Dual-radar point cloud fusion imaging method and radar system

Abstract

This invention discloses a dual-radar point cloud fusion imaging method and radar system. The imaging method includes: S100, setting up radar A and radar B, which employ linear array antennas, in an orthogonal arrangement of antenna elements, and configuring radar A and radar B to have identical radar parameters; S200, configuring interface A of radar A and interface B of radar B to report 2D point cloud information to the host, where each data point of the 2D point cloud information contains three feature information: distance, velocity, and angle; configuring control port A of radar A and control port B of radar B to receive high real-time frame trigger signals sent by the host to radar A and radar B, respectively; S300, triggering radar A and radar B, and controlling the host to acquire and fuse the 2D point cloud information output by radar A and radar B. This invention uses two synchronous 2D radars to construct a parallel radar system, and the host fuses the 2D point clouds of the two radars into a 3D point cloud, solving the problem of low cost in radar target imaging.

Description

Technical Field

[0001] This invention belongs to the field of radar data processing technology, specifically relating to a dual-radar point cloud fusion imaging method and radar system. Background Technology

[0002] Currently, there are various methods for monitoring human bodies in a scene, such as infrared sensing and image recognition; however, in some specific scenarios, such as smart bathrooms, these technologies have certain pain points, specifically: infrared sensors will fail in a steamy environment; and cameras pose a risk of privacy leaks.

[0003] Radar imaging can achieve target localization and attitude imaging while protecting target privacy. Millimeter-wave radar, in particular, offers all-weather stable sensing capabilities, providing strong support for target localization and attitude imaging. However, current imaging radars require at least an 8T8R antenna configuration. As the number of antennas increases, the resources required for radar signal processing increase exponentially, keeping the price of imaging radar high. Existing technologies attempt to use two freely operating single radars to perform single-point fusion using two single dimensions, but this can only solve single-target localization at a granularity of 0.5m, and its point cloud contains only one data point, failing to support fine-grained target imaging. Summary of the Invention

[0004] This invention provides a dual-radar point cloud fusion imaging method and radar system, achieving fine attitude imaging using low-cost radar and reducing system costs. This invention is achieved through the following technical solution:

[0005] In a first aspect, the present invention provides a dual-radar point cloud fusion imaging method, comprising:

[0006] S100. Radar A and Radar B, which use linear array antennas, are set up in a manner in which the antenna elements are arranged orthogonally to each other, and the radar parameters of Radar A and Radar B are configured to be the same.

[0007] S200: Configure interface A of radar A and interface B of radar B to report 2D point cloud information to the host respectively. Each data point of the 2D point cloud information contains three feature information: distance, speed, and angle. Configure control port A of radar A and control port B of radar B to receive high real-time frame trigger signals sent by the host to radar A and radar B respectively.

[0008] S300, Trigger radar A and radar B, control the host to acquire and fuse the 2D point cloud information output by radar A and radar B, the fusion method includes:

[0009] S310. Find all data points with the same distance and speed in the 2D point cloud information output by radar A and radar B as point pairs;

[0010] S320. Recombining the two data points in each point pair into an imaging point: taking the distance feature information, angle feature information of the data point from radar A and the angle feature information of the data point from radar B in the point pair, and using them as the distance feature information, horizontal angle feature information and pitch angle feature information of the imaging point spherical coordinate system, respectively.

[0011] S330, combine all the imaging points together to form an imaging point cloud.

[0012] As a preferred technical solution, the fusion method further includes:

[0013] S340. Transform the coordinate system of the imaging point cloud from the spherical coordinate system to the rectangular coordinate system to complete the target imaging in the rectangular coordinate system.

[0014] As a preferred technical solution, the specific method for triggering radar A and radar B in step S300 and controlling the host to acquire the 2D point cloud information output by radar A and radar B is as follows:

[0015] S301. After controlling radar A and radar B to power on and complete initialization, wait for the frame trigger signal to be received.

[0016] S302. After the host computer completes the startup, it sends a frame trigger signal to radar A and radar B.

