Multipath false target detection and identification method of vehicle-mounted millimeter-wave radar

Abstract

This application provides a method for multipath false target detection and identification using vehicle-mounted millimeter-wave radar. The method includes acquiring a single-frame echo signal from the target vehicle-mounted millimeter-wave radar; filtering the single-frame echo signal to generate a single-frame point set; performing multipath pattern judgment on the single-frame point set to generate a multipath judgment result; associating the single-frame over-detected points in the single-frame point set with flight paths based on the multipath judgment result and preset rules to generate a target flight path; calculating multipath confidence information on the target flight path to generate a comprehensive multipath confidence evaluation value; and performing multipath false target detection and identification processing based on the comprehensive multipath confidence evaluation value and a preset confidence threshold. In this way, accurate multipath judgment can be performed under complex road conditions.

Description

Technical Field

[0001] The embodiments of this application relate to the field of intelligent driving environment perception, and in particular to a method for multipath false target detection and recognition using vehicle-mounted millimeter-wave radar. Background Technology

[0002] Vehicle-mounted millimeter-wave radar has become a standard feature of ADAS / autonomous driving systems due to its advantages such as all-weather operation, accurate speed measurement, and moderate cost. According to the SAE J3016 standard, Level 2 and above autonomous driving systems require at least 3-5 millimeter-wave radars to achieve 360° environmental perception. However, radar waves encounter strong reflectors such as bridge railings, tunnel walls, and metal road signs during propagation, resulting in multipath interference. Common multipath interference includes signals indirectly reaching the real target via reflectors and then returning, as well as false signals formed by multiple reflections between the reflector and the target. Multipath interference is a major challenge in millimeter-wave radar target detection, leading to false alarms and false braking in real-world driving conditions, thus reducing the intelligent driving experience. Therefore, addressing this issue is necessary.

[0003] To address existing problems, current technologies often employ deep neural network-based training datasets or acquire track information of targets to be identified from millimeter-wave radar. However, these solutions suffer from numerous drawbacks. For instance, training dataset-based solutions rely on large amounts of labeled data, resulting in high costs for obtaining ground truth labels for multipath scenarios. Due to the diversity and variability of complex real-world road conditions, training sets are extremely complex and consume excessive memory, making these solutions unsuitable for practical applications. Alternatively, track information-based solutions utilize the track relationships between real targets and stationary reflectors to match real and false targets, thus identifying multipath false targets. However, if the initial clustering of track points is incorrect, or if multipath points merge with other targets, the multipath analysis fails. Therefore, there is an urgent need for a technology capable of accurately determining multipath in complex real-world road conditions. Summary of the Invention

[0004] According to an embodiment of this application, a method for multipath false target detection and identification using vehicle-mounted millimeter-wave radar is provided, which can accurately perform multipath judgment under actual complex road conditions.

[0005] In a first aspect of this application, a method for detecting and identifying multipath false targets using a vehicle-mounted millimeter-wave radar is provided. The method includes:

[0006] Acquire a single frame echo signal from the target vehicle's millimeter-wave radar;

[0007] The single-frame echo signal is filtered to generate a single-frame point set;

[0008] Perform multipath pattern judgment on the single-frame point trace set and generate multipath judgment results;

[0009] Based on the multipath judgment result and preset rules, the single-frame over-detected points in the single-frame point set are associated with the track to generate the target track;

[0010] The target trajectory is processed by multipath confidence information calculation to generate a comprehensive multipath confidence evaluation value;

[0011] Multipath false target detection and identification processing is performed based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

[0012] In one possible implementation, the step of filtering the single-frame echo signal to generate a single-frame point set includes:

[0013] The single-frame echo signal is subjected to Fourier transform processing to generate a two-dimensional frequency domain matrix;

[0014] Constant false alarm rate detection is performed in the two-dimensional frequency domain matrix to filter out single-frame over-detection points.

[0015] Calculate the target calculation data for each of the single-frame over-detection points, and sort the single-frame over-detection points according to the point distance to generate a single-frame point set, wherein the target calculation data includes distance, velocity, azimuth angle and energy value.

[0016] In one possible implementation, the step of performing multipath pattern judgment on the single-frame point set and generating multipath judgment results includes:

[0017] Determine whether the single-frame over-detected point trace in the single-frame point trace set satisfies the preset multipath law equation;

[0018] If the single-frame over-detection point satisfies the preset multipath law equation, then the single-frame over-detection point is determined to be a multipath false target and deleted.

[0019] If the single-frame over-detection point does not satisfy the preset multipath law equation but is within the redundancy range of the preset multipath law equation, then a multipath confidence level is assigned to the single-frame over-detection point.

