amplitude points of the first DBF directional diagram, wherein each first amplitude maximum point is an amplitude point with amplitude values larger than adjacent left and right amplitude values in all amplitude points of the first DBF directional diagram; the second acquisition module is used for acquiring i angle values corresponding to the first amplitude maximum points according to the sequence from the large to the small of the first amplitude maximum points in the plurality of first amplitude maximum points, wherein the first i first amplitude maximum points are points corresponding to the target to be identified, i is a positive integer, and i is larger than 1; the filtering module is used for selecting an angle value A _k from the i angle values, wherein the target corresponding to the A _k is a current target to be identified, and filtering the signal corresponding to the A _k from the radar target echo signal to obtain a residual signal component, wherein k is a positive integer, and k is less than or equal to i; a second determining module, configured to determine a second DBF pattern corresponding to the residual signal component; and a third determining module, configured to determine, in the second DBF pattern, that the current target to be identified is a real target if the a _k belongs to an angle value corresponding to a minimum amplitude value point, and the remaining i-1 angle values except the a _k are not angle values corresponding to the i-1 maximum amplitude value points with the largest angle values, where the minimum amplitude value point is an amplitude value point in the second DBF pattern, where the amplitude value is smaller than adjacent left and right amplitude values, and the second maximum amplitude value point is an amplitude value point in the second DBF pattern, where the amplitude value is greater than adjacent left and right amplitude values.

According to another aspect of an embodiment of the present application, there is provided an object recognition apparatus including: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is used to store executable instructions that cause the processor to perform the operations of the object recognition method as described above.

According to yet another aspect of embodiments of the present application, there is provided a computer-readable storage medium having stored therein executable instructions that, when executed, perform operations of the object recognition method as described above.

Since the target array original signal S _origin is the obtained actual signal, if a _k is the angle value corresponding to the actual target, the theoretical signal Y _k calculated by a _k is close to the actual signal value, so in the embodiment of the application, whether a _k is the angle value corresponding to the actual target can be accurately determined by filtering the signal component estimated value S _k of the theoretical signal Y _k corresponding to the angle value a _k in the target array original signal S _origin in the first DBF direction diagram, and then filtering the second DBF direction diagram obtained after filtering the signal component estimated value S _k corresponding to a _k.

The foregoing description is only an overview of the technical solutions of the embodiments of the present application, and may be implemented according to the content of the specification, so that the technical means of the embodiments of the present application can be more clearly understood, and the following specific embodiments of the present application are given for clarity and understanding.

Drawings

The drawings are only for purposes of illustrating embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:

Fig. 1 shows a flow chart of a target recognition method according to an embodiment of the present application;

FIG. 2 shows a schematic diagram of a radar array in a simulation experiment provided by an embodiment of the present application;

FIG. 3 is a schematic diagram of a first DBF pattern and a DBF pattern after filtering theoretical signals of a real target according to an embodiment of the present application;

FIG. 4 is a schematic diagram of another first DBF pattern and a DBF pattern after filtering out theoretical signals of a false peak target located on the left side of a real target according to an embodiment of the present application;

FIG. 5 is a schematic diagram of another first DBF pattern and a DBF pattern after filtering out theoretical signals of a false peak target located on the right side of a real target according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of an object recognition device according to an embodiment of the present application;

Fig. 7 is a schematic structural diagram of an object recognition device according to an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein.

Radar target echoes refer to signals reflected back by the surface of a target object after electromagnetic waves emitted by radar reach the target object. In general, the speed information (movement speed information relative to the radar), the distance information (distance information between the radar) and the angle information (angle and position information relative to the radar) of different objects are not identical, so that there is a certain difference in the signals reflected from the surfaces of different objects. Based on this, the target object can be identified by the radar target echo. For example, in urban traffic environment, if a vehicle is a target object of interest, a building or the like belongs to a non-target object, after electromagnetic waves are emitted to a detection area by a radar, the electromagnetic waves emitted by the radar reach a plurality of objects including the vehicle, the building or the like, and then the radar target echo signal includes signals reflected by the objects such as the vehicle, the building or the like. Since the reflection characteristics, speed, distance and angle information of the vehicle and the building are not exactly the same, the vehicle can be identified from the radar target echo.