[0017] S303. Radar A and Radar B receive a frame trigger signal, perform a detection operation, and generate 2D point cloud information to report to the host.

[0018] As a preferred technical solution, the specific steps in step S303 where radar A and radar B perform detection operations and generate 2D point cloud information to report to the host include:

[0019] S303a. Control radar A and radar B to send chirp waveforms according to the predetermined chirp length, frequency sweep speed, starting frequency, chirp interval and chirp number respectively;

[0020] S303b: Control radar A and radar B to perform 3D-FFT on the received echoes in the fast time dimension, slow time dimension and antenna dimension respectively to obtain 3D heat maps;

[0021] S303c: Control radar A and radar B to perform CFAR detection on their respective 3D thermal images to obtain 2D point cloud information.

[0022] As a preferred technical solution, when performing 3D-FFT in step S303b, windowing operation is performed using at least one window function among Hann window, Blackman window, Hamm window, Chebyshev window, and rectangular window.

[0023] As a preferred technical solution, in step S300, when triggering radar A and radar B, the frame triggering times of radar A and radar B are controlled to be the same, and the high-low phase difference of the chirp starting frequency points of radar A and radar B is controlled to be Δf, where Δf ≥ BW, and BW is the larger value of the receiver bandwidth of radar A and radar B.

[0024] As a preferred technical solution, when triggering radar A and radar B in step S300, the starting frequency of the chirp frames of radar A and radar B is the same. The frame trigger signals sent by the host to radar A and radar B through control port A and control port B respectively are spaced apart by a preset time τ, and the preset time τ does not exceed the chirp length.

[0025] As a preferred technical solution, the preset time τ is determined by the formula τ=2*BW / slop, where BW is the larger value of the receiver bandwidth of radar A and radar B, and slop is the frequency sweep speed.

[0026] As a preferred technical solution, when setting up radar A and radar B in step S100, the distance between the centers of the antenna arrays of radar A and radar B does not exceed half of their array dimensions.

[0027] Secondly, the present invention provides a radar system including a host, radar A and radar B, wherein the host, radar A and radar B cooperate to perform the dual radar point cloud fusion imaging method described above.

[0028] The dual-radar point cloud fusion imaging method and radar system provided by this invention have the following advantages: Two synchronous 2D radars are arranged in parallel using an orthogonal antenna element orientation to construct a parallel radar system. The host system fuses the 2D point clouds of the two radars into a 3D point cloud. The dense pixels of the 3D point cloud are used to outline the contour of the detected target, thus achieving target imaging. This invention uses only two 2D radars to achieve target imaging, solving the problem of low-cost radar target imaging. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 The main flowchart of the dual radar point cloud fusion imaging method provided in the embodiments of the present invention is shown.

[0031] Figure 2 This is a flowchart of the fusion method in the dual radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0032] Figure 3 This is a block diagram of the radar system adapted to the dual-radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0033] Figure 4 This is a schematic diagram of the radar waveforms of radar A and radar B when the frame triggering times are the same in the dual radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0034] Figure 5 This is a schematic diagram of the radar waveforms of radar A and radar B when the frame trigger times are different in the dual radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0035] Figure 6 The diagram shows the orthogonal arrangement of radar A and radar B using 8x1 linear array antennas when the array distance is zero in the dual radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0036] Figure 7 The diagram shows the orthogonal arrangement of radar A and radar B using 8x1 linear array antennas when the array distance is half the array length in the dual radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0037] Figure 8 This is a schematic diagram of the orthogonal arrangement of two 2T4R radars in the dual-radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0038] Figure 9 for Figure 8 The diagram shows the installation of the two 2T4R radars.

[0039] Figure 10 This is a schematic diagram of parabolic interpolation of antenna dimensions in the dual-radar point cloud fusion imaging method provided in the embodiments of the present invention.

[0040] Figure 11 This is a measured effect diagram of human posture imaging, which is a specific example of the dual radar point cloud fusion imaging method provided in the embodiments of the present invention. Detailed Implementation

[0041] To make the technical solution of the present invention clearer and its technical advantages more apparent, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of the present invention.

