Data processing method, apparatus, electronic device, computer storage medium

Abstract

An embodiment of the present application provides a data processing method, apparatus, electronic device, and computer storage medium. The data processing method includes: periodically transmitting a detection signal through a radar carried by a mobile device, and acquiring an echo signal generated by a reflection point of a target reflecting the detection signal when the detection signal irradiates the corresponding target; determining a frame of range-Doppler spectrum according to the echo signals received within one detection period; determining a stationary clutter clustering cluster composed of reflection points that are stationary relative to the ground according to the distance from the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative speed of the reflection points relative to the radar; and determining the traveling speed of the mobile device relative to the ground according to the relative speed of the stationary reflection points included in the stationary clutter clustering cluster. The traveling speed estimated by this method is more accurate.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of computer technologies, and in particular, to a data processing method, apparatus, electronic device, and computer storage medium. Background Art

[0002] Currently, mobile devices generally carry sensors (such as cameras, positioning modules, radars, etc.). Taking an intelligent driving vehicle as an example, the vehicle can detect its own driving speed and other moving objects (pedestrians or vehicles), obstacles, etc. around it to assist or replace the driver in making correct driving decisions and improve driving safety.

[0003] In the prior art, the detection of the driving speed of a mobile device requires multiple sensors to work together to achieve a relatively accurate measurement of the driving speed. However, the types and numbers of sensors carried by different mobile devices are different. Therefore, how to accurately measure the driving speed of a mobile device with fewer sensors is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a data processing solution to at least partially solve the above problems.

[0005] According to a first aspect of embodiments of the present application, there is provided a data processing method, including: periodically transmitting a detection signal through a radar carried by a mobile device, and obtaining an echo signal generated by a reflection point of a target reflecting the detection signal when the detection signal irradiates the corresponding target; determining a frame of range-Doppler spectrum according to the echo signals received within one detection period; determining a stationary clutter clustering cluster composed of reflection points that are stationary relative to the ground according to the distance from the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative speed of the reflection points relative to the radar; and determining the driving speed of the mobile device relative to the ground according to the relative speed of the stationary reflection points included in the stationary clutter clustering cluster.

[0006] According to a second aspect of embodiments of the present application, there is provided a data processing apparatus, including: an obtaining module, configured to periodically transmit a detection signal through a radar carried by a mobile device, and obtain an echo signal generated by a reflection point of a target reflecting the detection signal when the detection signal irradiates the corresponding target; a first determination module, configured to determine a frame of range-Doppler spectrum according to the echo signals received within one detection period; a second determination module, configured to determine a stationary clutter clustering cluster composed of reflection points that are stationary relative to the ground according to the distance from the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative speed of the reflection points relative to the radar; and a third determination module, configured to determine the driving speed of the mobile device relative to the ground according to the relative speed of the stationary reflection points included in the stationary clutter clustering cluster.

[0007] According to a third aspect of the embodiments of the present application, an electronic device is provided, including: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus; the memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the data processing method described in the first aspect.

[0008] According to a fourth aspect of the embodiments of the present application, a computer storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, it implements the data processing method described in the first aspect.

[0009] According to the data processing solution provided by the embodiments of the present application, when estimating the traveling speed of a mobile device, there is no need to additionally rely on other sensors. An echo signal of a detection signal emitted by a radar is used to obtain a frame of range-Doppler spectrum, and a stationary reflection point that is stationary relative to the ground is determined from multiple reflection points based on the range-Doppler spectrum. The speed of the radar relative to the ground is estimated based on the relative speed between the stationary reflection point and the radar indicated on the range-Doppler spectrum as the traveling speed of the mobile device carrying the radar. It not only realizes accurate estimation of the traveling speed, but also can be achieved with only one radar, which helps to reduce costs. Description of the Drawings

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0011] Figure 1a It is a flowchart of the steps of a data processing method according to Embodiment 1 of the present application;

[0012] Figure 1b It is Figure 1a a schematic diagram of a scenario example in the illustrated embodiment;

[0013] Figure 1c It is Figure 1a the range-Doppler spectrum in the illustrated embodiment;

[0014] Figure 2 It is a flowchart of the steps of a data processing method according to Embodiment 2 of the present application;

[0015] Figure 3a It is a flowchart of the steps of a data processing method according to Embodiment 3 of the present application;

[0016] Figure 3b A Figure 3a schematic diagram of a reference speed dimension in the illustrated embodiment;

[0017] Figure 3c A Figure 3a schematic diagram showing the change of driving speed before and after filtering over time in the illustrated embodiment;

[0018] Figure 3d A Figure 3a schematic diagram showing the change of acceleration over time in the illustrated embodiment;

[0019] Figure 4 A structural block diagram of a data processing device according to Embodiment 4 of the present application;

[0020] Figure 5 A structural schematic diagram of an electronic device according to Embodiment 5 of the present application. Specific embodiments

[0021] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art shall fall within the scope of protection of the embodiments of the present application.

[0022] The following further illustrates the specific implementation of the embodiments of the present application in conjunction with the accompanying drawings of the embodiments of the present application.

[0023] Embodiment 1

[0024] Referring to Figure 1a , a step - flow schematic diagram of the data processing method according to Embodiment 1 of the present application is shown, and the method includes:

[0025] Step S102: Periodically transmit detection signals through a radar carried by a mobile device, and obtain echo signals generated by the reflection points of the target reflecting the detection signals when the detection signals irradiate the corresponding target.