Digital beam forming (Digital Beam Forming, DBF) techniques can weight sum echo signals received by a radar array to form beams in specific directions, so as to achieve the receiving and enhancing of desired signals, wherein the specific directions are directions of objects of interest (i.e., directions and angles of real objects). Therefore, if the radar target echo signal is used for DBF angle measurement, after the DBF directional diagram is obtained, the angle value corresponding to the real target can be determined according to the amplitude value in the DBF directional diagram, so that the positioning of the real target can be further realized. Wherein, in the DBF pattern, the abscissa represents the azimuth; the ordinate represents the amplitude and phase result, which is a complex number, and after the obtained complex number of amplitude and phase results are modulo, the obtained modulo is the amplitude, and the amplitude represents the signal intensity in the direction.

In order to facilitate analysis, in the embodiment of the present application, the magnitudes of the magnitude points in the first DBF pattern and the second DBF pattern are obtained by performing a modulo operation on a complex-form magnitude-phase result in the DBF pattern obtained by the DBF technique, and converting the modulo value into dB units.

The present inventors found in the study that if the angle value corresponding to the maximum value (i.e., peak value) of the amplitude value in the DBF pattern is directly determined as the angle value of the real target, there is a certain error, because the radar itself may have a certain error, so the peak value in the DBF pattern determined based on the radar target echo signal may be a false peak value caused by a side lobe or a grating lobe, i.e., the peak value is not the peak value corresponding to the real target signal.

Based on the above situation, the present application provides a target recognition method, in the method, a first DBF pattern obtained after DBF angle measurement is performed on a radar target echo is obtained, a residual signal component is obtained by filtering a signal corresponding to the angle value a from a radar target echo signal according to a peak value and a corresponding angle value a of the first DBF pattern, and further DBF angle measurement is performed on the residual signal component to obtain a second DBF pattern, so that whether the angle value a is an angle value of a real target can be accurately determined based on the second DBF pattern.

Fig. 1 shows a flowchart of a target recognition method provided by an embodiment of the present application, where the method is performed by a terminal device. The terminal device may be a radar or a device in communication with a radar to obtain a radar detection signal. The terminal device may be a terminal device comprising one or more processors, which may be Central Processing Units (CPUs), or Application SPECIFIC INTEGRATED Circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present application, without limitation. The one or more processors comprised by the terminal device may be the same type of processor, such as one or more CPUs; but may be different types of processors such as, without limitation, one or more CPUs and one or more ASICs. As shown in fig. 1, the method comprises the steps of:

step 110: a first DBF (direct current) directional diagram of a target to be identified is obtained, wherein the first DBF directional diagram is obtained by DBF angle measurement on radar target echo signals.

As previously described, DBF angle measurement can typically be performed using radar target echo signals to obtain a DBF pattern. In some embodiments, if all objects within the radar detection range need not be identified, but only the target object (e.g. the object at the set target distance and target speed) is concerned, the following steps a 1-a 3 may be performed to obtain the target array original signal S _origin including only the target object before performing the present step 110, and then the first DBF pattern is generated based on S _origin.

Step a1: and acquiring radar target echo signals.

The radar target echo signal is a signal reflected by an object in the detection area after the radar array transmits electromagnetic waves to the detection area.

Step a2: and carrying out Fourier transformation of the distance dimension and the speed dimension on the radar target echo signal to obtain distance-Doppler RD spectrum data.

Where the fourier transform is a linear integral transform used for the transformation of a function (commonly referred to as a "signal" applied) between the time and frequency domains. The inventor finds that the existing method for filtering radar echo signals mostly filters non-attention signals such as interference, clutter and the like, and the method for filtering radar array signals based on two-dimensional (distance dimension and speed dimension) Fourier transform mostly utilizes Doppler effect of the echo signals, so that less research is carried out on filtering radar array signals. In the embodiment of the application, the radar array signals obtained based on the two-dimensional Fourier transform are utilized for filtering, and the target is accurately identified according to the filtered signals.

The Doppler effect refers to the phenomenon of frequency change caused by relative motion between a transmitting source and a receiver, and can be divided into a static receiver and a motion transmitting source, a motion receiver and a static transmitting source, and the like, and the Doppler effect has application in the fields of radar, sonar, astronomy, optics and the like.

In this step, after performing fourier transform in a distance dimension on the radar target echo signal, fourier transform in a velocity dimension is continued on the basis, so as to obtain distance-Doppler (RD) data. The RD spectrum data comprises distance information of each object from the radar in the radar detection range and speed information of each object relative to the radar.