[0042] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0043] It should be understood that although the steps in the flowcharts of this application's embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0044] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0045] See Figure 1 As shown, as a basic implementation method, the dual-radar point cloud fusion imaging method provided in this embodiment includes:

[0046] S100. Radar A and Radar B, which use linear array antennas, are set up in a manner where the antenna elements are arranged orthogonally to each other, and the radar parameters of Radar A and Radar B are the same; among which, the radar parameters include: frequency sweep rate, chirp length, chirp spacing, and number of chirps.

[0047] S200, interface A of radar A and interface B of radar B respectively report 2D point cloud information to the host (2D point cloud information refers to point cloud information with only distance and angle in spatial coordinates, and its spatial coordinates all fall on the surface of three-dimensional physical space). Each data point of the 2D point cloud information contains three feature information: distance, velocity, and angle. Control port A of radar A and control port B of radar B respectively receive high real-time frame trigger signals sent by the host to radar A and radar B, and combine them with... Figure 3 As shown; among them, the radar interface refers to the physical channel responsible for high-speed data output and power supply, while the radar control port refers to the communication link responsible only for high real-time command issuance, parameter configuration and status management. The two may share the same port physically, but they are strictly distinguished in terms of logical function.

[0048] S300 triggers radar A and radar B, and the control host acquires and fuses the 2D point cloud information output by radar A and radar B; combined with Figure 2 As shown, the fusion method specifically includes:

[0049] S310. Find all data points with the same distance and velocity in the 2D point cloud information output by radar A and radar B and treat them as point pairs. Discard any points that do not form point pairs as noisy points.

[0050] S320. Reassemble the two data points in each point pair (from radar A and radar B respectively) into one imaging point: Take the distance feature information and angle feature information of the data point from radar A and the angle feature information of the data point from radar B in the point pair, and use them as the distance feature information, horizontal angle feature information and pitch angle feature information of the imaging point in the spherical coordinate system respectively; that is, the coordinates of the imaging point in the spherical coordinate system are three dimensions: {distance, horizontal angle, pitch angle}.

[0051] S330, combine all imaging points together to form an imaging point cloud.

[0052] As a preferred embodiment, if the aforementioned imaged point cloud needs to be subsequently applied, processed, or displayed in a Cartesian coordinate system, the fusion method further includes:

[0053] S340. Transform the coordinate system of the imaging point cloud from the spherical coordinate system to the Cartesian coordinate system, i.e., the X, Y, Z three-dimensional coordinate system, to complete the target imaging in the Cartesian coordinate system. In this way, the host can transform the imaging point cloud from the spherical coordinate system to the Cartesian coordinate system, and then send it to the subsequent processing or display it on the GUI to realize imaging.

[0054] As a preferred embodiment, the specific method for triggering radar A and radar B in step S300 and controlling the host to acquire the 2D point cloud information output by radar A and radar B is as follows:

[0055] S301. After control radar A and radar B complete initialization upon power-on, wait for the receive frame trigger signal:

[0056] S302. After the control host completes the startup, it sends a frame trigger signal to radar A and radar B.

[0057] S303, Radar A, and Radar B receive the frame trigger signal, perform detection operations, and generate 2D point cloud information to report to the host.

[0058] More specifically, the specific steps in step S303 where radar A and radar B perform detection operations and generate 2D point cloud information to report to the host include:

[0059] S303a, control radar A and radar B respectively transmit chirp waveforms according to the predetermined chirp length, frequency sweep speed, starting frequency, chirp interval and chirp number;

[0060] S303b, control radar A, and radar B perform 3D-FFT on the received echoes in the fast time dimension, slow time dimension, and antenna dimension, respectively, to obtain 3D heatmaps. Windowing operations can be optionally performed when performing 3D-FFT, and window functions include Hann window, Blackman window, Hamm window, Chebyshev window, rectangular window, etc.

[0061] S303c, control radar A, and radar B each perform CFAR detection on their respective 3D thermal images to obtain 2D point cloud information.