[0026] The mobile device in the present application can be a movable device equipped with sensors (such as cameras, positioning modules, or radars, etc.), including automobiles, robots, etc. Taking the mobile device as an intelligent driving vehicle as an example, at least one radar is carried on the intelligent driving vehicle. The radar can periodically emit electromagnetic waves as detection signals. When there are targets (such as pedestrians, vehicles, buildings, and trees, etc.) within the detection range of the radar, the targets will reflect the detection signals to form echo signals. The position where the target reflects the detection signal can be called the reflection point.

[0027] According to different requirements, the electromagnetic wave signals emitted by the radar can be modulated into different frequencies, signal bandwidths, and frame periods. For example, in one example, in order to meet the requirements of the detection range and minimize the data volume as much as possible to reduce the computational load, for a 77 GHz millimeter-wave radar, the signal bandwidth of the detection signal can be modulated to 0.75 GHz, the sampling frequency to 10 MHz, the frame period to 100 ms, and the number of accumulation periods to 256. Among them, the frame period of 100 ms means that the time difference between two adjacent detection periods of the radar is 100 ms, and the number of accumulation periods of 256 means that 256 detection signals are emitted within one detection period to obtain the corresponding echo signals.

[0028] Based on the above detection signals, the corresponding range resolution can be calculated as 0.2 m, the maximum detection range as 50 m, the velocity resolution as 0.23 m / s, and the maximum detection velocity as 14.76 m / s.

[0029] Of course, in other examples, based on the principle that the range resolution and velocity resolution are positively correlated with the data volume, and the signal bandwidth is negatively correlated with the detection range, the detection signals can be appropriately modulated according to the required detection range and the storable data volume, and appropriate detection signals can be selected for detection.

[0030] In one detection period, at the beginning stage of it, the radar emits multiple detection signals outward. If there are targets (such as walls or trees, etc.) within the detection range of the radar, the reflection points on these walls or trees will reflect the detection signals to form echo signals. There can be one or more reflection points on a target, and this embodiment does not limit this.

[0031] Step S104: Determine a frame of range-Doppler spectrum according to the echo signals received within one detection period.

[0032] Taking the example of emitting 256 detection signals in one detection period as mentioned above, within this detection period, N echo signals reflected by different targets are received, N is greater than or equal to 1, and the maximum value of N is 256.

[0033] By processing the received echo signals, a frame of range-Doppler spectrum (denoted as RD spectrum) can be obtained. For example, determine the beat signal according to the received N echo signals and the corresponding detection signals. Obtain the range-Doppler spectrum by processing the beat signal (such as fast Fourier transform). Since the number of beat signals is less, the computational load and storage requirements are lower.

[0034] The range-Doppler spectrum includes a range dimension (i.e., the abscissa), a velocity dimension (i.e., the ordinate), and multiple points. These points correspond to the reflection points in the target and are used to indicate the distance and relative velocity of the reflection points relative to the radar. In addition, the color of the points in the range-Doppler spectrum is used to indicate the amplitude of the corresponding echo signal.

[0035] In this embodiment, since different targets have different distances, positions, and speeds relative to the radar on the mobile device, the resulting echo signals are also different. Based on the time delay of the echo signal relative to the detection signal, the distance and azimuth angle of the target corresponding to the echo signal relative to the radar can be determined. Based on the amplitude of the echo signal and the Doppler frequency between the echo signal and the detection signal, the relative speed of the target relative to the radar can be determined.

[0036] The Doppler frequency can be understood as the frequency difference between the echo signal and the detection signal. For example, when the radar emits a detection signal with a fixed frequency (such as a pulse wave) for air scanning, if a target is encountered, the frequency of the echo signal reflected by the target will have a frequency difference from the frequency of the emitted detection signal, and this frequency difference is caused by the relative motion between the target and the radar.

[0037] According to the magnitude of the Doppler frequency, the radial motion speed of the target relative to the radar can be determined. According to the time difference between the emission time of the detection signal and the reception time of the echo signal, the distance of the target relative to the radar can be determined. By using a frequency filtering method to detect the Doppler frequency spectrum line of the target and filtering out the spectrum lines of interference clutter, the radar can distinguish the echo signal corresponding to the target from strong clutter, so that the pulse Doppler radar has stronger anti-clutter interference ability than ordinary radar and can detect moving targets hidden in the environment.

[0038] Based on the foregoing principle, the distance and relative speed of the corresponding target relative to the radar can be determined based on the frequency and phase of the beat signal, etc.

[0039] Generally, the targets within the detection range of the radar include noise targets and actual targets. Noise targets such as air, rain, snow, sea waves, and clutter, etc. The reflection points corresponding to these noise targets are denoted as noise reflection points. Due to the insufficient reflection ability of the corresponding noise targets, the amplitude of the reflected echo signal is relatively low, and the color in the range-Doppler spectrum is darker.

[0040] Actual targets such as trees, vehicles, pedestrians, and fixed objects on the ground, etc. The reflection points corresponding to these actual targets are denoted as candidate reflection points. Due to the better reflection ability of the corresponding actual targets, the amplitude of the reflected echo signal is relatively high, and the color in the range-Doppler is brighter.

[0041] Step S106: Determine a stationary clutter clustering cluster composed of reflection points that are stationary relative to the ground based on the distance of the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative speed of the reflection points relative to the radar.

[0042] When estimating the traveling speed of a mobile device relative to the ground, a fixed object on the ground and located directly in front of the radar can be used as a reference, and the traveling speed of the radar relative to the fixed object directly in front can be estimated as the traveling speed of the mobile device relative to the ground.