Step a3: and obtaining a target array original signal S _origin of the target distance and the target speed in the RD spectrum data.

Wherein the target distance and target speed may be set as desired, specifically, if a target object at a certain distance and a certain speed is to be focused on or identified, the distance and the speed are set as the target distance and the target speed.

In some embodiments, the target array raw signal S _origin is obtained by detecting a target object at a target distance and a target speed in the RD spectrum data through Constant false alarm (Constant FALSE ALARM RATE, CFAR) detection. The CFAR detection is to judge the signal and noise output by the receiver under the condition of keeping the false alarm probability constant so as to determine whether the target signal exists, the input is the result obtained by the radar echo signal after the fourier transformation of the distance dimension and the speed dimension, and the output is the detected target result.

In step 110, a first DBF pattern is obtained by performing DBF angle measurement on the target array original signal S _origin. The specification of the first DBF pattern is 1×m, where m is the number of angular search points of the DBF angle measurement, and may be set as required, for example, 300, 500, 900, etc.

Step 120: and determining a plurality of first amplitude maximum points according to all amplitude points of the first DBF directional diagram, wherein each first amplitude maximum point is an amplitude point with amplitude larger than the adjacent left and right amplitude values in all amplitude points of the first DBF directional diagram.

Wherein, the first DBF pattern is continuous (beam shape is continuous), and for the first DBF pattern, if the amplitude of a point is greater than the amplitude of the point adjacent to the first DBF pattern and located on the left side and the amplitude of the point on the right side, the point is the first amplitude maximum point. From the beam shape, the peak point of each wave in the first DBF pattern is the first amplitude maximum point. It is understood that the plurality of first amplitude maxima points includes a maximum amplitude point in the first DBF pattern.

Step 130: and acquiring i angle values corresponding to the first i first amplitude maximum points according to the sequence from the large to the small of the first amplitude maximum points in the first amplitude maximum points, wherein i is a positive integer, and i is greater than 1.

As described above, since the beam in the specific direction can be formed by the DBF technology, the receiving and enhancing of the expected signal are realized, in this step, after the i angle values corresponding to the first i maximum amplitude value points are found, whether the angle value corresponding to the real target exists in the i angle values can be further judged subsequently. Where i may be set as desired, e.g., i is 2, 3, etc.

In the case of a false peak caused by grating lobes, there will be a feature that is symmetrical about the target. I.e. two false peaks typically occur, distributed on the left and right sides of the object, respectively. Therefore, in the embodiment of the present application, considering a real target and 1 side lobe or grating lobe decoy in the left-right angle search range, it is preferable to set i to 3.

Step 140: selecting a target corresponding to an angle value A _k,A_k from i angle values as a current target to be identified, and filtering a signal corresponding to A _k from radar target echo signals to obtain a residual signal component, wherein k is a positive integer, and k is less than or equal to i.

If the current target to be identified is a real target, that is, a _k is an angle value corresponding to the real target, in this step, after filtering signals corresponding to a _k from the radar target echo signals, signals corresponding to false targets caused by side lobes and grating lobes are filtered together, so in this step, after filtering signals corresponding to a _k from the radar target echo signals, whether a _k is an angle value corresponding to the real target can be further analyzed according to the obtained residual signal component.

In some embodiments, in order to improve the accuracy of the obtained residual signal component, the residual signal component is determined by the following steps b1 to b 3.

Step b1: and determining a theoretical signal Y _k of the current target to be identified under the detection of the radar array according to A _k.

Wherein, one skilled in the art can calculate Y _k by using a correlation formula or perform a correlation simulation experiment to obtain Y _k according to the need.

Step b2: and determining a signal component estimated value S _k of Y _k in the first DBF pattern according to the value corresponding to A _k in the first DBF pattern and Y _k, wherein the value is a complex value.

The signal component estimated value S _k is an estimated value corresponding to the theoretical signal Y _k corresponding to a _k in the first DBF pattern. The skilled person can calculate S _k using a correlation formula or perform a correlation simulation experiment to obtain S _k as required.