[0020] In one possible implementation, determining whether the single-frame over-detected point trace in the single-frame point trace set satisfies the preset multipath law equation includes:

[0021] Obtain all the judgment conditions of the preset multipath law equation;

[0022] Determine whether the single-frame over-detection point meets all the determination conditions;

[0023] If the single-frame over-detection point meets all the judgment conditions, then the single-frame over-detection point is determined to satisfy the preset multipath law equation.

[0024] If the single-frame over-detection point does not meet all the judgment conditions, then it is determined that the single-frame over-detection point does not satisfy the preset multipath law equation.

[0025] In one possible implementation, assigning a multipath confidence score to the single-frame over-detection trace includes:

[0026] When the number of the judgment conditions for the single-frame over-detection traces meets the first data quantity value;

[0027] Based on the target calculation data, the threshold of the single-frame over-detection point is relaxed, and it is determined whether the single-frame over-detection point of the relaxed threshold process satisfies all the judgment conditions.

[0028] If the single-frame over-detection trace processed by relaxing the threshold satisfies all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the first confidence level.

[0029] If the single-frame over-detection trace processed by relaxing the threshold does not meet all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the second confidence level.

[0030] When the number of the judgment conditions for the single-frame over-detection traces meets the second data quantity value;

[0031] Based on the target calculation data, the threshold of the single-frame over-detection point is relaxed, and it is determined whether the single-frame over-detection point of the relaxed threshold process satisfies all the judgment conditions.

[0032] If the single-frame over-detection trace processed by relaxing the threshold satisfies all the above judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the third confidence level.

[0033] If the single-frame over-detection trace processed by relaxing the threshold does not meet all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the fourth confidence level.

[0034] When the number of over-detection points in a single frame meets the judgment condition is the third data value, the multipath confidence level assigned to the over-detection point in the single frame is the fifth confidence level.

[0035] In one possible implementation, the step of calculating and processing the multipath confidence information of the target trajectory to generate a comprehensive multipath confidence evaluation value includes:

[0036] Calculate the arithmetic mean of the multipath confidence of all associated points in a single frame for the target track, and use the arithmetic mean as the track confidence of the over-detected points in the single frame;

[0037] An exponentially weighted average of the track confidence scores across multiple consecutive frames is used to obtain a comprehensive multipath confidence score.

[0038] In one possible implementation, the multipath false target detection and identification process based on the comprehensive multipath confidence evaluation value and the preset confidence threshold includes:

[0039] The comprehensive multipath confidence evaluation value is compared with a preset confidence threshold to determine whether the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold.

[0040] If the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold, then the target track is determined to be a multipath false target track and is removed.

[0041] If the comprehensive multipath confidence evaluation value is less than the preset confidence threshold, the target trajectory is determined to be a real trajectory and is retained for reporting.

[0042] In a second aspect of this application, a multipath false target detection and identification device related to vehicle-mounted millimeter-wave radar is provided. The device includes:

[0043] The echo signal acquisition module is used to acquire a single frame echo signal from the target vehicle-mounted millimeter-wave radar.

[0044] The dot set generation module is used to filter and process the single-frame echo signal to generate a single-frame dot set.

[0045] The multipath judgment module is used to judge the multipath pattern of the single-frame point set and generate multipath judgment results.

[0046] The target trajectory generation module is used to associate the single-frame over-detected points in the single-frame point set with the multipath judgment result and preset rules to generate a target trajectory.

[0047] The comprehensive evaluation generation module is used to calculate and process the multipath confidence information of the target trajectory and generate a comprehensive multipath confidence evaluation value.

[0048] The false target identification module is used to perform multipath false target detection and identification processing based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

[0049] In a third aspect of this application, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described above.

[0050] In a fourth aspect of this application, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the method according to the first aspect of this application.

[0051] The multipath false target detection and identification method for vehicle-mounted millimeter-wave radar provided in this application first obtains precise point information including distance, speed, azimuth, and energy values ​​by performing Fourier transform and constant false alarm rate detection on single-frame echo signals, providing a reliable data foundation for multipath analysis. Next, multipath pattern judgment is performed at the single-frame level, and points that strictly conform to the multipath equation are immediately deleted, directly reducing the number of false targets and subsequent computational load. For uncertain points within the redundancy range, multipath confidence is introduced for quantification and labeling, improving the adaptability of single-frame processing. Then, the points with confidence are associated to form tracks. By calculating the arithmetic mean of the single-frame confidence and then performing an exponentially weighted average of multiple consecutive frames, a comprehensive evaluation value is generated. Finally, by comparing the comprehensive evaluation value with a threshold, multipath false target tracks are reliably identified and eliminated. Through the close connection from precise point extraction to flexible single-frame judgment and robust multi-frame fusion, accurate multipath judgment can be performed under complex road conditions while ensuring low computational complexity.

