A method for mitigating multipath interference for lane logic based on millimeter wave traffic radar

Abstract

The application discloses a kind of lane logic multipath interference suppression method and system based on millimeter wave traffic radar, wherein the method comprises: obtaining the point cloud data collected by millimeter wave traffic radar;The point cloud data is clustered to obtain point cloud clustering result, and the point cloud clustering result is condensed to obtain condensed point trace;Load high-definition map file, extract map data, and create lane model based on the map data;The condensed point trace is associated with the current managed track or historical point trace to form a new track;Based on the new track, update the lane model or filter out the logically unreasonable track to suppress multipath interference.

Description

Technical Field

[0001] This invention relates to the field of millimeter-wave radar in the transportation sector, and more specifically, to a multipath interference suppression method based on lane logic of millimeter-wave traffic radar. Background Technology

[0002] Millimeter-wave radar is a high-precision measuring instrument used in the transportation field, widely applied in vehicle trajectory detection, traffic flow statistics across road sections, and incident detection on highways. Because electromagnetic waves are scattered by structures such as guardrails, signs, noise barriers, lampposts, and tunnels around the road, especially when radar-emitted electromagnetic waves are scattered by vehicles and then refracted back to the radar receiving antenna, the multipath effect causes false targets, known as "ghosting." The multipath effect is caused by reflection, refraction, and scattering of the target's scattered signal during propagation, resulting in the radar receiving multiple signals from different paths. In real-world environments, the echo signal from millimeter-wave traffic radar includes not only the direct wave signal from the target but also multipath echo signals reflected from different structures. Multipath signals cause aliasing and interference of parameters such as the target's position, velocity, and size, thus affecting the measurement accuracy and target detection capability of millimeter-wave traffic monitoring radar.

[0003] Methods for identifying and suppressing multipath interference (ghosting) are currently mainly divided into two categories: model-based methods and data-based methods. Model-based methods exhibit good robustness when dealing with multipath ghosting that satisfies specular reflection conditions; however, the complexity of the electromagnetic environment in real-world scenarios leads to a mismatch between the multipath ghosting and the model, severely impacting the effectiveness of the methods. In contrast, data-based methods are less affected by the environment and can handle more complex multipath ghosting, but they face challenges such as limited labeled datasets and lower recognition accuracy. Summary of the Invention

[0004] According to the present invention, a multipath interference suppression method and system based on millimeter-wave traffic radar lane logic is provided to solve the technical problems of limited labeled datasets and low recognition accuracy.

[0005] According to a first aspect of the present invention, a multipath interference suppression method based on millimeter-wave traffic radar lane logic is provided, comprising: Acquire point cloud data collected by millimeter-wave traffic radar; Cluster the point cloud data to obtain point cloud clustering results, then aggregate the point cloud clustering results to obtain aggregated point traces; Load a high-resolution map file, extract map data, and create a lane model based on the map data; The aggregated points are associated with the currently managed track or historical points to form a new track; Based on this, new trajectories can be used to update lane models or filter out logically illogical trajectories to suppress multipath interference.

[0006] Optionally, it also includes preprocessing operations such as filtering, denoising, and extracting outliers from the point cloud data.

[0007] Optionally, the point cloud data includes the target's position information in the X and Y directions, radial distance, radial velocity, and scattering power in the radar coordinate system.

[0008] Optionally, loading a high-resolution map file, extracting map data, and creating a lane model based on the map data includes: Load a high-resolution map file, extract a section from the high-resolution map file, and obtain the point set from the high-resolution map file; The point sets in the high-resolution map file based on the WGS84 coordinate system are converted to the North-Sky-East coordinate system, and the points in the North-Sky-East coordinate system are converted to the radar coordinate system. The conversion formula is as follows: Right now in , , Here, E, N, and U are the local radar coordinates, E, N, and U are the northeast-sky coordinates, and L is the transformation matrix. The deflection angle and its value is equal to , This is the radar's local north-north angle; The point set data of each lane in the radar field of view is extracted based on the lane attributes in the high-definition map file, and the start position, end position and lane dividing line distance are read by reading the configuration file. Starting from the initial position, draw a dividing line at regular intervals until the end position. Combined with the lane lines on the map, divide the map in the radar field of view into several areas, label them, and create a spatial model of the lanes.