[0043] Since there may be moving targets such as vehicles and pedestrians in the target, to prevent the adverse effects of these targets on speed estimation, a stationary clutter clustering cluster composed of reflection points stationary relative to the ground can be determined based on the distances from the reflection points indicated by the range-Doppler spectrum to the radar and their relative speeds relative to the radar. Since the reflection points stationary relative to the ground (denoted as stationary reflection points) come from stationary fixed objects, determining the stationary clutter clustering cluster also determines the stationary fixed objects.

[0044] A feasible way to determine the stationary clutter clustering cluster can be: screening out reflection points with the same relative speed and continuous distances from the radar in the range-Doppler spectrum to form the stationary clutter clustering cluster. The principle of doing this is that: the detection period corresponding to one frame of the range-Doppler spectrum is very short (for example, 100 ms in this embodiment), so it can be considered that the moving speed of the mobile device is basically uniform within the time range corresponding to one frame of the range-Doppler spectrum. That is to say, within one frame of the range-Doppler spectrum, the relative speed between the stationary reflection points and the radar remains unchanged or changes very little. Also, since the mobile device moves continuously, the distances from the stationary reflection points to the radar change continuously.

[0045] Step S108: Determine the traveling speed of the mobile device relative to the ground according to the relative speeds of the stationary reflection points included in the stationary clutter clustering cluster.

[0046] In a feasible way, to ensure the accuracy of the estimated traveling speed, the mean value can be calculated based on the relative speeds of the stationary reflection points as the estimated traveling speed. Of course, in other feasible ways, other methods can be used to determine the traveling speed, and this embodiment does not limit this.

[0047] The following combines a specific usage scenario to illustrate the process of estimating the traveling speed as follows: As Figure 1b shown, in this usage scenario, taking the case where there is a radar installed on a vehicle as an example, the radar can be a 77 GHz millimeter-wave radar, the radar has an antenna with 1 transmit and 4 receive channels, and the mobile device equipped with the radar is an intelligent driving vehicle, which is installed on the front bumper of the vehicle.

[0048] During the vehicle's driving process, the radar periodically emits detection signals. For example, if the frame period is 100 ms and the number of accumulation periods is 256, 256 electromagnetic wave pulses need to be emitted at the start stage of 100 ms to form 256 detection signals. The detection signal is a linear frequency modulation sawtooth wave signal, with a signal bandwidth of 0.75 GHz and a sampling rate of 10 MHz.

[0049] After the radar receives the echo signal corresponding to the detection signal, based on the detection signal and the echo signal, the corresponding beat signal is determined, which can reduce the data storage amount without losing information on relative speed, relative distance, and azimuth angle, thereby reducing the load on the storage space.

[0050] By processing the beat signal (such as performing a fast Fourier transform), a frame of range-Doppler spectrum corresponding to this detection period is obtained, as Figure 1c shown. The abscissa of this range-Doppler spectrum is the relative distance, and the ordinate is the relative speed, Figure 1c where the white dots correspond to the reflection points of the actual targets, and the gray dots correspond to the reflection points of the noise targets.

[0051] After obtaining a frame of range-Doppler spectrum, based on the distance from the reflection point indicated in the range-Doppler spectrum to the radar (denoted as the relative distance), the relative speed between the reflection point and the radar, and the amplitude of the echo signal, the stationary reflection points located directly in front of the radar and stationary relative to the ground are selected from these reflection points. Then, based on the relative speed of the stationary reflection points, the speed of the radar relative to the stationary reflection points is determined, and further, the driving speed of the vehicle relative to the ground is estimated.

[0052] Through this embodiment, when estimating the driving speed of a mobile device, there is no need to additionally rely on other sensors. A frame of range-Doppler spectrum is obtained through the echo signal of the detection signal emitted by the radar, and the stationary reflection points stationary relative to the ground are determined from multiple reflection points based on the range-Doppler spectrum. The speed of the radar relative to the ground is estimated based on the relative speed between the stationary reflection points and the radar indicated on the range-Doppler spectrum as the driving speed of the mobile device equipped with this radar. It not only achieves accurate estimation of the driving speed but also can be realized with only one radar, which helps to reduce costs. Further, if the mobile device is an autonomous mobile vehicle, robot, etc., it can remind the surrounding pedestrians to pay attention in the form of sound, light, body text display, or projection text display based on the driving speed.

[0053] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as tablet computers, mobile phones, etc.), and PC machines, etc.

[0054] Embodiment 2

[0055] Refer toFigure 2 , which shows the schematic flow chart of the data processing method according to the second embodiment of the present application.

[0056] In this embodiment, the data processing method can be executed by a chip with data processing capabilities provided on the mobile device, so as to estimate the driving speed. Or the data processing method can also be executed by other devices with data processing capabilities that are data-connected to the mobile device.

[0057] The data processing method includes the aforementioned steps S102 to S108. Among them, step S104 may include the following sub-steps:

[0058] Sub-step S1041: Determine the beat signal according to the echo signal and the detection signal.

[0059] After receiving the echo signal through the receiving antenna, the corresponding beat signal can be determined according to the echo signal and the detection signal.

[0060] For example, the high-frequency linear frequency modulation sawtooth wave signal generated by the transmitter of the radar is directly mixed with the echo signal reflected by the reflection point of the received target. Since there is a time difference between the detection signal and the echo signal, the frequency and phase of the beat signal obtained after mixing are related to the distance from the target to the radar, the relative speed with the radar, and the azimuth angle. Subsequently, by processing the beat signal, the distance and relative speed from the reflection point on the target to the radar can be obtained.

[0061] Sub-step S1042: Perform fast Fourier transform (i.e., FFT) on the beat signal in both the fast time dimension and the slow time dimension to obtain the range-Doppler spectrum.