[0052] It should be understood that the description in the Summary Section is not intended to limit the key or essential features of the embodiments of this application, nor is it intended to restrict the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0053] The above and other features, advantages, and aspects of the embodiments of this application will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0054] Figure 1 The flowchart below shows a method for detecting and identifying multipath false targets using vehicle-mounted millimeter-wave radar according to an embodiment of this application.

[0055] Figures 2(a) and 2(b) are example diagrams illustrating multipath false target detection and identification using vehicle-mounted millimeter-wave radar according to embodiments of this application;

[0056] Figure 3 This is a block diagram of a multipath false target detection and identification device for vehicle-mounted millimeter-wave radar according to an embodiment of this application;

[0057] Figure 4 This is a schematic diagram of the structure of a terminal device or server suitable for implementing the embodiments of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0059] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0060] Figure 1 A flowchart of a multipath false target detection and identification method involving vehicle-mounted millimeter-wave radar according to an embodiment of the present disclosure is shown.

[0061] like Figure 1 As shown, the main process of this method is described below (steps S101 to S106):

[0062] Step S101: Acquire a single frame echo signal from the target vehicle-mounted millimeter-wave radar.

[0063] In some embodiments, the millimeter-wave radar sensor, used in in-vehicle intelligent driving applications, operates powered by the vehicle's system. The millimeter-wave radar sensor continuously emits electromagnetic waves, which are reflected back upon encountering a target. The sensor receives these echo signals, thus achieving target detection. Each radar frame comprises the transmitted signal, the received signal, and data processing, forming a complete frame. Therefore, when the in-vehicle millimeter-wave radar is operating, it continuously emits electromagnetic waves, which are reflected upon encountering a target, resulting in a single frame of echo signal.

[0064] Step S102: The single-frame echo signal is filtered to generate a single-frame point set.

[0065] For step S102, Fourier transform processing is performed on the single-frame echo signal to generate a two-dimensional frequency domain matrix; constant false alarm rate detection is performed in the two-dimensional frequency domain matrix to filter out single-frame over-detection points; target calculation data for each single-frame over-detection point is calculated, and the single-frame over-detection points are sorted according to the point distance to generate a single-frame point set, wherein the target calculation data includes distance, velocity, azimuth angle and energy value.

[0066] In some embodiments, a Fourier transform is performed on the real-time acquired single-frame echo signal to obtain a two-dimensional frequency domain matrix. Constant false alarm rate (CFAR) peak detection is then performed on this matrix to filter out frequency domain peak points. These peak points are used as single-frame over-detection points. Based on these over-detection points, target calculation data such as distance, velocity, azimuth, and energy values ​​between the target vehicle-mounted millimeter-wave radar and the target are obtained. Simultaneously, all single-frame over-detection points are sorted in ascending order according to their distance to obtain a single-frame point set. Each single-frame overcheck point trace contains distance. (m), velocity (m / s), angular velocity of azimuth (rad / s), energy value (dB) contains 4 pieces of information, represented as a 4-dimensional feature vector (1):

[0067] (1)

[0068] Step S103: Perform multipath pattern judgment on the single frame point set and generate multipath judgment result.

[0069] For step S103, determine whether the single-frame over-detected point trace in the single-frame point trace set satisfies the preset multipath law equation; if the single-frame over-detected point trace satisfies the preset multipath law equation, then the single-frame over-detected point trace is determined as a multipath false target and deleted; if the single-frame over-detected point trace does not satisfy the preset multipath law equation but is within the redundancy range of the preset multipath law equation, then a multipath confidence score is assigned to the single-frame over-detected point trace.

[0070] In some embodiments, the overall execution method is to process the single-frame point set. Perform two nested for loops to iterate through the data. Since the multipath distance of the target is always larger than the direct round-trip distance of the actual target, this is applied to the single-frame point set that has already been sorted by distance. Simplify the for loop for dots. ,from Start iterating and comparing to reduce the number of iterations in the for loop.

[0071] Specifically, the distance, velocity, azimuth, and energy values ​​of the single-frame over-detected points in the single-frame point set are compared using a preset multipath law equation. During this process, if a single-frame over-detected point strictly satisfies the preset multipath law equation, the confidence level of the single-frame over-detected point is set to 1, and it is determined to be a multipath false target that can be directly deleted. If a single-frame over-detected point does not satisfy the preset multipath law equation but is within the redundancy range of the preset multipath law equation, the confidence level of the single-frame over-detected point is set to be greater than 0 and less than 1, and further multipath confidence assignment processing is performed.