In some embodiments, in order to improve the accuracy of the obtained S _k, S _k is calculated by using the formula S _k＝Y_k*S_map(A_k), where S _k has a specification of 1×n, n is the total number of units of the radar array that emits electromagnetic waves to the detection area, and S _map(A_k) is a value corresponding to a _k in the first DBF pattern, where the value is a complex-form value, and the value is also referred to as an amplitude-phase result corresponding to a _k in the first DBF pattern. As described above, in the first DBF pattern, the amplitude corresponding to A _k is the result of modeling the complex number and converting the modeling value into dB units.

Step b3: and filtering S _k from the radar target echo signal to obtain a residual signal component.

The residual signal component S _k 'may be obtained after filtering S _k from the radar target echo signal, where the residual signal component S _k' represents the remaining signal component after filtering the theoretical signal component estimated value corresponding to a _k from the target array original signal S _origin. In this step, the formula can be usedS _k' is calculated.

Step 150: a second DBF pattern corresponding to the residual signal component is determined.

And performing DBF angle measurement on the residual signal component to obtain a second DBF directional diagram.

Step 160: in the second DBF direction diagram, if A _k belongs to an angle value corresponding to an amplitude minimum value point, and the remaining i-1 angle values are not the angle values corresponding to the largest i-1 second amplitude maximum value points except A _k, determining the current target to be identified as a real target, wherein the amplitude minimum value point is an amplitude point of which the amplitude value is smaller than the adjacent left and right amplitude values in the second DBF direction diagram, and the second amplitude maximum value point is an amplitude point of which the amplitude value is larger than the adjacent left and right amplitude values in the second DBF direction diagram.

Since the target array original signal S _origin is the obtained actual signal, if a _k is the angle value corresponding to the actual target, the theoretical signal Y _k obtained by calculating a _k is close to the actual signal value, so in the embodiment of the application, after filtering the signal component estimated value S _k of the theoretical signal Y _k corresponding to the angle value a _k in the target array original signal S _origin in the first DBF direction diagram, the residual signal component is subjected to DBF angle measurement, so as to obtain the second DBF direction diagram, and then, based on the second DBF direction diagram, whether the theoretical signal component estimated value S _k corresponding to a _k is filtered out can be judged, and whether a _k is the angle value corresponding to the actual target can be further accurately judged. Specifically, the magnitudes of all the magnitude points in the second DBF pattern may be sorted according to magnitude points corresponding to the i angle values in the second DBF pattern, if a _k belongs to an angle value corresponding to a magnitude minimum point, and the remaining i-1 angle values are not the angle values corresponding to the largest i-1 second magnitude maximum points except a _k in the i angle values, it may be determined that a _k is the angle value corresponding to the real target, and i-1 angle values except a _k are the angle values of the false target caused by the sidelobe and the grating lobe in the i angle values.

In the case of a false peak caused by grating lobes, there will be a feature that is symmetrical about the target. I.e. two false peaks typically occur, distributed on the left and right sides of the object, respectively. Based on this, in order to further improve the accuracy of target recognition, in the embodiment of the present application, i is 3, and step 160 includes:

Step c1: and determining a plurality of second amplitude maximum points and a plurality of amplitude minimum points according to all the amplitudes of the second DBF pattern.

The present step is similar to step 120, and thus, the implementation of the present step may refer to step 120, which is not described herein.

Step c2: and sequencing the plurality of second amplitude maximum points from large to small to obtain the first i-1 second amplitude maximum points.

Step c3: if A _k belongs to the angle value corresponding to one amplitude minimum point in the amplitude minimum points, and the remaining i-1 angle values except A _k in the i angle values do not belong to the angle values corresponding to the first i-1 second amplitude maximum points, determining that the current target to be identified is a real target.

If a _k in the second DBF pattern belongs to the angle value corresponding to the minimum point of the amplitude, it is indicated that the target component corresponding to a _k has been effectively filtered. If the two angle values except the A _k in the 3 angle values do not belong to the angle values corresponding to the first two second amplitude maximum value points in the second DBF directional diagram, determining that the two angle values are the angle values of false targets caused by side lobes and grating lobes.

As described above, since the false peak value caused by the grating lobe has a feature of symmetry about the target, after three first amplitude maximum points arranged in the first DBF direction diagram are found, three angle values corresponding to the three first amplitude maximum points are obtained, and if the angle value corresponding to one of the three first amplitude maximum points is the angle value corresponding to the real target, the other two first amplitude maximum points are the false peak value caused by the grating lobe. Therefore, in the embodiment of the present application, if a _k is the angle value corresponding to the real target, after filtering out the theoretical signal component estimated value S _k corresponding to the angle value a _k in the original signal of the target array, the remaining two signals caused by the grating lobes are also filtered out, and the amplitude corresponding to the two false peak values caused by the grating lobes in the second DBF direction diagram is no longer the peak value, so that it can be accurately determined whether a _k is the angle value corresponding to the real target.