[0009] Optionally, the aggregated points can be associated with currently managed tracks or historical points to form a new track, including: Determine whether the agglomerated point falls within the tracking gate. The tracking gate is the area centered on the latest state of the current managed track, with distance and speed as the dimensions. That is, the horizontal and vertical distances between the agglomerated point and the center point meet the set threshold values, and the interpolation of the speed and the center point meets the set threshold values. If the agglomerated point falls into the tracking gate, the agglomerated point is associated with the track. If the association is successful, a new track is formed. Otherwise, the agglomerated point is associated with the historical point to form a new track. If the trace does not fall within the tracking gate, the process returns to the next trace for evaluation.

[0010] Optionally, based on the new trajectory, updating the lane model or filtering out logically illogical trajectories to suppress multipath interference includes: If the new track is successfully associated with the old track, then update the track information and the lane model. If the new track fails to associate with the old track, multipath interference is suppressed by filtering out logically illogical tracks.

[0011] Optionally, it also includes: Filter out logically illogical tracks, update the existing tracks with the retained tracks, update the track information and lane information in the lane model, and output the track.

[0012] According to another aspect of the present invention, a multipath interference suppression system based on millimeter-wave traffic radar lane logic is also provided, comprising: The acquisition module is used to acquire point cloud data collected by millimeter-wave traffic radar; The aggregation module is used to cluster the point cloud data to obtain point cloud clustering results, and aggregate the point cloud clustering results to obtain aggregated point traces. A module is created to load high-definition map files, extract map data, and create lane models based on the map data. The association module is used to associate the aggregated points with currently managed tracks or historical points to form a new track; The suppression module is used to suppress multipath interference by updating the lane model or filtering out logically illogical tracks based on the new track.

[0013] According to another aspect of the invention, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method as described in any one of the claims.

[0014] According to another aspect of the invention, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method as described in any one of the claims.

[0015] Therefore, this invention provides a multipath interference suppression method based on millimeter-wave traffic radar lane logic. It associates the aggregated points with currently managed tracks or historical points to form new tracks. Based on these new tracks, the lane model is updated or logically flawed tracks are filtered out, thereby suppressing multipath interference. This method is less affected by environmental factors and has advantages such as high efficiency, high reliability, and ease of implementation. It can effectively suppress multipath interference, thereby improving the accuracy of target detection. Attached Figure Description

[0016] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures: Figure 1 This is a flowchart illustrating a multipath interference suppression method based on millimeter-wave traffic radar for lane logic as described in this embodiment. Figure 2 This is a schematic diagram of the structure of a multipath interference suppression system based on lane logic of millimeter-wave traffic radar as described in this embodiment; Figure 3 This is a schematic diagram illustrating the workflow of a multipath interference suppression system based on millimeter-wave traffic radar lane logic as described in this embodiment. Figure 4 This is a schematic diagram of the overall workflow of a multipath interference suppression method based on lane logic of millimeter-wave traffic radar as described in this embodiment. Figure 5 This is a schematic diagram of the North-East coordinate transformation and radar coordinate system described in this embodiment; Figure 6 This is a schematic diagram of the lane model described in this embodiment; Figure 7 This is a schematic diagram illustrating the calculation of the occlusion coefficient and control range described in this embodiment; Figure 8 This is a schematic diagram illustrating the calculation of the currently occupied cell of the track as described in this embodiment; Figure 9 This is a schematic diagram illustrating the principle of the sliding window algorithm described in this embodiment; Figure 10 This is a schematic diagram of the track and track association described in this embodiment; Figure 11 This is a schematic diagram of the forward multi-track competition lanes described in this embodiment; Figure 12 This is a schematic diagram of the reverse multi-track competition lane described in this embodiment; Figure 13 This is a schematic diagram of a multipath interference suppression system based on millimeter-wave traffic radar lane logic as described in this embodiment. Detailed Implementation

[0017] Exemplary embodiments of the invention will now be described with reference to the accompanying drawings. However, the invention may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the invention and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the invention. In the drawings, the same units / elements are referred to by the same reference numerals.