[0072] Furthermore, determining whether a single-frame over-checked point in the single-frame point set satisfies the preset multipath law equation includes: obtaining all the judgment conditions of the preset multipath law equation; determining whether a single-frame over-checked point meets all the judgment conditions; if a single-frame over-checked point meets all the judgment conditions, then the single-frame over-checked point is determined to satisfy the preset multipath law equation; if a single-frame over-checked point does not meet all the judgment conditions, then the single-frame over-checked point does not satisfy the preset multipath law equation.

[0073] The pre-set multipath law equation uses a double multipath law as an example. The actual target's point information is: The dot information of the false target is The distance and velocity of the false dots are twice that of the real dots, and their azimuth angles are similar. The energy value of the false dots decreases to varying degrees depending on the distance to the real dots; the energy decreases more with greater distance. Different energy thresholds need to be set for near, medium, and far distances. The following relationship is satisfied:

[0074] (2)

[0075] (3)

[0076] (4)

[0077] (5)

[0078] Where Wthred1 is the energy threshold for close-range (di≤Rnear) scenarios, Rnear is the close-range distance threshold, Wthred2 is the energy threshold for medium-range (Rnear≤di≤Rfar) scenarios, Rfar is the far-range distance threshold, Wthred3 is the energy threshold for far-range (di≥Rfar) scenarios, ωj is the angular velocity of the false dots, and ωi is the angular velocity of the real dots.

[0079] If all four conditions (2), (3), (4), and (5) above are met, then the dot pattern is considered to be... For dots If the number of false points detected by a multipath event is twice that of the number of false points, and the over-detected points in a single frame satisfy the preset multipath law equation, then the points are considered to be false. For dots Multiple-path false points, confidence level Set to 1, delete the dots, and exit the current page. The traversal begins. The multipath pattern is rigorously compared. If any of the four conditions is not met, the single-frame over-detection point is determined to not satisfy the preset multipath pattern equation, and further investigation is required. Redundancy judgment of multipath pattern.

[0080] Furthermore, assigning a multipath confidence score to a single-frame over-detection trace includes: when the number of judgment conditions met by a single-frame over-detection trace is a first data quantity value; performing threshold relaxation processing on the single-frame over-detection trace based on the target calculated data, and determining whether the threshold-relaxed single-frame over-detection trace meets all judgment conditions; if the threshold-relaxed single-frame over-detection trace meets all judgment conditions, then the multipath confidence score assigned to the single-frame over-detection trace is the first confidence score; if the threshold-relaxed single-frame over-detection trace does not meet all judgment conditions, then the multipath confidence score assigned to the single-frame over-detection trace is the second confidence score; when the judgment conditions of a single-frame over-detection trace are met... When the number of items that meet the criteria is the second data value; based on the target calculation data, the threshold of the single-frame over-inspection point is relaxed, and it is determined whether the single-frame over-inspection point that has been relaxed meets all the judgment conditions; if the single-frame over-inspection point that has been relaxed meets all the judgment conditions, then the multipath confidence level assigned to the single-frame over-inspection point is the third confidence level; if the single-frame over-inspection point that has been relaxed does not meet all the judgment conditions, then the multipath confidence level assigned to the single-frame over-inspection point is the fourth confidence level; when the number of items that meet the judgment conditions of the single-frame over-inspection point is the third data value, the multipath confidence level assigned to the single-frame over-inspection point is the fifth confidence level.

[0081] If the number of over-detected traces in a single frame meets the judgment condition of the first data quantity value, that is, three of the above conditions (2), (3), (4), and (5) are met, then the thresholds for the remaining unmet conditions are relaxed. Assuming that the distance, velocity, and energy values ​​are all met, and only the azimuth threshold is not met, the azimuth threshold can be appropriately relaxed, that is, expression (4) can be changed to:

[0082] (4-1);

[0083] If relaxing the azimuth threshold can satisfy the requirement, then for the point trace confidence level Set the confidence level to 0.8. If relaxing the azimuth threshold still does not meet the requirements, then... confidence level Set it to the second confidence level, i.e., 0.3. Similarly, if the distance, azimuth, and energy values ​​are satisfied, but the speed threshold is not, the speed threshold can be appropriately relaxed.