For the second DBF pattern, if a _k belongs to an angle value corresponding to the amplitude minimum point, it may be determined that the theoretical signal component estimated value S _k corresponding to a _k has been filtered out, and further according to whether the amplitude point corresponding to i-1 angle values does not belong to the first i-1 second amplitude maximum points in the second DBF pattern except for a _k in the i angle values, it may be determined whether a _k is an angle value corresponding to a real target. If the magnitudes corresponding to the i-1 angle values are not the magnitudes of the first i-1 second maximum magnitude points in the second DBF pattern, it may be determined that a _k is the angle value corresponding to the real target, otherwise a _k may be a side lobe, a grating lobe decoy, or multiple targets with the same distance and the same speed may exist in the radar target echo component, and a _k is only one of the real target angle values.

In some embodiments, based on the target recognition method provided in fig. 1, after step 160, the method further includes: in the second DBF pattern, if a _k does not belong to the angle value corresponding to the amplitude minimum point, or if the remaining i-1 angle values are not the angle values corresponding to the largest i-1 second amplitude maximum points except a _k in the i angle values, selecting an unselected angle value a _k from the i angle values, and switching to the step of filtering the signal corresponding to a _k from the radar target echo signal.

The angle value a _k may be selected randomly, or may be selected in a specific order, for example, in order of decreasing angle value or in order of increasing angle value. If i=3, among the 3 angle values corresponding to the first amplitude maximum points obtained in the order from large to small of the first amplitude maximum points, among the angle values corresponding to the first 3 first amplitude maximum points, the angle value between the other two angle values of the first DBF pattern is most likely to be the angle value corresponding to the target object, so that step 140 can first select the angle value to perform subsequent judgment, thereby determining the angle value corresponding to the real target as soon as possible, and improving the target recognition efficiency.

When step 140 is performed for the first time, k is 1; the first time this step is performed, k is 2, the second time this step is performed, k is 3, and so on.

In some embodiments, to improve the accuracy of the obtained theoretical signal Y _k, step b1 includes:

step b11: and calculating an index value according to the first preset parameter, the A _k, the radar array arrangement position, the pitching angle value of the real target and the wavelength of the radar target echo signal.

Step b12: and calculating a theoretical signal Y _k by taking a preset numerical value as a base and an index value as an index.

If only targets with different azimuth dimensions under the same pitching dimension are considered, the DBF angle measurement range is-90 degrees, and the default pitching angle is 0 degrees, the formula can be utilizedCalculating a theoretical signal Y _k, wherein the Array _TR is a radar Array arrangement position, the specification of the Array is 1 x N, N is the total number of units of a radar Array for transmitting electromagnetic waves to a detection area, and N is the product of a transmitting Array unit and a receiving Array unit for a multiple-transmitting multiple-receiving radar Array; lambda is the wavelength of the signal reflected back by the object in the detection area, i.e. the wavelength of the radar target echo signal, which can be calculated by the formula/>And c is the light speed, and f is the center frequency of the radar echo signal. Wherein the first preset parameter is-2pi.j, and the preset value is e.

If only targets of different azimuth dimensions under the same pitching dimension are considered, the DBF angle measurement range is-90 degrees, and the default pitching angle is not 0 degrees, the formula can be utilizedThe theoretical signal Y _k is calculated according to a formula, wherein x _arrayTR is the position of an x axis (horizontal direction) of the radar array arrangement, Y _arrayTR is the position of a Y axis (vertical pitching direction) of the radar array arrangement, and the position of the x axis (vertical pitching direction)The target pitching angle is a target pitching angle, wherein for a radar array with a pitching angle measuring function, the pitching angle/>, can be measured through DBF, a amplitude comparison method, a phase comparison method and other angle measuring methods

If only the targets of different azimuth dimensions under the same pitching dimension are considered, the DBF angle measurement range is 0-180 degrees, and the default pitching angle is 0 degrees, the formula can be utilizedThe theoretical signal Y _k is calculated.