[0062] In an example, the detection period of the radar can be called the frame period. For example, if the driving speed of the mobile device is detected every 100 ms, then a 100 ms can be considered as a detection period. During this detection period, the radar emits 256 detection signals. For the echo signal of each detection signal, it is stored in one row, and the detection signals emitted at different times are stored in different rows. In this way, one echo signal (that is, one row) is the fast time dimension, and different echo signals (that is, multiple rows) are the slow time dimension.

[0063] In this embodiment, since the number of detection signals (which can also be called the number of periods) is 256, the range-Doppler spectrum obtained after performing fast Fourier transform on the fast time dimension and the slow time dimension respectively is a 128x512-dimensional range-Doppler spectrum.

[0064] In this way, a frame of range-Doppler spectrum can be obtained quickly, and the traveling speed of the mobile device can be estimated based on the range-Doppler spectrum in the subsequent steps. When estimating the traveling speed in this way, there is no need to additionally rely on other sensors, and the traveling speed of the mobile device can be accurately estimated by processing the echo signal of the detection signal transmitted by the radar. Moreover, the computational complexity of obtaining the range-Doppler spectrum is small and the computational speed is fast.

[0065] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as tablet computers, mobile phones, etc.), and PC machines, etc.

[0066] Embodiment III

[0067] Referring to Figure 3a , a schematic flowchart of the steps of the data processing method of the embodiment of the present application is shown.

[0068] In this embodiment, the data processing method includes the aforementioned steps S102 to S108. Among them, step S104 can be implemented in the manner of Embodiment I or II. Step S106 includes the following sub-steps:

[0069] Sub-step S1061: Screen out at least one candidate reflection point other than the noise reflection points from the multiple reflection points indicated by the range-Doppler spectrum.

[0070] In the range-Doppler spectrum, the amplitudes of the echo signals of the reflection points corresponding to different points are different. For example, the amplitude of the echo signal generated by the reflection point of a noise target is usually much smaller than the amplitude of the echo signal generated by the reflection point of an actual target. Therefore, the candidate reflection points corresponding to the actual target can be screened out based on this.

[0071] When screening the reflection points, since the characteristics of noises such as rain, snow, sea waves, air, and clutter interference are different, a constant false alarm detection algorithm appropriate to the characteristics can be used to screen out the possible candidate reflection points for subsequent determination of the stationary reflection points from the candidate reflection points. Constant false alarm detection algorithms such as CA-CFAR, GO-CFAR, SO-CFAR, OS-CFAR, etc.

[0072] In the urban road environment of this embodiment, candidate reflection points are screened out in the range dimension and the speed dimension respectively through a specific constant false alarm detection algorithm (such as OS-CFAR). The obtained candidate reflection points can be represented as discrete or continuous points.

[0073] Specifically, sub-step S1061 includes the following process:

[0074] Process A1: Obtain a set protection threshold value and a reference threshold value according to the total number of speed dimensions of the range-Doppler spectrum.

[0075] Among them, the protection threshold value and the reference threshold value are used to determine the screening conditions corresponding to each candidate reflection point when determining the candidate reflection points.

[0076] Taking the 128-dimensional velocity dimension as an example, since the moving speeds of targets such as pedestrians, vehicles, and fixed objects are different, the velocity dimensions in which different types of targets are distributed in the range-Doppler spectrum may be quite different, while the moving speeds of the same type of targets may be relatively close, so they may be distributed in adjacent velocity dimensions or the same velocity dimension in the range-Doppler spectrum.

[0077] For example, among vehicles, pedestrians, and fixed objects, a vehicle may move at the same speed as the radar, so the vehicle is distributed in the velocity dimension with a relative velocity of 0 in the range-Doppler spectrum. The relative velocity between a fixed object and the radar is relatively large, and it is distributed in the velocity dimension of -5 and adjacent velocity dimensions.

[0078] Since stationary reflection points may be distributed in adjacent different velocity dimensions, in order to accurately screen out candidate reflection points and prevent missing candidate reflection points, a protection threshold value needs to be set. The reference threshold value is used to determine the velocity dimension used in screening. The protection threshold value and the reference threshold value can be determined according to specific circumstances. For example, the protection threshold value can be 2, and the reference threshold value can be 5.

[0079] Process A2: Determine at least one candidate reflection point according to the protection threshold value, the reference threshold value, and the amplitudes of the echo signals corresponding to the multiple reflection points indicated by the range-Doppler spectrum.

[0080] Next, in combination with Figure 3b , the implementation of Process A2 will be described exemplarily. For any reflection point indicated in the range-Doppler spectrum, it has a corresponding amplitude.

[0081] During screening, for each reflection point indicated by the range-Doppler spectrum, the following actions can be performed to determine whether it is a candidate reflection point.

[0082] Process A2 can be implemented as follows: For the multiple reflection points indicated by the range-Doppler spectrum, determine the excluded velocity dimensions according to the velocity dimension of the current reflection point and the protection threshold value; determine the velocity boundary of the current reflection point according to the excluded velocity dimensions; determine the reference velocity dimension of the current reflection point according to the velocity boundary and the reference threshold value; determine the screening condition of the current reflection point according to the amplitudes of the echo signals of the multiple reflection points indicated in the reference velocity dimension, and determine whether the current reflection point is a candidate reflection point according to the screening condition and the amplitude of the echo signal of the current reflection point.