[0084] If the number of over-detected traces in a single frame meets the criteria for the second data value (i.e., two of the conditions in (2), (3), (4), and (5) are met), then the thresholds for the remaining unmet conditions are relaxed. For example, if the distance and velocity are met, but the azimuth and energy thresholds are not, the energy and azimuth thresholds can be appropriately relaxed. If the relaxed thresholds are met, then the traces are... confidence level Set to the third confidence level, i.e., 0.5. If the relaxed threshold is still not met, then the trace... confidence level Set to the fourth confidence level, i.e., 0.1. Similarly, if the distance and azimuth are satisfied, but the speed and energy thresholds are not, the speed and energy thresholds can be appropriately relaxed in the same way. It should be noted that the specific amount of relaxation needs to be set according to actual needs when appropriately relaxing the thresholds, and no specific limit is made here.

[0085] If the number of over-detected traces in a single frame meets the criteria is the third data value, that is, less than two of the conditions in (2), (3), (4), and (5) above are met, then the traces are... confidence level The confidence level is set to the fifth level, which is 0. At this point, the single-frame tracking device has made a preliminary judgment on multipath patterns. Tracks that strictly satisfy multipath patterns have been deleted, and potentially spurious multipath tracks have been assigned different confidence levels and transmitted to the tracking device. Confidence level The larger the value, the more likely the dot feature is a multipath spurious feature.

[0086] Step S104: Based on the multipath judgment result and preset rules, the single-frame over-detected points in the single-frame point set are associated with the target point to generate the target point.

[0087] In some embodiments, a track refers to associating multiple frames of points that satisfy the same motion law to form a continuous target motion trajectory that can match the actual motion state of the target. Therefore, based on the multipath judgment result, the track association of single-frame over-detected points that satisfy the same motion law in multiple frames can be obtained.

[0088] Step S105: Perform multipath confidence information calculation and processing on the target trajectory to generate a comprehensive multipath confidence evaluation value.

[0089] For step S105, the arithmetic mean of the multipath confidence of all associated points in a single frame of the target track is calculated, and the arithmetic mean is used as the track confidence of the over-checked points in a single frame; the track confidence of multiple consecutive frames is calculated by exponential weighted average to obtain the comprehensive multipath confidence evaluation value.

[0090] In some embodiments, for target track The multipath confidence information of the points associated in a single frame is processed. If the number of points is 0, then the target track is considered to be... Multipath confidence If the number of associated points is 0, then... Then, the multipath confidence scores of the track points are averaged to obtain the m-th frame of the track. The confidence expression is as follows:

[0091] (6)

[0092] The weighted average of the multi-frame multipath confidence index of the target trajectory is the confidence score of multiple consecutive frames. When performing an exponentially weighted average, since the tracks of multipath false targets will enter the system decision once they are officially reported and output, the multipath judgment needs to be completed before the tracks are reported and output.

[0093] If the track is reported and output after 4 frames of stability, the track confidence score needs to be calculated by an exponentially weighted average of the track confidence scores of the previous 4 frames. The expression is as follows:

[0094] (7);

[0095] in, The smoothing coefficient satisfies , which is a weighted average of all past sequence values, with the weights decaying exponentially.

[0096] Step S106: Perform multipath false target detection and identification processing based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

[0097] For step S106, the comprehensive multipath confidence evaluation value is compared with the preset confidence threshold to determine whether the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold. If the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold, the target track is determined to be a false multipath target track and is removed. If the comprehensive multipath confidence evaluation value is less than the preset confidence threshold, the target track is determined to be a real track and is retained for reporting.

[0098] In some embodiments, the preset reliability threshold is set to 0.7, if the calculated reliability threshold is 0.7. If the value is greater than or equal to 0.7, the track is judged as a multipath false target and directly deleted without being reported. If the calculated value is... If the value is less than 0.7, the track is considered a real target and can be reported and output normally.

[0099] To illustrate the above points, let's take a specific example:

[0100] Referring to Figure 2, the two-dimensional frequency domain matrix of the millimeter-wave radar is shown. The peak points of the real target and the false target are outlined in the figure. (a) The range, velocity, azimuth, and energy values ​​in the figure strictly satisfy the multipath law. (b) The velocity, azimuth, and energy values ​​in the figure strictly satisfy the multipath law, and the range is within the redundancy range of the multipath law.

[0101] Taking the specific data in Figure 2(b) as an example, single-frame point detection is performed and sorted by distance d from smallest to largest. The point information is shown in Table 1 below:

[0102]

[0103] Table 1 Single Frame Detected Track Information Table

[0104] As shown in Table 1, the specific multipath pattern lookup is performed using the first for loop. arrive The second for loop iterates through... arrive That is, for In other words, from traversing to the next Find its multipath false points; for In other words, from traversing to the next Find its multipath false points.