In order to verify the accuracy of the target recognition method provided by the embodiment of the present application, description will be made herein with reference to relevant simulation experiment data. Fig. 2 is a schematic diagram of a radar array in a simulation experiment according to an embodiment of the present application. As shown in fig. 2, the radar array is a 3-transmit and 4-receive array, and the arrangement of the transmitting array units in the figure refers to the arrangement of the radar transmitting array units; the arrangement of the receiving array units refers to the arrangement of the radar receiving array units. The simulation experiment sets the simulated real target angle at the azimuth 6 degree (default pitching is 0 degree) position, and the angle scanning range of the DBF angle measurement setting is-90 degrees: 0.2 °:90 °, the number of angular search points m=900, random noise is added, and the target signal-to-noise ratio is set to 20dB. Fig. 3 is a schematic diagram of a first DBF pattern and a DBF pattern after filtering theoretical signals of a real target according to an embodiment of the present application. As shown in fig. 3, 1 is a first DBF pattern, and 2 is a DBF pattern after filtering theoretical signals of a real target. In the first DBF directional diagram, the point a ₁ is a point corresponding to a real target, the angle is 6 degrees, and the amplitude is 21.9789dB; the point b ₁ is a false peak value point which is positioned at the left side of a real target and is caused by grating lobes, the angle is 15.8 degrees below zero, and the amplitude is 19.5943dB; the point c ₁ is a false peak point on the right side of the real target caused by grating lobes, the angle is 28.8 degrees, and the amplitude is 19.5909dB. In the DBF directional diagram after filtering the theoretical signal of the real target, the magnitudes of the point b ₂, the point a ₂ and the point c ₂ corresponding to-15.8 degrees, 6 degrees and 28.8 degrees are all lower, and are not the maximum magnitude point in the directional diagram, and the angle value of the point a ₂ is 6 degrees which belongs to the angle value corresponding to the minimum magnitude point in the DBF directional diagram, which indicates that the point a1 corresponding to 6 degrees belongs to the real target point, -b ₁ corresponding to 15.8 degrees is the false peak value point which is caused by grating lobes and is positioned on the left side of the real target, and the point c ₁ corresponding to 28.8 degrees is the false peak value point which is caused by grating lobes and is positioned on the right side of the real target.

Fig. 4 is a schematic diagram of another first DBF pattern and a DBF pattern after filtering out a theoretical signal of a false peak target located on the left side of a real target according to an embodiment of the present application. As shown in fig. 4, 11 is a first DBF pattern, and 31 is a DBF pattern after filtering out the theoretical signal of the false peak target located at the left side of the real target. In the figure, the point d ₁ is a point corresponding to a real target, the angle is 6 degrees, and the amplitude is 21.9789dB; the point e ₁ is a false peak value point which belongs to the grating lobe and is positioned on the left side of a real target, the angle is 15.8 degrees below zero, and the amplitude is 19.5943dB; the point f ₁ is a false peak point which belongs to the grating lobe and is positioned on the right side of a real target, the angle is 28.8 degrees, and the amplitude is 19.5909dB. In the 31 of the DBF pattern after filtering the theoretical signal of the false peak value target positioned at the left side of the real target, the angle of the f ₂ point is 28.8 degrees, the amplitude is 15.6778dB, the amplitudes of the d ₂ point and the f ₂ point corresponding to the 6 degrees and the 28.8 degrees are still peaks, and the e ₁ point can be proved to be the false peak value point positioned at the left side of the real target caused by grating lobes by combining the DBF pattern results before and after filtering the a ₁ point of the real target in the figure 3.

Fig. 5 is a schematic diagram of another first DBF pattern and a DBF pattern after filtering out a theoretical signal of a false peak target located on the right side of a real target according to an embodiment of the present application. As shown in fig. 5, 11 is a first DBF pattern, which is identical to the first DBF pattern in fig. 4; 32 is the DBF pattern after filtering out the theoretical signal of the false peak target located to the right of the real target. In the 32 of the DBF directional diagram after filtering the theoretical signal of the false peak value target positioned on the right side of the real target, the angle of the e ₂ point is 15.8 degrees below zero, and the amplitude is 15.6836dB; the magnitudes of the e ₂ point and the d ₂ point corresponding to 15.8 degrees and 6 degrees are still peaks, and the DBF direction diagram results before and after filtering out the a ₁ point of the real target in combination with the figure 3 can show that the f ₁ point is a false peak value point which belongs to the grating lobe and is positioned on the right side of the real target.