[0062] Ideally, both radar A and radar B are FMCW radars and are high-resolution radars, capable of achieving high range resolution, high velocity resolution, and high angular resolution. Radar A and radar B typically have no fewer than four antenna elements, preferably eight (if it's a 2T4R configuration using MIMO to virtually generate eight antennas, that's also considered eight antennas; similar configurations include 3T4R with 12 virtual antennas, 4T4R with 16 virtual antennas, 8T8R with 64 virtual antennas, etc.). Radar A and radar B are placed compactly together, ensuring that the center-to-center distance between their arrays does not exceed half the sum of their array dimensions. This ensures that the difference in radar viewing angles is sufficiently small, meaning that the target viewing angles are the same (if the two radars are far apart, the target will appear from different sides, and the 2D point cloud information will lose its basis for fusion).

[0063] Furthermore, the interval between the host sending frame trigger signals to radars A and B must be small enough to ensure that radars A and B measure the target's motion at the same time (if the time difference is too large, the target's attitude will change, and the 2D point cloud information will lose its basis for fusion). Because radars A and B are physically adjacent, to avoid their electromagnetic signals interfering with each other, it is necessary to ensure that their electromagnetic signal frequencies are outside the receiver bandwidth of the other at any given time. There are two specific methods for this:

[0064] Method (1): See Figure 4 As shown, the frame trigger times for control radar A and radar B are the same, and radar A and radar B start working simultaneously. The high and low phase difference between the starting frequencies of the chirps for control radar A and radar B is Δf, where Δf ≥ BW, and BW is the larger value of the receiver bandwidth for radar A and radar B. A frame consists of one or more chirps, and the frame trigger time is the start time of the first chirp in a frame.

[0065] Method (2): See Figure 5 As shown, radars A and B share the same chirp start frequency. The host sends frame trigger signals to radars A and B respectively via control ports A and B, with a preset time interval τ between them. That is, the frame trigger times of radars A and B differ by a preset time τ. Radar A starts operating first, and radar B starts operating after a certain delay τ, ensuring that the preset time τ is long enough but does not exceed the chirp length, allowing radar A's frequency sweep to reach the out-of-band area of ​​radar B. Preferably, the preset time τ is determined by the formula τ = 2 * BW / slop, where BW is the larger of the receiver bandwidths of radars A and B, and slop is the frequency sweep rate.

[0066] To make the above implementation methods clearer and easier to understand, a specific example is given, taking the above method (1) as an example:

[0067] Radar A and Radar B are both 2T4R MIMO radars operating in the 60GHz band. Their array antennas are virtually configured as an 8x1 linear array. Radar A is placed horizontally, and Radar B is placed vertically. Figure 8 As shown, its array after MIMO expansion is equivalent to Figure 6 and Figure 7 The celestial array shown.

[0068] The radar parameter configurations for Radar A and Radar B are as follows:

[0069] Frequency sweep rate = 195MHz / µs;

[0070] chirp length = 25.6us;

[0071] chirp interval = 417us;

[0072] Number of chirps = 64.

[0073] The starting frequency and trigger time of radar A and radar B are configured as follows:

[0074] Radar A's starting frequency = 59 GHz;

[0075] Radar B's starting frequency = 59.008 GHz;

[0076] Δf = 8MHz;

[0077] τ=0.

[0078] The interface and control port configurations for Radar A and Radar B are as follows:

[0079] Interface A and Interface B: UART interfaces with a baud rate of 921600;

[0080] Control port A and control port B: GPIO.

[0081] like Figure 9 As shown, the radar system, including radar A, radar B, and the main unit, is installed with the beam pointing diagonally downwards to fully utilize the ultra-high range resolution characteristics of the 60GHz millimeter-wave radar. Figure 9 In this context, FOV stands for Field of View; it refers to the far-field movement of the human body within the imaging radar.