[0105] in To ensure accurate identification, from traversing to the next To find its multipath false points, according to formulas (2), (3), (4), and (5), we can see that... The distance range of the false dots corresponding to the dots is [ 2-0.5, [2 + 0.5] = [15.24, 16.24], the speed range is [-4.25]. 2-0.5, -4.25 2+0.5]=[8.00, 9.00], azimuth range is [0.58-0.05, 0.58+0.5]=[0.53, 0.63], energy threshold is 132.05-10=122.05. Based on the multipath logic, , , , , Since only one condition, energy, is satisfied, the confidence level of the false dot is 0. This is a false point. Three conditions—velocity, azimuth, and energy—must be strictly met. By appropriately relaxing the distance threshold and performing redundant multipath pattern checks, if the distance difference is within 2 meters, then... Confidence level of being considered a fake mark It is 0.8. , , , , Since only one condition, energy, is satisfied, the confidence level of the false dot is 0.

[0106] The rigorous comparison of multipath patterns and the redundancy judgment of multipath patterns in a single frame have been completed, only... If a point is identified as a false point and does not meet the strict comparison of multipath rules, it cannot be directly deleted; however, if it meets the redundancy judgment of multipath rules, further multipath false judgment at the track end is required.

[0107] A track refers to associating points that satisfy the same motion pattern across multiple frames to form a continuous target motion trajectory that can match the actual motion state of the target. In this example, the peak value of false points exists continuously for many frames, and only one false point is detected in each frame. Therefore, the number of points associated with track i in each frame is... The arithmetic mean of the multipath confidence scores of the associated points is calculated to obtain the m-th frame of the track. The confidence level is Furthermore, if the continuous correlation points of the track start normally with M=4 frames, that is, if the track confidence scores of the first 4 frames are calculated using an exponentially weighted average, then... .

[0108] ;

[0109] The system identifies the flight path as a multipath false target, thus enabling the deletion of multipath false targets.

[0110] According to the embodiments of this disclosure, the following technical effects are achieved: First, by performing Fourier transform and constant false alarm rate detection on the single-frame echo signal, precise trace information including distance, speed, azimuth angle, and energy value is obtained, providing a reliable data foundation for multipath analysis. Then, multipath pattern judgment is performed at the single-frame level, and traces that strictly conform to the multipath equation are immediately deleted, directly reducing the number of false targets and subsequent computational load. For uncertain traces within the redundancy range, multipath confidence is introduced for quantification and marking, improving the adaptability of single-frame processing. Then, traces with confidence are associated to form tracks. By calculating the arithmetic mean of the single-frame confidence and then performing an exponential weighted average on multiple consecutive frames, a comprehensive evaluation value is generated. Finally, by comparing the comprehensive evaluation value with a threshold, multipath false target tracks are reliably identified and eliminated. Through the close connection from precise trace extraction to flexible single-frame judgment and then to robust multi-frame fusion, accurate multipath judgment can be performed under actual complex road conditions while ensuring low computational complexity.

[0111] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0112] The above is an introduction to the method embodiments. The following describes the solution of this application further through device embodiments.

[0113] Figure 3 A block diagram of a multipath false target detection and identification device involving an automotive millimeter-wave radar according to an embodiment of this application is shown, such as... Figure 3 The following are included:

[0114] The echo signal acquisition module 201 is used to acquire a single frame echo signal of the target vehicle-mounted millimeter-wave radar;

[0115] The dot set generation module 202 is used to filter and process the single-frame echo signal to generate a single-frame dot set;

[0116] The multipath judgment module 203 is used to judge the multipath pattern of a single frame point set and generate multipath judgment results.

[0117] The target trajectory generation module 204 is used to associate the single-frame over-detected points in the single-frame point set with the multipath judgment result and preset rules to generate a target trajectory.

[0118] The comprehensive evaluation generation module 205 is used to calculate and process the multipath confidence information of the target trajectory and generate a comprehensive multipath confidence evaluation value.

[0119] The false target identification module 206 is used to perform multipath false target detection and identification processing based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

[0120] As an optional implementation of this embodiment, the point set generation module 202 is specifically used to perform Fourier transform processing on the single-frame echo signal to generate a two-dimensional frequency domain matrix; perform constant false alarm rate detection in the two-dimensional frequency domain matrix to filter out single-frame over-detected points; calculate the target calculation data of each single-frame over-detected point, and sort the single-frame over-detected points according to the point distance to generate a single-frame point set, wherein the target calculation data includes distance, velocity, azimuth angle and energy value.

[0121] As an optional implementation of this embodiment, the multipath determination module 203 includes:

[0122] The rule satisfaction judgment module is used to determine whether the single-frame over-checked point trace in the single-frame point trace set satisfies the preset multipath law equation;

[0123] The false detection and deletion module is used to identify and delete over-detected points in a single frame as multipath false targets.

[0124] The multipath confidence assignment module is used to assign a multipath confidence level to a single frame over-detection point trace.