[0018] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.

[0019] According to a first aspect of the present invention, a multipath interference suppression method 100 based on millimeter-wave traffic radar lane logic is provided, with reference to... Figure 1 As shown, the method 100 includes: S101: Acquire point cloud data collected by millimeter-wave traffic radar; S102: Cluster the point cloud data to obtain point cloud clustering results, and agglomerate the point cloud clustering results to obtain agglomerated point traces; S103: Load the high-definition map file, extract the map data, and create a lane model based on the map data; S104: Associate the aggregated points with the currently managed track or historical points to form a new track; S105: Based on the new trajectory, update the lane model or filter out logically illogical trajectories to suppress multipath interference.

[0020] Specifically, to address at least one of the problems described in the background section, this invention provides a multipath interference suppression method based on millimeter-wave traffic radar and lane logic. By associating aggregated data points with currently managed tracks or historical data points to form new tracks, and then updating the lane model or filtering out logically flawed tracks based on these new tracks, multipath interference is suppressed.

[0021] Operating mode configuration: This includes configuring system operating parameters, radio frequency module parameters, data preprocessing parameters, point clustering and agglomeration parameters, track management module parameters, and track output parameters. Workflow reference. Figure 2 and Figure 3 As shown.

[0022] Point cloud data acquisition module: This module processes and converts the radar echo signals received by the radio frequency module into usable point cloud data. The acquired radar point cloud data mainly includes the target's position information in the X and Y directions, radial distance R, radial velocity V, and scattering power P in the radar coordinate system.

[0023] Point cloud data preprocessing module: performs preliminary processing on the data, including filtering, noise reduction, and removal of outlier data.

[0024] Point cloud data clustering and agglomeration: When a radar detects a target, it receives several point cloud data points reflected from the same target. If tracking filtering is applied to each point cloud, a single target will generate a large number of false tracks, affecting the target's tracking performance. Therefore, it is necessary to merge all point cloud data points of the same target into a single point cloud set; this process is called point cloud clustering. In this embodiment of the invention, the DBSCAN clustering algorithm is used to perform point cloud clustering. The DBSCAN clustering algorithm is a commonly used technique in this field.

[0025] If the aforementioned point cloud dataset is directly input for correlation processing without any processing, it will not only increase the computational load but also potentially result in multiple tracks for a single target. Since these tracks belong to the same target and are close in both distance and velocity dimensions, they will compete for points, leading to track interruptions and preventing stable tracking. Therefore, it is necessary to preprocess the measurement information and utilize point aggregation to consolidate multiple points from the same target into a single target point track, thereby reducing data processing complexity and improving target tracking stability. In this embodiment of the invention, point aggregation is achieved through the centroid algorithm.

[0026] Track Management Module: Loads high-precision map files, extracts map data, creates lane models, and first associates the aggregated points with the track. If the association is successful, the track is updated. For points that are not successfully associated, the association calculation between the points is performed. If a new track is formed, the logical rationality of the track is evaluated in conjunction with the lane model. Multipath interference is suppressed by filtering out logically illogical tracks.

[0027] Track output module: Outputs track information through an interface.

[0028] In the practical environment described in this invention, the echo signal of millimeter-wave traffic radar includes not only the direct wave signal of the target, but also multipath echo signals reflected from different buildings. Multipath signals can cause aliasing and interference of parameters such as the target's position, velocity, and size, thereby affecting the measurement accuracy and target detection capability of the millimeter-wave traffic monitoring radar.