Fig. 6 shows a schematic structural diagram of an object recognition device according to an embodiment of the present application. As shown in fig. 6, the apparatus 200 includes: a first acquisition module 201, a first determination module 202, a second acquisition module 203, a filtering module 204, a second determination module 205, and a third determination module 206. The first obtaining module 201 is configured to obtain a first digital beam forming DBF pattern of an object to be identified, where the first DBF pattern is obtained by performing DBF goniometry on a radar target echo signal. The second module 202 is configured to determine a plurality of first amplitude maxima points according to all amplitude points of the first DBF pattern, where each first amplitude maxima point is an amplitude point with an amplitude greater than adjacent left and right amplitude values in all amplitude points of the first DBF pattern. The second obtaining module 203 is configured to obtain i angle values corresponding to the first i first amplitude maximum points according to a sequence from the first amplitude maximum point to the second amplitude maximum point, where i is a positive integer, and i is greater than 1. The filtering module 204 is configured to select a target corresponding to an angle value a _k,A_k from i angle values as a current target to be identified, and filter a signal corresponding to a _k from the radar target echo signal to obtain a residual signal component, where k is a positive integer, and k is less than or equal to i. The second determining module 205 is configured to determine a second DBF pattern corresponding to the residual signal component. The third determining module 206 is configured to determine, in the second DBF pattern, if a _k belongs to an angle value corresponding to a magnitude minimum point, and the remaining i-1 angle values except a _k in the i angle values are not the angle values corresponding to the largest i-1 second magnitude maximum points, where the magnitude minimum point is a magnitude point in the second DBF pattern where the magnitude value is smaller than the adjacent left and right magnitudes, and the second magnitude maximum point is a magnitude point in the second DBF pattern where the magnitude value is greater than the adjacent left and right magnitudes.

The target recognition device provided in this embodiment is configured to execute the technical scheme of the target recognition method in the foregoing method embodiment, and its implementation principle and technical effects are similar, and are not described herein again.

It should be noted that, the object recognition device provided in this embodiment further includes other modules for executing the steps of the above object recognition method embodiment, which are not described herein in detail.

Fig. 7 is a schematic structural diagram of an object recognition device according to an embodiment of the present application, and the specific embodiment of the present application is not limited to the specific implementation of the object recognition device.

As shown in fig. 7, the object recognition apparatus may include: a processor (processor) 302, a communication interface (Communications Interface) 304, a memory (memory) 306, and a communication bus 308.

Wherein: processor 302, communication interface 304, and memory 306 perform communication with each other via communication bus 308. A communication interface 304 for communicating with network elements of other devices, such as clients or other servers. The processor 302 is configured to execute the program 310, and may specifically perform the relevant steps in the above-described embodiment of the target recognition method.

In particular, program 310 may include program code comprising computer-executable instructions.

The processor 302 may be a central processing unit CPU, or an Application-specific integrated Circuit ASIC (Application SPECIFIC INTEGRATED Circuit), or one or more integrated circuits configured to implement embodiments of the present application. The one or more processors comprised by the object recognition device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.

Memory 306 for storing program 310. Memory 306 may comprise high-speed RAM memory or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

Embodiments of the present application provide a computer-readable storage medium storing executable instructions that, when executed on an object recognition device, cause the object recognition device to perform the object recognition method in any of the method embodiments described above.

Embodiments of the present application provide a computer program that is callable by a processor to cause an object recognition device to perform the object recognition method in any of the method embodiments described above.

An embodiment of the present application provides a computer program product comprising a computer program stored on a computer readable storage medium, the computer program comprising program instructions which, when run on a computer, cause the computer to perform the object recognition method of any of the method embodiments described above.