[0083] Taking Figure 3bTaking point A shown in [figure] as an example, if the protection threshold is 2 and the reference threshold is 5, then taking point A as the current emission point, the velocity dimension where point A is located is denoted as velocity dimension A. According to the velocity dimension where point A is located and the protection threshold, determine the excluded velocity dimensions, that is, exclude the 2 adjacent velocity dimensions above velocity dimension A and the 2 adjacent velocity dimensions below it (a total of 4 velocity dimensions). As shown in Figure 3b the dashed line B1 shown in [figure] as the upper velocity boundary, Figure 3b the dashed line B2 shown in [figure] as the lower velocity boundary. If there are less than 2 adjacent velocity dimensions below velocity dimension A, then all the velocity dimensions below can be directly excluded. When there are less than 2 adjacent velocity dimensions above, then all the velocity dimensions above can be excluded.

[0084] According to the determined upper and lower velocity boundaries and the reference threshold, determine the reference velocity dimension corresponding to point A. For example, if the reference threshold is 5, then select 5 adjacent velocity dimensions upward from the upper velocity boundary as the upper reference velocity dimension ( Figure 3b the area shown as B1 to C1 in [figure]), and select 5 adjacent velocity dimensions downward from the lower velocity boundary as the lower reference velocity dimension ( Figure 3b the area shown as B2 to C2 in [figure]).

[0085] Determine the screening condition for the current reflection point according to the amplitudes of the echo signals of multiple reflection points indicated in the reference velocity dimension. For example, calculate the mean value of the amplitudes according to the amplitudes of the echo signals of each reflection point indicated in the reference velocity dimension, and use the mean value as the screening condition.

[0086] According to this screening condition, if the amplitude of point A is greater than or equal to the mean value, then determine point A as a candidate reflection point. Conversely, if the amplitude of point A is less than the mean value, then determine that point A is not a candidate reflection point.

[0087] In this way, each reflection point indicated by the range-Doppler spectrum can be traversed to screen out candidate reflection points. For different current reflection points, corresponding protection thresholds and reference thresholds can be determined.

[0088] Sub-step S1062: Determine the stationary clutter clustering clusters composed of stationary reflection points that are stationary relative to the earth from the at least one candidate reflection point according to the distance from the at least one candidate reflection point to the radar and the relative velocity of the at least one candidate reflection point relative to the radar.

[0089] Due to the characteristic that when a stationary reflection point that is stationary relative to the earth is in the direct front of the radar, its distance from the radar changes continuously but its relative velocity remains unchanged. That is to say, it forms a straight line in a certain velocity dimension in the range-Doppler spectrum. Therefore, when determining the stationary clutter clustering clusters composed of stationary reflection points, sub-step S1062 can be implemented through the following process:

[0090] Process B1: Determine a distance cell threshold based on the distance of the at least one candidate reflection point relative to the radar and the relative velocity.

[0091] For example, for a certain velocity dimension, based on the distance of the candidate reflection points to the radar, determine the candidate reflection points that are continuous in the distance dimension among the candidate reflection points in this velocity dimension, and then determine the region where these continuously adjacent candidate reflection points in the distance dimension are located as the stationary clutter region.

[0092] To avoid errors in velocity estimation caused by the azimuth angle between the candidate reflection points and the radar, for each candidate reflection point in the stationary clutter region, determine a distance threshold to exclude the Doppler-azimuth coupling part in the stationary clutter region.

[0093] For example, in the part of the candidate reflection points with Doppler-azimuth coupling, they are not a straight line in a certain velocity dimension, but there are also candidate reflection points in the distance dimension (see the Doppler-azimuth coupling part in Figure 1c ). The relative velocity of the candidate reflection points in this coupling part is affected by the azimuth angle, so the radar's traveling speed cannot be accurately estimated using them, and they need to be excluded. Based on this, the distance threshold can be determined according to the distance dimension corresponding to the Doppler method coupling part.

[0094] In Figure 1c the shown range-Doppler spectrum, the distance threshold can be set to 257 to 512. The detected targets in this region form a straight line in the velocity dimension, indicating that these candidate reflection points are in front of the radar, and the echo signals they reflect will not affect the frequency, amplitude, time delay, etc. due to the position deviation relative to the radar, thus avoiding errors and ensuring the accuracy of velocity estimation.

[0095] Process B2: Determine the pre-screened reflection points that meet the distance cell threshold from the at least one candidate reflection point.

[0096] According to the distance threshold, the candidate reflection points within this distance threshold among the candidate reflection points can be used as the pre-screened reflection points. There may be stationary reflection points or moving pre-screened reflection points (such as reflection points on a vehicle, etc.) among the pre-screened reflection points.

[0097] Process B3: Perform density clustering on the relative velocities of the pre-screened reflection points, and determine the clustering result with the longest span as the stationary clutter clustering cluster, and the pre-screened reflection points included in the stationary clutter clustering cluster are used as stationary reflection points.

[0098] In order to determine stationary reflection points from the initially screened reflection points, density clustering is performed on the initially screened reflection points. For example, the initially screened reflection points are clustered by the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) density clustering method. The DBSCAN density clustering can set the radius of the clustering parameter to 0.3 m / s and the minimum number to 3. By traversing all the initially screened reflection points, the initially screened reflection points are clustered to obtain at least one clustering cluster.

[0099] The density-based DBSCAN clustering algorithm can set reasonable threshold parameters according to the distance from the initially screened reflection points to the radar, and divide the initially screened reflection points that tend to cluster in different dimensions into the same clustering cluster. This clustering feature is very suitable for the relatively high-precision initially screened reflection points obtained by the millimeter-wave radar.

[0100] For the obtained clustering clusters, the clustering cluster with the longest span is determined as the stationary clutter clustering cluster.