[0082] The host sends a frame trigger signal to radar A and radar B every 33.33ms to provide a refresh rate of 30fps. Upon receiving the frame trigger signal, radar A and radar B immediately begin chirp transmission. After receiving a full frame (i.e., 64 chirs), they perform the following operations:

[0083] a) Perform a fast time dimension FFT (also known as distance FFT) on it, and here we choose to apply a Hann window;

[0084] b) Then perform a slow-time FFT (also known as a Doppler FFT) on it, here we choose to apply a Hann window;

[0085] c) Finally, perform zero-padding FFT on the antenna dimension. Here, we choose not to add a window and pad with 8 zeros, which means that the data from 8 antennas are sent to perform 16-point zero-padding FFT to synthesize 3D-FFT.

[0086] After performing 3D-FFT, CA-CFAR (Cell-Averaging Constant False Alarm Rate) is applied to the data. For each point satisfying the CA-CFAR condition, interpolation along the angular dimension is performed. This involves taking the logarithmic power values ​​of the current point cloud and the two preceding / following points along the antenna dimension, and then performing parabolic interpolation. Figure 10 As shown, precise angles are obtained. After CA-CFAR processing, a 2D point cloud is obtained, where each data point contains {range, velocity, angle}. The distance is represented by the range gate number, the velocity by the velocity gate number, and the angle by the precise angle value output by parabolic interpolation. Because there are range resolution, velocity resolution, and angle resolution, based on actual radar testing experience, the number of point clouds in the 2D point clouds of radar A and radar B is approximately 50 to 200.

[0087] Radar A and Radar B each report their 2D point cloud information to the host via the UART interface. After receiving the 2D point cloud data from Radar A and Radar B, the host immediately performs a pairwise comparison point by point. The pseudocode is as follows:

[0088] for k=1 to P;

[0089] x = the kth point in the 2D point cloud reported by radar A;

[0090] for h=1 to Q;

[0091] y = the h-th point in the 2D point cloud reported by radar B;

[0092] If (the distance of x equals the distance of y) and (the velocity of x equals the velocity of y)

[0093] The imaging point z = {distance x, angle x, angle y};

[0094] The imaging point z is accumulated and incorporated into the imaging point cloud storage space;

[0095] end;

[0096] end;

[0097] end.

[0098] Where P represents the number of data points in the 2D point cloud reported by radar A, and Q represents the number of data points in the 2D point cloud reported by radar B. After fusion processing, an imaging point cloud in spherical coordinates is obtained, which can be converted to Cartesian coordinates if necessary. Based on actual radar testing experience, the number of points in the imaging point cloud is approximately 30 to 50.

[0099] The host performs sliding window merging on the imaged point clouds to increase point cloud density and improve image quality. Specifically, when reporting the imaged point cloud of the current frame, it merges it with the imaged point clouds of the two most recent frames. That is, the host reports the union of the imaged point clouds of frames s-2, s-1, and s in frame s. A measured effect image of human pose imaging is shown below. Figure 11 As shown.

[0100] Combination Figure 3 As shown, this embodiment of the invention also provides a radar system, including a host, radar A and radar B, wherein the host, radar A and radar B cooperate to perform the dual radar point cloud fusion imaging method described above.

[0101] The dual-radar point cloud fusion imaging method and radar system provided in the above embodiments mainly address the problem of low-cost 3D target imaging, with low cost being the core objective. This is because commercially available 3D imaging radars typically require numerous antennas and substantial computing resources. In the above embodiments of this invention, two synchronous 2D radars are arranged in parallel using an orthogonal antenna element configuration. The host system fuses the 2D point clouds of the two radars into a 3D point cloud, using the dense pixels of the 3D point cloud to outline the contour of the detected target, thus achieving target imaging. In other words, the above embodiments of this invention achieve target imaging using only two 2D radars, solving the problem of low-cost radar target imaging. For example, in 2026, a commercially available 8x8 antenna radar costs approximately $10, while the two 2x4 radars in this embodiment cost only approximately $4-5, saving at least half the price.