[0029] refer to Figure 4 As shown, the specific implementation process is as follows: Step 1: Load the high-resolution map file. Using the latitude and longitude coordinates of the radar installation location as the origin and the radar's scanning range as the radius, the high-resolution map file is cropped and processed. The processing involves first converting the point set in the high-resolution map file (based on the WGS84 coordinate system) to the North-Eastern (ENU) coordinate system. This conversion method is a commonly used technique in this field. Then, the points in the ENU coordinate system are converted to the radar coordinate system (a Cartesian coordinate system with the radar array normal to the Y-axis). Figure 5 As shown. The conversion formula for transforming the north-sky-east coordinate system to the radar coordinate system is as follows: Right now in , , Here, E, N, and U are the local radar coordinates, E, N, and U are the northeast-sky coordinates, and L is the transformation matrix. The deflection angle and its value is equal to , This is the radar's local north-north angle.

[0030] Based on the lane attributes in the map data, the point set data of each lane in the radar field of view is extracted. Parameters such as the start and end positions and the distance to lane dividers are read from the configuration file. Starting from the start position, a lane divider line is drawn at regular intervals until the end position. Combined with the lane lines in the map, the map in the radar field of view is divided into several regions, which are then labeled to create a spatial model of the lanes. Figure 6 As shown, in this embodiment, the distance between lane dividing lines is 5m, the starting position is 50m, and the ending position is determined according to the maximum detection range of the radar. In this embodiment, a 500m millimeter-wave traffic radar is used, and the ending position is 450m.

[0031] Step 2: Obtain the aggregated point data and determine whether the point falls within the tracking gate. If not, return and then determine the next point. Otherwise, associate the point with the existing track. If the association is successful, update the track information, which mainly includes track length, track density, track instantaneous speed, track average speed, track direction, track average acceleration, occlusion coefficient, control range, and track confidence.

[0032] Among them, the tracking gate is the region centered on the latest state of the current managed track, with distance and speed as the dimensions. That is, the horizontal and vertical distances between the aggregated track and the center point meet the set threshold values, and the interpolation of the speed and the center point meets the set threshold values.

[0033] Among them, the track length is the distance between the track at the farthest point from the radar and the track at the nearest point. Y The difference in coordinates of the direction, i.e. Track density The number of all points contained in the track; the instantaneous velocity of the track. The velocity of the most recently associated point in the track; the track average velocity is the average velocity of all points in the track, i.e. ; flight path direction The value is +1 when the target is far from the radar and -1 when it is close to the radar; the average acceleration of the track is the velocity at the current point of the track. Speed ​​relative to the starting point of the track The difference between the two points and the time difference between them The ratio, i.e. ; Occlusion coefficient From the height of the target The ratio of the radar installation height to the radar installation height. The target's height information is derived from its length through experience. Control range. , where y is the current position Y The coordinates of the direction. For the above occlusion coefficient, such as Figure 7 As shown. The cell occupied by the target in the road model is determined based on the target's current location and control range, such as... Figure 8 As shown, the current cells occupied by the track are 1-1 and 1-2.

[0034] track confidence Initial trajectory confidence level of the flight path Confidence of process trajectory Speed ​​confidence The sum of, i.e. .

[0035] Among them, the initial trajectory confidence for: When the trajectory is positive, the calculation formula is: When the flight path is reversed, the calculation formula is: Process trajectory confidence for: Where, is the starting point of the trajectory. Y The coordinates of the direction, when hour, ,when hour, ; The preset starting position of the track in step 1, The preset end position of the track in step 1; For the length of the track, when hour, ; The direction of the flight path; For track density, As the trajectory stability threshold, when hour, .

[0036] Speed ​​confidence for: in, The instantaneous velocity of the trajectory described in step 2; The average speed of the trajectory as described in step 2; Let be the variance of all points in the trajectory in the velocity dimension; when hour, .

[0037] In the above, To preset the initial reliability parameter, it is set to 20 in this embodiment; The preset process reliability parameter is 70 in this embodiment; The velocity confidence parameter is set to 10 in this embodiment; where =100.