[0101] For example, calculate the difference in the relative distances between any two initially screened reflection points in the clustering cluster, and determine the maximum difference in the clustering cluster. Select the clustering cluster with the maximum difference as the stationary clutter clustering cluster, and the initially screened reflection points in this stationary clutter clustering cluster are used as stationary reflection points.

[0102] In this way, stationary reflection points can be determined from the reflection points indicated by the range-Doppler spectrum, that is, the reflection points located directly in front of the radar and stationary relative to the ground. The echo signals reflected by these stationary reflection points will not generate errors due to the azimuth angle, etc., and can be used to accurately estimate the driving speed.

[0103] After obtaining the stationary reflection points, subsequent steps can be executed to estimate the driving speed of the mobile device according to the relative speed of the stationary reflection points, so as to obtain the accurate driving speed.

[0104] For example, in this embodiment, step S108 can be implemented as: using the centroid algorithm for the stationary clutter clustering cluster to determine the average speed of the stationary reflection points relative to the mobile device according to the relative speed of the stationary reflection points included in the stationary clutter clustering cluster, as the driving speed of the mobile device relative to the ground. In this way, the relative speed of the radar relative to the stationary reflection points can be accurately estimated.

[0105] Optionally, in this embodiment, in order to improve the accuracy of speed estimation, in addition to estimating the driving speed corresponding to the current detection period through the echo signal of the radar in each detection period, the driving speeds estimated in multiple detection periods can also be processed to obtain a more accurate driving speed. To this end, when the number of detection periods is more than two, the method further includes step S110: performing Kalman filtering processing based on the driving speeds of the mobile device obtained in a plurality of consecutive detection periods, and determining the driving speed of the mobile device relative to the ground according to the processing result.

[0106] For example, by using the measurement vectors of multiple consecutive frames, a state equation and an observation equation are established, and then Kalman filtering is used to improve the accuracy of driving speed estimation and obtain a more accurate driving speed.

[0107] In this embodiment, taking the example of obtaining 800 frames of range-Doppler spectra in 80 s, 1 frame of range-Doppler spectrum corresponds to a time of 100 ms. In the starting stage of 100 ms (such as the first 10 ms), 256 detection signals are continuously emitted, and according to the obtained echo signals, beat signals are determined, and the beat signals are processed to obtain the range-Doppler spectrum of the corresponding frame.

[0108] When the number of frames of the obtained range-Doppler spectra is greater than or equal to 2, a state equation and an observation equation can be constructed based on at least two adjacent frames of range-Doppler spectra, and Kalman filtering is used to process the state equation, thereby improving the accuracy of estimating the driving speed of the mobile device.

[0109] For example, the observation vector can be expressed as Z = [v]. Since the driving speed of the mobile device is estimated in this embodiment, the state vector is described by the speed. At the k-th moment, the state vector can be expressed as:

[0110] The corresponding observation equation at the (k + 1)-th moment can be expressed as: Z(k + 1) = HX(k + 1), where Z(k + 1) is the observation vector at the (k + 1)-th moment, H is the weight coefficient, which can be appropriately determined according to needs, for example, [1, 0]. X(k + 1) is the state vector at the (k + 1)-th moment.

[0111] Substituting the state vector and the observation vector into the observation equation gives:

[0112] The state equation can be expressed as: X(k + 1) = AX(k), where X(k + 1) is the state vector at the (k + 1)-th moment, A is the weight coefficient, which can be calculated according to needs, for example, T is the time interval between two adjacent frames, which is 0.1 s in this embodiment. X(k) is the state vector at the k-th moment. That is to say, the state equation can be expressed as:

[0113] After establishing the state equation and the observation equation, the Kalman filtering method can be adopted to calculate the Kalman filtering coefficients, determine the observed speed according to the observation equation at the (k + 1)-th moment, determine the predicted speed at the (k + 1)-th moment through the state equation at the (k + 1)-th moment, weight the observed speed using the Kalman filtering coefficients, and then sum it with the predicted speed to obtain the driving speed at the (k + 1)-th moment.

[0114] Through Kalman filtering, an optimal estimated speed under linear Gaussian white noise interference can be obtained. Using Kalman filtering estimation can improve the accuracy of the self-vehicle speed estimation. Figure 3c The schematic diagram showing the change of the driving speed before and after filtering with respect to time is shown. Figure 3d The schematic diagram showing the change of the acceleration is shown.

[0115] In this way, within one detection period, using the radar fixed at the front end of the mobile device, a detection signal is transmitted, and the obtained echo signal is processed in the range dimension and the Doppler dimension to obtain a range-Doppler spectrum. In a scenario with continuous stationary clutter, through constant false alarm rate (CFAR) detection and density clustering of the obtained range-Doppler spectrum, combining the CFAR detection result with the density clustering result, and setting a certain range threshold, stationary reflection points whose features conform to those of stationary clutter are extracted from the range-Doppler spectrum. Then, based on the relative speed of the stationary reflection points, the speed of the radar relative to the ground is estimated, which is the driving speed of the mobile device relative to the ground. To make the estimation more accurate, a state equation and an observation equation are constructed for the driving speeds estimated in multiple consecutive detection periods, and then filtering is performed to improve the estimation accuracy of the driving speed.

[0116] This process uses a millimeter-wave radar to measure the driving speed of the self-vehicle, and uses the Kalman filtering method to improve the estimation accuracy. Without relying on other sensors, the speed of the radar relative to the ground can be obtained, which is convenient for the radar to subsequently output the speed compensation of the self-vehicle and improve the accuracy of self-vehicle information detection. While ensuring the basic functions of radar ranging, speed measurement, and angle measurement, the stationary clutter information in the range-Doppler spectrum is fully utilized to estimate the self-vehicle speed. On the basis of ensuring the speed estimation of each frame of the range-Doppler spectrum, the Kalman filtering algorithm is introduced to use multi-frame information to improve the self-vehicle speed estimation accuracy.