Claims

1. A method of target identification, the method comprising:

acquiring a first Digital Beam Forming (DBF) directional diagram of a target to be identified, wherein the first DBF directional diagram is obtained by DBF angle measurement of radar target echo signals;

determining a plurality of first amplitude maximum points according to all amplitude points of the first DBF directional diagram, wherein each first amplitude maximum point is an amplitude point with amplitude values larger than adjacent left and right amplitude values in all amplitude points of the first DBF directional diagram;

According to the sequence from large to small of the first amplitude maximum points in the plurality of first amplitude maximum points, i angle values corresponding to the first i first amplitude maximum points are obtained, wherein the first i first amplitude maximum points are points corresponding to the target to be identified, i is a positive integer, and i is larger than 1;

Selecting an angle value A _k from the i angle values, wherein the target corresponding to the A _k is a current target to be identified, and filtering a signal corresponding to the A _k from the radar target echo signal to obtain a residual signal component, wherein k is a positive integer, and k is less than or equal to i;

determining a second DBF pattern corresponding to the residual signal component;

In the second DBF pattern, if the a _k belongs to an angle value corresponding to an amplitude minimum value point, and the remaining i-1 angle values except the a _k in the i angle values are not angle values corresponding to the largest i-1 second amplitude maximum value points, determining that the current target to be identified is a real target, where the amplitude minimum value point is an amplitude point in the second DBF pattern, where the amplitude value is smaller than adjacent left and right amplitude values, and the second amplitude maximum value point is an amplitude point in the second DBF pattern, where the amplitude value is greater than adjacent left and right amplitude values.

2. The method according to claim 1, wherein the radar target echo signal is a signal reflected by an object in a detection area after the radar array emits electromagnetic waves to the detection area, the filtering the signal corresponding to a _k from the radar target echo signal to obtain a residual signal component includes:

determining a theoretical signal Y _k of the current target to be identified under the detection of the radar array according to the A _k;

Determining a signal component estimated value S _k of the Y _k in the first DBF directional diagram according to the value corresponding to the A _k in the first DBF directional diagram and the Y _k, wherein the value is a complex-form value;

And filtering the S _k from the radar target echo signal to obtain the residual signal component.

3. The method according to claim 1, wherein if the a _k belongs to an angle value corresponding to a minimum value point of amplitude, and the remaining i-1 angle values except the a _k are not angle values corresponding to the maximum i-1 second maximum value points of amplitude, determining the current target to be identified as a real target includes:

Determining a plurality of second amplitude maximum points and a plurality of amplitude minimum points according to all the amplitudes of the second DBF directional diagram;

Sequencing the second amplitude maximum points from large to small to obtain first i-1 second amplitude maximum points;

If the a _k belongs to an angle value corresponding to one amplitude minimum point of the amplitude minimum points, and the remaining i-1 angle values except the a _k in the i angle values do not belong to an angle value corresponding to the first i-1 second amplitude maximum points, determining that the current target to be identified is the real target.

4. The method of claim 1, wherein i is 3.

5. The method according to claim 1, wherein the method further comprises:

In the second DBF pattern, if the a _k does not belong to an angle value corresponding to a minimum value point of amplitude, or if the remaining i-1 angle values other than the a _k are not the angle values corresponding to the maximum i-1 second maximum value points of amplitude, selecting an unselected angle value a _k from the i angle values;

And turning to the step of filtering the signal corresponding to the A _k from the radar target echo signal to obtain a residual signal component.

6. The method of claim 1, wherein prior to the acquiring the first DBF pattern of the object to be identified, the method further comprises:

Acquiring the radar target echo signal;

Performing Fourier transform of distance dimension and speed dimension on the radar target echo signal to obtain distance-Doppler RD spectrum data;

Obtaining a target array original signal S _origin of a target distance and a target speed in the RD spectrum data;

The obtaining the first DBF pattern of the object to be identified includes:

and performing DBF angle measurement on the S _origin to obtain the first DBF directional diagram.

7. The method according to claim 2, wherein said determining a theoretical signal Y _k of the current object to be identified under detection by the radar array according to the a _k comprises:

Calculating an index value according to a first preset parameter, the A _k, the radar array arrangement position, the pitching angle value of the real target and the wavelength of the radar target echo signal;

and calculating the theoretical signal Y _k by taking a preset numerical value as a base number and the index value as an index.

8. The method of claim 2, wherein the determining the signal component estimate S _k of the Y _k in the first DBF pattern based on the value corresponding to the a _k in the first DBF pattern and the Y _k comprises:

The step S _k) is calculated by the formula S _k＝Y_k*S_map(A_k), wherein S _map(A_k) is the numerical value.

9. An object recognition apparatus, characterized by comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;

The memory is configured to store executable instructions that cause the processor to perform the operations of the object recognition method of any one of claims 1-7.

10. A computer readable storage medium, characterized in that the storage medium has stored therein executable instructions which, when run, perform the object recognition method according to any one of claims 1-7.