[0038] Then, the basic information of each lane in the lane logic model is updated, mainly including lane density, lane occupancy range, average lane speed, average lane acceleration, and average track confidence. Here, lane density *m* represents the number of tracks within the lane, and lane occupancy range is the current distance of the track furthest from the radar within the lane. Y The current position of the direction and the nearest track Y The position of direction, i.e. When there is only one track, ,in The length of the target; the average lane speed is the ratio of the sum of the products of the average speeds and directions of all paths within the lane to the lane density, i.e. Lane-average acceleration is the ratio of the sum of the products of the average acceleration of all tracks within a lane and their direction to the lane density, i.e. The lane average confidence level is the ratio of the sum of the products of the confidence levels of all tracks within a lane and their directions to the lane density. .

[0039] Step 3: For points that failed to be associated with existing tracks in Step 2, associate them with each other. This involves associating the current point with points from previous times. A successful association forms a new track. In track initiation algorithms, the most widely used in engineering is the "sliding window" m / n logical nearest neighbor initiation method. This method has low computational cost and is easy to implement in engineering. The "sliding window" method processes the current time and several consecutive historical periods of point sequences that have failed to be associated with existing tracks. The basic idea of ​​this algorithm is: taking the current time as the endpoint, if the number of successful associations between points is greater than or equal to m in n consecutive periods, a new track is output. In engineering applications, n and m are often 3 and 2, respectively. The principle is as follows: Figure 9 As shown.

[0040] Step 4: Perform multi-cycle correlation calculations between the new track formed in Step 3 and existing tracks. The main purpose is to address the issue of multiple tracks failing to correlate when the same target's track is interrupted due to obstruction or other reasons, and the target is re-detected stably. (Reference) Figure 10 As shown, the old track is forward extrapolated to the start time of the new track, and the new track is backward extrapolated to the end time of the old track. The data from the two tracks are then correlated. If the points in multiple consecutive cycles fall within the correlation threshold, the new track is considered successfully correlated with the old track. The track information is then updated, followed by updating the track information within each lane of the lane model.

[0041] Step 5: For new tracks that failed to associate with existing tracks in Step 4, obtain basic lane information from the lane model based on the lane number of the new track. Transform the rationality judgment problem of the new track into a problem of multiple tracks competing for lanes, mainly including the following situations: like Figure 11 In scenario 1, the new track appears within the cell range occupied by an existing track, indicating that multiple tracks are competing for the same space. This can be determined by comparing the confidence levels of the existing and new tracks. To filter tracks with low confidence; like Figure 11 In scenario 2, if track information exists within the lane, and the new track appears after the last track, or if no track information exists within the lane, the starting point of the new track is after the starting position set in step 1. If a rectangle is drawn with the current position of the track as the upper left point and the radar position as the lower right point, and there are no other targets within the rectangular area, it indicates that there is no obstruction, and the track is directly filtered out. Otherwise, it indicates that there may be lateral obstruction. If the new track is interrupted and no longer updated before the confidence level meets the set threshold, it is considered that the new track is caused by interference and is directly filtered out; otherwise, the track is retained. like Figure 11In case 3, regardless of whether there is track information in the lane, if the confidence of the new track is greater than the set threshold before the starting position set in step 1, and there is no competition or obstruction, the track is retained. like Figure 11 In scenario 4, track information exists within the lane, and a new track appears before the current position of an existing track. The target altitude of the first existing track following the new track... If the value is less than the set threshold, it means that the target could not possibly block a target until the blocked target escapes the blocking area before being detected by the radar. Therefore, the appearance of the new track is illogical, and the track is directly filtered out in this embodiment. Otherwise, it means that the new track may have been detected just because it was blocked. If the new track is interrupted and no longer updated before the confidence level meets the set threshold, it is considered to be caused by interference and is directly filtered out. If the set threshold is met, the track is considered to be a real target and is retained. like Figure 12 In case 5, there is a track information in the lane. The starting point of the new track is before the ending position set in step 1, and the direction is reverse. Since there is no obstruction for the reverse moving target at this time, the occurrence of this track is unreasonable, and we consider filtering out the track directly. like Figure 12 In scenario 6, there is a track information within the lane. The starting point of the new track is after the ending position set in step 1, and the direction is reverse. At this time, the feasibility of a reverse vehicle is analyzed based on the average speed, average acceleration, and lane occupancy range of the track within the lane. For example, a reverse vehicle is considered to have appeared only when the average speed in the lane information in step 2 tends to be less than a preset threshold, the decrease in lane acceleration exceeds a preset threshold, and the lane occupancy length is less than a preset threshold. Otherwise, it is illogical for a reverse vehicle to appear in the lane, and it can be directly filtered out. like Figure 12 In case 7, there is no track information in the lane. The starting point of the new track is after the ending position set in step 1, and the direction is reverse. When the confidence of the new track is greater than the set threshold, the track is retained and it is considered that a vehicle is traveling in the opposite direction. like Figure 12 In scenario 8, where no track information exists within the lane, the starting point of the new track is before the termination position set in step 1, with the direction reversed. A rectangle is drawn with the current position of the track as the upper left point and the radar position as the lower right point. If there are no other targets within the rectangular area, it indicates no obstruction, and the track is directly filtered out. Otherwise, it indicates possible lateral obstruction. If the new track is interrupted and no longer updated before the confidence level meets the set threshold, it is considered to be caused by interference and is directly filtered out; otherwise, the track is retained.