[0117] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as tablet computers, mobile phones, etc.), and PC machines, etc.

[0118] Embodiment 4

[0119] Refer to Figure 4 , which shows the structural block diagram of the data processing device according to Embodiment 4 of the present application.

[0120] In this embodiment, the data processing device includes:

[0121] An acquisition module 402, configured to periodically transmit detection signals through a radar carried by a mobile device, and acquire echo signals generated by reflection points of a target reflecting the detection signals when the detection signals irradiate the corresponding target;

[0122] A first determination module 404, configured to determine a frame of range-Doppler spectrum according to the echo signals received within one detection period;

[0123] A second determination module 406, configured to determine a stationary clutter clustering cluster composed of reflection points stationary relative to the ground according to the distance from the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative velocity of the reflection points relative to the radar;

[0124] A third determination module 408, configured to determine the traveling speed of the mobile device relative to the ground according to the relative velocities of the stationary reflection points included in the stationary clutter clustering cluster.

[0125] Optionally, the second determination module 406 includes:

[0126] A screening module 4061, configured to screen out at least one candidate reflection point other than noise reflection points from the multiple reflection points indicated by the range-Doppler spectrum;

[0127] A fourth determination module 4062, configured to determine a stationary clutter clustering cluster composed of stationary reflection points stationary relative to the ground from the at least one candidate reflection point according to the distance from the at least one candidate reflection point to the radar and the relative velocity of the at least one candidate reflection point relative to the radar.

[0128] Optionally, the screening module 4061 is configured to obtain a set protection threshold value and a reference threshold value according to the total number of velocity dimensions of the range-Doppler spectrum, where the protection threshold value and the reference threshold value are used to determine screening conditions corresponding to each candidate reflection point when determining the candidate reflection points; and determine at least one candidate reflection point according to the protection threshold value, the reference threshold value, and the amplitudes of the echo signals corresponding to the multiple reflection points indicated by the range-Doppler spectrum.

[0129] Optionally, when determining at least one candidate reflection point based on the amplitudes of the echo signals corresponding to multiple reflection points indicated by the protection threshold value, the reference threshold value, and the range-Doppler spectrum, the screening module 4061 determines, for the multiple reflection points indicated by the range-Doppler spectrum, the velocity dimension to be excluded according to the velocity dimension of the current reflection point and the protection threshold value; determines the velocity boundary of the current reflection point according to the excluded velocity dimension; determines the reference velocity dimension of the current reflection point according to the velocity boundary and the reference threshold value; determines the screening condition of the current reflection point according to the amplitudes of the echo signals of the multiple reflection points indicated in the reference velocity dimension, and determines whether the current reflection point is a candidate reflection point according to the screening condition and the amplitude of the echo signal of the current reflection point.

[0130] Optionally, the fourth determination module 4062 is configured to determine a range cell threshold according to the distance of the at least one candidate reflection point relative to the radar and the relative velocity; determine, from the at least one candidate reflection point, the pre-screened reflection points that meet the range cell threshold; perform density clustering on the relative velocities of the pre-screened reflection points, and determine the clustering result with the longest span as the stationary clutter clustering cluster, and the pre-screened reflection points included in the stationary clutter clustering cluster are used as stationary reflection points.

[0131] Optionally, the third determination module 408 uses the centroid algorithm for the stationary clutter clustering cluster to determine the average velocity of the stationary reflection points relative to the mobile device according to the relative velocities of the stationary reflection points included in the stationary clutter clustering cluster, as the driving velocity of the mobile device relative to the ground.

[0132] Optionally, when the number of detection periods is two or more, the apparatus further includes: a filtering module 410, configured to perform Kalman filtering processing according to the driving velocities of the mobile device obtained in a plurality of consecutive detection periods, and determine the driving velocity of the mobile device relative to the ground according to the processing result.

[0133] The data processing apparatus in this embodiment is used to implement the corresponding data processing methods in the foregoing multiple method embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here. In addition, the functions of each module in the data apparatus in this embodiment can be referred to the descriptions of the corresponding parts in the foregoing method embodiments, which will not be elaborated here either.

[0134] Embodiment 5

[0135] Refer to Figure 5 , which shows a schematic structural diagram of an electronic device according to Embodiment 5 of the present application. The specific implementation of the electronic device in the specific embodiments of the present application is not limited.

[0136] As Figure 5As shown in the figure, the electronic device may include: a processor 502, a communications interface 504, a memory 506, and a communication bus 508.

[0137] Among them:

[0138] The processor 502, the communications interface 504, and the memory 506 communicate with each other through the communication bus 508.

[0139] The communications interface 504 is used to communicate with other electronic devices or servers.

[0140] The processor 502 is used to execute the program 510, and specifically can execute the relevant steps in the above data processing method embodiments.

[0141] Specifically, the program 510 may include program code, and the program code includes computer operation instructions.

[0142] The processor 52 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application. One or more processors included in the intelligent device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0143] The memory 506 is used to store the program 510. The memory 506 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0144] The program 510 is specifically used to cause the processor 502 to execute the operations corresponding to any of the foregoing data processing methods.

[0145] For the specific implementation of each step in the program 510, reference may be made to the corresponding steps and descriptions in the corresponding units in the above data processing method embodiments, which will not be elaborated here. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, which will not be repeated here.