[0125] In this optional implementation, the rule satisfaction judgment module is specifically used to obtain all the judgment conditions of the preset multipath law equation; to judge whether the single frame over-detection point trace meets all the judgment conditions; if the single frame over-detection point trace meets all the judgment conditions, then the single frame over-detection point trace is judged to satisfy the preset multipath law equation; if the single frame over-detection point trace does not meet all the judgment conditions, then the single frame over-detection point trace is judged not to satisfy the preset multipath law equation.

[0126] In this optional embodiment, the multipath confidence assignment module is specifically used when the number of judgment conditions met by a single-frame over-detection trace is a first data value; based on the target calculated data, the single-frame over-detection trace undergoes threshold relaxation processing, and it is determined whether the single-frame over-detection trace with relaxed threshold processing meets all judgment conditions; if the single-frame over-detection trace with relaxed threshold processing meets all judgment conditions, then the multipath confidence assigned to the single-frame over-detection trace is a first confidence level; if the single-frame over-detection trace with relaxed threshold processing does not meet all judgment conditions, then the multipath confidence assigned to the single-frame over-detection trace is a second confidence level; when the judgment conditions of a single-frame over-detection trace... When the number of conditions met is the second data value; based on the target calculation data, the threshold of the single-frame over-checked trace is relaxed, and it is determined whether the single-frame over-checked trace with relaxed threshold meets all the judgment conditions; if the single-frame over-checked trace with relaxed threshold meets all the judgment conditions, the multipath confidence assigned to the single-frame over-checked trace is the third confidence level; if the single-frame over-checked trace with relaxed threshold does not meet all the judgment conditions, the multipath confidence assigned to the single-frame over-checked trace is the fourth confidence level; when the number of judgment conditions met by the single-frame over-checked trace is the third data value, the multipath confidence assigned to the single-frame over-checked trace is the fifth confidence level.

[0127] As an optional implementation of this embodiment, the comprehensive evaluation generation module 205 is specifically used to calculate the arithmetic mean of the multipath confidence of all associated points in a single frame of the target track, and use the arithmetic mean as the track confidence of the over-checked points in a single frame; and to calculate the exponential weighted average of the track confidence of multiple consecutive frames to obtain the comprehensive multipath confidence evaluation value.

[0128] As an optional implementation of this embodiment, the false target identification module 206 is specifically used to compare the comprehensive multipath confidence evaluation value with a preset confidence threshold, and determine whether the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold; if the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold, the target track is determined to be a multipath false target track and is removed; if the comprehensive multipath confidence evaluation value is less than the preset confidence threshold, the target track is determined to be a real track and is retained and reported.

[0129] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0130] Figure 4 A schematic diagram of a terminal device or server suitable for implementing embodiments of this application is shown.

[0131] like Figure 4 As shown, the terminal device or server includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 302 or a program loaded from storage section 308 into random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the terminal device or server. The CPU 301, ROM 302, and RAM 303 are interconnected via bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0132] The following components are connected to I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to I / O interface 305 as needed. A removable medium 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 310 as needed so that computer programs read from it can be installed into storage section 308 as needed.

[0133] Specifically, according to embodiments of this application, the above method flow steps can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a machine-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the functions defined in the system of this application.

[0134] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0135] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0136] The units or modules described in the embodiments of this application can be implemented in software or hardware. The described units or modules can also be located in a processor. The names of these units or modules do not, in certain circumstances, constitute a limitation on the unit or module itself.

[0137] In another aspect, this application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the methods described in this application.

[0138] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A method for detecting and identifying multipath false targets using vehicle-mounted millimeter-wave radar, characterized in that, include: Acquire a single frame echo signal from the target vehicle's millimeter-wave radar; The single-frame echo signal is filtered to generate a single-frame point set; Perform multipath pattern judgment on the single-frame point trace set and generate multipath judgment result; determine whether the single-frame over-detected point trace in the single-frame point trace set satisfies the preset multipath pattern equation; If the single-frame over-detection point satisfies the preset multipath law equation, then the single-frame over-detection point is determined to be a multipath false target and deleted. If the single-frame over-detection point trace does not satisfy the preset multipath law equation but is within the redundancy range of the preset multipath law equation, then a multipath confidence score is assigned to the single-frame over-detection point trace; Based on the multipath judgment result and preset rules, the single-frame over-detected points in the single-frame point set are associated with the track to generate the target track; The target trajectory is processed by multipath confidence information calculation to generate a comprehensive multipath confidence evaluation value; Calculate the arithmetic mean of the multipath confidence of all associated points in a single frame for the target track, and use the arithmetic mean as the track confidence of the over-detected points in the single frame; An exponentially weighted average of the track confidence scores across multiple consecutive frames is used to obtain a comprehensive multipath confidence evaluation value. Multipath false target detection and identification processing is performed based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