[0102] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A dual-radar point cloud fusion imaging method, characterized in that, include: S100. Radar A and Radar B, which use linear array antennas, are set up in a manner in which the antenna elements are arranged orthogonally to each other, and the radar parameters of Radar A and Radar B are configured to be the same. S200: Configure interface A of radar A and interface B of radar B to report 2D point cloud information to the host respectively. Each data point of the 2D point cloud information contains three feature information: distance, speed and angle. Configure control port A of radar A and control port B of radar B to receive high real-time frame trigger signals sent by the host to radar A and radar B respectively. S300, Trigger radar A and radar B, and control the host to acquire and fuse the 2D point cloud information output by radar A and radar B. The fusion method includes: S310. Find all data points with the same distance and speed in the 2D point cloud information output by radar A and radar B as point pairs; S320. Recombining the two data points in each point pair into an imaging point: taking the distance feature information, angle feature information of the data point from radar A and the angle feature information of the data point from radar B in the point pair, and using them as the distance feature information, horizontal angle feature information and pitch angle feature information of the imaging point spherical coordinate system, respectively. S330, combine all the imaging points together to form an imaging point cloud.

2. The dual-radar point cloud fusion imaging method according to claim 1, characterized in that, The fusion method further includes: S340. Transform the coordinate system of the imaging point cloud from the spherical coordinate system to the rectangular coordinate system to complete the target imaging in the rectangular coordinate system.

3. The dual-radar point cloud fusion imaging method according to claim 1, characterized in that, The specific method for triggering radar A and radar B in step S300 and controlling the host to acquire the 2D point cloud information output by radar A and radar B is as follows: S301. Control radar A and radar B to power on and complete initialization, and wait to receive frame trigger signal; S302. After the host computer completes the startup, it sends a frame trigger signal to radar A and radar B. S303. Radar A and Radar B receive a frame trigger signal, perform a detection operation, and generate 2D point cloud information to report to the host.

4. The dual-radar point cloud fusion imaging method according to claim 3, characterized in that, The specific steps in step S303 where radar A and radar B perform detection operations and generate 2D point cloud information to report to the host include: S303a. Control radar A and radar B to send chirp waveforms according to the predetermined chirp length, frequency sweep speed, starting frequency, chirp interval and chirp number respectively; S303b: Control radar A and radar B to perform 3D-FFT on the received echoes in the fast time dimension, slow time dimension and antenna dimension respectively to obtain 3D heat maps; S303c: Control radar A and radar B to perform CFAR detection on their respective 3D thermal images to obtain 2D point cloud information.

5. The dual-radar point cloud fusion imaging method according to claim 4, characterized in that, When performing 3D-FFT in step S303b, windowing operations are performed using at least one of the following window functions: Hann window, Blackman window, Hamm window, Chebyshev window, and rectangular window.

6. The dual-radar point cloud fusion imaging method according to claim 1, characterized in that, In step S300, when radar A and radar B are triggered, the frame triggering times of radar A and radar B are controlled to be the same, and the high-low phase difference of the chirp starting frequency points of radar A and radar B is controlled to be Δf, where Δf ≥ BW, and BW is the larger value of the receiver bandwidth of radar A and radar B.

7. The dual-radar point cloud fusion imaging method according to claim 4, characterized in that, In step S300, when radar A and radar B are triggered, the starting frequency of chirp for radar A and radar B is the same. The frame trigger signals sent by the host to radar A and radar B through control port A and control port B respectively are spaced apart by a preset time τ, and the preset time τ does not exceed the chirp length.

8. The dual-radar point cloud fusion imaging method according to claim 7, characterized in that, The preset time τ is determined by the formula τ=2*BW / slop, where BW is the larger value of the receiver bandwidth of radar A and radar B, and slop is the frequency sweep rate.

9. The dual-radar point cloud fusion imaging method according to claim 1, characterized in that, When setting up radar A and radar B in step S100, the distance between the centers of the antenna arrays of radar A and radar B shall not exceed half of the sum of their array sizes.

10. A radar system, comprising a main unit, radar A, and radar B, characterized in that: The host computer, radar A, and radar B work together to perform the dual-radar point cloud fusion imaging method according to any one of claims 1-9.

Images