[0146] It should be noted that according to the needs of implementation, each component / step described in the embodiments of the present application may be split into more components / steps, or two or more components / steps or parts of the operations of the components / steps may be combined into new components / steps to achieve the purpose of the embodiments of the present application.

[0147] The method according to the embodiments of the present application can be implemented in hardware, firmware, or be implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or be implemented as computer code that is originally stored in a remote recording medium or a non-transitory machine-readable medium and downloaded through a network and will be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component (such as RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the data processing method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the data processing method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the data processing method shown herein.

[0148] Those of ordinary skill in the art can realize that the units and method steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments of the present application.

[0149] The above embodiments are only used to illustrate the embodiments of the present application, rather than to limit the embodiments of the present application. Those of ordinary skill in the relevant technical field can also make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application. The patent protection scope of the embodiments of the present application shall be defined by the claims.

Claims

1. A data processing method, comprising: Periodically transmitting a detection signal through a radar carried by a mobile device to obtain an echo signal generated by a reflection point of a target reflecting the detection signal when the detection signal irradiates the corresponding target; Determining a frame of range-Doppler spectrum according to the echo signals received within a detection period; Determining a stationary clutter clustering cluster composed of reflection points stationary relative to the earth according to the distance from the reflection point indicated by the frame of range-Doppler spectrum to the radar and the relative velocity of the reflection point relative to the radar, including: screening out at least one candidate reflection point other than noise reflection points from the multiple reflection points indicated by the range-Doppler spectrum; determining a range cell threshold according to the distance of the at least one candidate reflection point relative to the radar and the relative velocity; determining, from the at least one candidate reflection point, a pre-screened reflection point that satisfies the range cell threshold to exclude a part of Doppler-azimuth coupling in the stationary clutter region; performing density clustering on the relative velocities of the pre-screened reflection points, and determining the clustering result with the longest span as the stationary clutter clustering cluster, and using the pre-screened reflection points included in the stationary clutter clustering cluster as stationary reflection points; Determining the traveling speed of the mobile device relative to the earth according to the relative velocities of the stationary reflection points included in the stationary clutter clustering cluster.

2. The method according to claim 1, wherein The screening out at least one candidate reflection point other than noise reflection points from the reflection points indicated by the range-Doppler spectrum includes: Obtaining a set protection threshold value and a reference threshold value according to the total number of velocity dimensions of the range-Doppler spectrum, where the protection threshold value and the reference threshold value are used to determine the screening conditions corresponding to each candidate reflection point when determining the candidate reflection points; Determining at least one candidate reflection point according to the protection threshold value, the reference threshold value, and the amplitudes of the echo signals corresponding to the multiple reflection points indicated by the range-Doppler spectrum.

3. The method according to claim 2, wherein, The determining at least one candidate reflection point according to the protection threshold value, the reference threshold value, and the amplitudes of the echo signals corresponding to the multiple reflection points indicated by the range-Doppler spectrum includes: For the multiple reflection points indicated by the range-Doppler spectrum, determining the excluded velocity dimension according to the velocity dimension of the current reflection point and the protection threshold value; Determining the velocity boundary of the current reflection point according to the excluded velocity dimension; Determining the reference velocity dimension of the current reflection point according to the velocity boundary and the reference threshold value; Determining the screening condition of the current reflection point according to the amplitudes of the echo signals of the multiple reflection points indicated in the reference velocity dimension, and determining whether the current reflection point is a candidate reflection point according to the screening condition and the amplitude of the echo signal of the current reflection point.

4. The method according to claim 1, wherein, The determining the traveling speed of the mobile device relative to the earth according to the relative velocities of the stationary reflection points included in the stationary clutter clustering cluster includes: Using a centroid algorithm for the stationary clutter clustering cluster to determine the average velocity of the stationary reflection points relative to the mobile device according to the relative velocities of the stationary reflection points included in the stationary clutter clustering cluster, as the traveling speed of the mobile device relative to the earth.

5. According to any one of the methods described in claims 1-3, wherein When the number of detection periods is two or more, the method further includes: Performing Kalman filtering processing on the traveling speeds of the mobile device obtained in a plurality of consecutive detection periods, and determining the traveling speed of the mobile device relative to the ground according to the processing result.

6. A data processing device, comprising: An acquisition module, configured to periodically transmit detection signals through a radar carried by a mobile device, and acquire echo signals generated by reflection points of a target reflecting the detection signals when the detection signals irradiate the corresponding target; A first determination module, configured to determine a frame of range-Doppler spectrum according to the echo signals received within one detection period; A second determination module, configured to determine a stationary clutter clustering cluster composed of reflection points stationary relative to the ground according to the distance from the reflection points indicated by the frame of range-Doppler spectrum to the radar and the relative speed of the reflection points relative to the radar; A third determination module, configured to determine the traveling speed of the mobile device relative to the ground according to the relative speeds of the stationary reflection points included in the stationary clutter clustering cluster; The second determination module is specifically configured to screen out at least one candidate reflection point other than the noise reflection points from the multiple reflection points indicated by the range-Doppler spectrum; determine a range cell threshold according to the distance of the at least one candidate reflection point relative to the radar and the relative speed; Determine preliminary screening reflection points that satisfy the range cell threshold from the at least one candidate reflection point to exclude a part of Doppler-azimuth coupling in the stationary clutter region; Perform density clustering on the relative speeds of the preliminary screening reflection points, and determine the clustering result with the longest span as the stationary clutter clustering cluster, and the preliminary screening reflection points included in the stationary clutter clustering cluster are used as stationary reflection points.

7. An electronic device, comprising: A processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the data processing method according to any one of claims 1-5.

8. A computer storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the data processing method according to any one of claims 1-5.

Images