[0042] By using logical relationships within the lanes to filter newly emerging tracks, tracks generated by multipath interference can be effectively removed.

[0043] Step 6: Delete the tracks that were determined to be filtered out in Step 5, and update the tracks that were determined to be retained into the existing tracks. Simultaneously update the track information and lane information in the lane model, and finally output the tracks.

[0044] Therefore, this invention provides a multipath interference suppression method based on millimeter-wave traffic radar lane logic. It associates the aggregated points with currently managed tracks or historical points to form new tracks. Based on these new tracks, the lane model is updated or logically flawed tracks are filtered out, thereby suppressing multipath interference. This method is less affected by environmental factors and has advantages such as high efficiency, high reliability, and ease of implementation. It can effectively suppress multipath interference, thereby improving the accuracy of target detection.

[0045] According to another aspect of the present invention, a multipath interference suppression system 1300 based on millimeter-wave traffic radar lane logic is also provided, with reference to... Figure 13 As shown, the system 1300 includes: The acquisition module 1310 is used to acquire point cloud data collected by millimeter-wave traffic radar; The aggregation module 1320 is used to cluster the point cloud data to obtain point cloud clustering results, aggregate the point cloud clustering results, and obtain aggregated point traces. Module 1330 is used to load high-definition map files, extract map data, and create lane models based on the map data. The association module 1340 is used to associate the aggregated points with the currently managed track or historical points to form a new track; The suppression module 1350 is used to suppress multipath interference by updating the lane model or filtering out logically illogical tracks based on the new track.

[0046] The multipath interference suppression system 1300 based on lane logic of millimeter-wave traffic radar in one embodiment of the present invention corresponds to the multipath interference suppression method 100 based on lane logic of millimeter-wave traffic radar in another embodiment of the present invention, and will not be described again here.