2. The method according to claim 1, characterized in that, The step of filtering the single-frame echo signal to generate a single-frame point set includes: The single-frame echo signal is subjected to Fourier transform processing to generate a two-dimensional frequency domain matrix; Constant false alarm rate detection is performed in the two-dimensional frequency domain matrix to filter out single-frame over-detection points. Calculate the target calculation data for each of the single-frame over-detection points, and sort the single-frame over-detection points according to the point distance to generate a single-frame point set, wherein the target calculation data includes distance, velocity, azimuth angle and energy value.

3. The method according to claim 2, characterized in that, The step of determining whether the over-detected single-frame point in the single-frame point set satisfies the preset multipath law equation includes: Obtain all the judgment conditions of the preset multipath law equation; Determine whether the single-frame over-detection point meets all the determination conditions; If the single-frame over-detection point meets all the judgment conditions, then the single-frame over-detection point is determined to satisfy the preset multipath law equation. If the single-frame over-detection point does not meet all the judgment conditions, then it is determined that the single-frame over-detection point does not satisfy the preset multipath law equation.

4. The method according to claim 3, characterized in that, Assigning a multipath confidence score to the single-frame over-detection point trace includes: When the number of the judgment conditions for the single-frame over-detection traces meets the first data quantity value; Based on the target calculation data, the threshold of the single-frame over-detection point is relaxed, and it is determined whether the single-frame over-detection point of the relaxed threshold process satisfies all the judgment conditions. If the single-frame over-detection trace processed by relaxing the threshold satisfies all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the first confidence level. If the single-frame over-detection trace processed by relaxing the threshold does not meet all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the second confidence level. When the number of the judgment conditions for the single-frame over-detection traces meets the second data quantity value; Based on the target calculation data, the threshold of the single-frame over-detection point is relaxed, and it is determined whether the single-frame over-detection point of the relaxed threshold process satisfies all the judgment conditions. If the single-frame over-detection trace processed by relaxing the threshold satisfies all the above judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the third confidence level. If the single-frame over-detection trace processed by relaxing the threshold does not meet all the judgment conditions, then the multipath confidence level assigned to the single-frame over-detection trace is the fourth confidence level. When the number of over-detection points in a single frame meets the judgment condition is the third data value, the multipath confidence level assigned to the over-detection point in the single frame is the fifth confidence level.

5. The method according to claim 1, characterized in that, The multipath false target detection and identification process based on the comprehensive multipath confidence evaluation value and the preset confidence threshold includes: The comprehensive multipath confidence evaluation value is compared with a preset confidence threshold to determine whether the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold. If the comprehensive multipath confidence evaluation value is not less than the preset confidence threshold, then the target track is determined to be a multipath false target track and is removed. If the comprehensive multipath confidence evaluation value is less than the preset confidence threshold, the target trajectory is determined to be a real trajectory and is retained for reporting.

6. A multipath false target detection and identification device for vehicle-mounted millimeter-wave radar, characterized in that, include: The echo signal acquisition module is used to acquire a single frame echo signal from the target vehicle-mounted millimeter-wave radar. The dot set generation module is used to filter and process the single-frame echo signal to generate a single-frame dot set. The multipath judgment module is used to perform multipath pattern judgment on the single-frame point trace set and generate a multipath judgment result; and to determine whether the single-frame over-detected point trace in the single-frame point trace set satisfies the preset multipath pattern equation. If the single-frame over-detection point satisfies the preset multipath law equation, then the single-frame over-detection point is determined to be a multipath false target and deleted. If the single-frame over-detection point trace does not satisfy the preset multipath law equation but is within the redundancy range of the preset multipath law equation, then a multipath confidence score is assigned to the single-frame over-detection point trace; The target trajectory generation module is used to associate the single-frame over-detected points in the single-frame point set with the multipath judgment result and preset rules to generate a target trajectory. The comprehensive evaluation generation module is used to calculate and process the multipath confidence information of the target trajectory and generate a comprehensive multipath confidence evaluation value. Calculate the arithmetic mean of the multipath confidence of all associated points in a single frame for the target track, and use the arithmetic mean as the track confidence of the over-detected points in the single frame; An exponentially weighted average of the track confidence scores across multiple consecutive frames is used to obtain a comprehensive multipath confidence evaluation value. The false target identification module is used to perform multipath false target detection and identification processing based on the comprehensive multipath confidence evaluation value and the preset confidence threshold.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method as described in any one of claims 1 to 5.

Images