[0047] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0048] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0049] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0051] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0052] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A multipath interference suppression method based on lane logic of millimeter-wave traffic radar, characterized in that, include: Acquire point cloud data collected by millimeter-wave traffic radar; Cluster the point cloud data to obtain point cloud clustering results, then aggregate the point cloud clustering results to obtain aggregated point traces; Load a high-resolution map file, extract map data, and create a lane model based on the map data; The aggregated points are associated with the currently managed track or historical points to form a new track; Based on the new trajectory, the lane model is updated or logically illogical trajectories are filtered out to suppress multipath interference. The point set data of each lane in the radar field of view is extracted based on the lane attributes in the high-definition map file, and the start position, end position and lane dividing line distance are read by reading the configuration file. Starting from the starting position, draw a dividing line at regular intervals until the ending position. Combined with the lane lines on the map, divide the map in the radar field of view into several areas, label them, and create a spatial model of the lanes. The aggregated points are associated with currently managed tracks or historical points to form a new track, including: Determine whether the agglomerated point falls within the tracking gate. The tracking gate is the area centered on the latest state of the current managed track, with distance and speed as the dimensions. That is, the horizontal and vertical distances between the agglomerated point and the center point meet the set threshold values, and the interpolation of the speed and the center point meets the set threshold values. If the agglomerated point falls into the tracking gate, the agglomerated point is associated with the track. If the association is successful, a new track is formed. Otherwise, the agglomerated point is associated with the historical point to form a new track. If the trace does not fall within the tracking gate, the process returns to the next trace for evaluation.

2. The method according to claim 1, characterized in that, Also includes: Preprocessing operations such as filtering, noise reduction, and outlier extraction are performed on point cloud data.

3. The method according to claim 1, characterized in that, The point cloud data includes the target's position information in the X and Y directions, radial distance, radial velocity, and scattering power in the radar coordinate system.

4. The method according to claim 1, characterized in that, Loading a high-resolution map file, extracting map data, and creating a lane model based on the map data includes: Load a high-resolution map file, extract a section from the high-resolution map file, and obtain the point set from the high-resolution map file; The point sets in the high-resolution map file based on the WGS84 coordinate system are converted to the North-Sky-East coordinate system, and the points in the North-Sky-East coordinate system are converted to the radar coordinate system. The conversion formula is as follows: Right now in , , Here, E, N, and U are the local radar coordinates, E, N, and U are the northeast-sky coordinates, and L is the transformation matrix. The deflection angle and its value is equal to , This is the radar's local north-north angle.

5. The method according to claim 1, characterized in that, Based on the new trajectory, update the lane model or filter out logically illogical trajectories to suppress multipath interference, including: If the new track is successfully associated with the old track, then update the track information and the lane model. If the new track fails to associate with the old track, multipath interference is suppressed by filtering out logically illogical tracks.

6. The method according to claim 5, characterized in that, Also includes: Filter out logically illogical tracks, update the existing tracks with the retained tracks, update the track information and lane information in the lane model, and output the track.

7. A multipath interference suppression system based on lane logic of millimeter-wave traffic radar, characterized in that, include: The acquisition module is used to acquire point cloud data collected by millimeter-wave traffic radar; The aggregation module is used to cluster the point cloud data to obtain point cloud clustering results, and aggregate the point cloud clustering results to obtain aggregated point traces. A module is created to load high-definition map files, extract map data, and create lane models based on the map data. The association module is used to associate the aggregated points with currently managed tracks or historical points to form a new track; The suppression module is used to suppress multipath interference by updating the lane model or filtering out logically illogical tracks based on the new track. The point set data of each lane in the radar field of view is extracted based on the lane attributes in the high-definition map file, and the start position, end position and lane dividing line distance are read by reading the configuration file. Starting from the starting position, draw a dividing line at regular intervals until the ending position. Combined with the lane lines on the map, divide the map in the radar field of view into several areas, label them, and create a spatial model of the lanes. The aggregated points are associated with currently managed tracks or historical points to form a new track, including: Determine whether the agglomerated point falls within the tracking gate. The tracking gate is the area centered on the latest state of the current managed track, with distance and speed as the dimensions. That is, the horizontal and vertical distances between the agglomerated point and the center point meet the set threshold values, and the interpolation of the speed and the center point meets the set threshold values. If the agglomerated point falls into the tracking gate, the agglomerated point is associated with the track. If the association is successful, a new track is formed. Otherwise, the agglomerated point is associated with the historical point to form a new track. If the trace does not fall within the tracking gate, the process returns to the next trace for evaluation.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1-6.

Images