Dynamic calibration method, device, equipment and medium for roadside millimeter wave radar

Abstract

The present disclosure provides a method, device, equipment, and medium for dynamic calibration of roadside millimeter wave radar, which belongs to the field of radar calibration technology. The method includes: obtaining the radar coordinates of the calibration vehicle based on first data, and obtaining the longitude and latitude coordinates of the calibration vehicle based on second data, wherein the first data is data acquired by the millimeter wave radar, and the second data is data acquired by the navigation of the calibration vehicle; correlating the radar coordinates and the longitude and latitude coordinates based on the similarity of the first data and the second data; transforming the radar coordinates based on the longitude and latitude coordinates to obtain a calibration matrix; and obtaining the target longitude and latitude coordinates of the target vehicle based on the calibration matrix. The method, device, equipment, and medium for dynamic calibration of roadside millimeter wave radar provided by the present disclosure can improve the accuracy of radar positioning data.

Description

Technical Field

[0001] The present disclosure belongs to the field of radar calibration technology, and more specifically, relates to a dynamic calibration method, device, equipment, and medium for a roadside millimeter-wave radar. Background Art

[0002] In the field of smart transportation, roadside perception technology can complete functions such as real-time monitoring of traffic conditions and traffic flow simply by installing millimeter-wave radars on the roadside. Compared with traditional vehicle-mounted radars, roadside perception can expand the perception blind spots of single-vehicle intelligence and achieve the function of global road perception. Through the analysis and processing of traffic big data, it is helpful for intelligent traffic control and flow management. The radar can obtain the relative position of the target from the position where the radar is installed. In order to conduct technical research such as radar-vision fusion traffic target tracking, or to further obtain the WGS-84 coordinates of the target detected by the radar, the calibration process is essential. Most traditional calibration methods use static calibration, that is, calibrating the radar at a fixed position. Although this method is simple, in a dynamic environment, the traditional calibration method has the problem of poor data accuracy. Summary of the invention

[0003] The purpose of the present disclosure is to provide a dynamic calibration method, device, equipment, and medium for roadside millimeter-wave radar to improve the accuracy of radar positioning data.

[0004] According to a first aspect of an embodiment of the present disclosure, a dynamic calibration method for a roadside millimeter-wave radar is provided, comprising:

[0005] Obtaining radar coordinates of the calibrated vehicle based on first data, and obtaining longitude and latitude coordinates of the calibrated vehicle based on second data, wherein the first data is data acquired by a millimeter-wave radar, and the second data is data acquired by a navigation system of the calibrated vehicle;

[0006] associating the radar coordinates with the latitude and longitude coordinates based on similarity between the first data and the second data;

[0007] Transforming the radar coordinates based on the latitude and longitude coordinates to obtain a calibration matrix;

[0008] The target latitude and longitude coordinates of the target vehicle are obtained based on the calibration matrix.

[0009] A second aspect of the embodiments of the present disclosure provides a dynamic calibration device for a roadside millimeter-wave radar, including:

[0010] A data acquisition module, used to obtain the radar coordinates of the calibration vehicle based on first data, and obtain the longitude and latitude coordinates of the calibration vehicle based on second data, wherein the first data is data obtained by the millimeter wave radar, and the second data is data obtained by the navigation of the calibration vehicle;

[0011] A data association module, configured to associate the radar coordinates with the latitude and longitude coordinates based on similarity between the first data and the second data;

[0012] A data conversion module, used for transforming the radar coordinates based on the latitude and longitude coordinates to obtain a calibration matrix;

[0013] The data calibration module is used to obtain the target latitude and longitude coordinates of the target vehicle based on the calibration matrix.

[0014] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned dynamic calibration method for roadside millimeter-wave radar when executing the computer program.

[0015] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned dynamic calibration method for roadside millimeter-wave radar are implemented.

[0016] The beneficial effects of the dynamic calibration method, device, equipment, and medium for roadside millimeter-wave radar provided by the embodiments of the present disclosure are: the embodiments of the present disclosure achieve accurate acquisition of the radar coordinates and longitude and latitude coordinates of the calibrated vehicle by combining the millimeter-wave radar with the vehicle navigation data. On this basis, the two are associated by using the similarity of the data, which effectively solves the problem of coordinate system mismatch. Furthermore, the embodiments of the present disclosure accurately transform the radar coordinates through the longitude and latitude coordinates to obtain a reliable calibration matrix, which can reflect the conversion relationship between the radar coordinate system and the longitude and latitude coordinate system in real time. Finally, using this calibration matrix, the target longitude and latitude coordinates of the target vehicle can be quickly and accurately obtained, and the vehicle position can be accurately tracked and located. Thereby improving the accuracy of radar positioning data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of a flow chart of a dynamic calibration method for a roadside millimeter-wave radar provided in an embodiment of the present disclosure;

[0019] Figure 2A radar target point cloud lateral coordinate histogram provided by an embodiment of the present disclosure;

[0020] Figure 3 A schematic diagram of dynamic density clustering provided by an embodiment of the present disclosure;

[0021] Figure 4 A flowchart of the iterative process of location-guided dynamic density clustering provided by an embodiment of the present disclosure;

[0022] Figure 5 A schematic diagram of a flow chart of a dynamic calibration method for a roadside millimeter-wave radar provided in another embodiment of the present disclosure;

[0023] Figure 6 A structural block diagram of a dynamic calibration device for a roadside millimeter-wave radar provided in an embodiment of the present disclosure;

[0024] Figure 7 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0026] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0027] Please refer to Figure 1 , Figure 1 A schematic flow chart of a dynamic calibration method for a roadside millimeter-wave radar provided in an embodiment of the present disclosure, the method comprising:

[0028] S101: obtaining radar coordinates of a calibrated vehicle based on first data, and obtaining longitude and latitude coordinates of the calibrated vehicle based on second data, wherein the first data is data obtained by a millimeter wave radar, and the second data is data obtained by a navigation system of the calibrated vehicle.

[0029] In this embodiment, the roadside millimeter wave radar transmits a millimeter wave signal in a specific direction, and the signal will be reflected back after encountering an object during the propagation process. Subsequently, the radar receives the reflected signal and determines the distance, relative speed and other information between the target object and the radar by measuring the time difference, frequency change and other parameters between the transmitted signal and the received signal. Based on the installation position and angle of the radar, as well as the distance, angle and other information of the target object received, the position of the target object in a specific coordinate system with the radar as the center and a specific direction as the coordinate axis can be calculated, thereby obtaining the radar coordinates of the calibrated vehicle.

[0030] Exemplarily, data is obtained by roadside millimeter-wave radar and integrated navigation. Among them, the millimeter-wave radar data of the calibration vehicle detected by the roadside millimeter-wave radar is the first data, and the real-time dynamic positioning (RTK) data of the calibration vehicle obtained by the integrated navigation is the second data. The roadside millimeter-wave radar is installed on the roadside traffic crossbar to detect vehicles traveling on the road. The calibration vehicle is equipped with a corner reflector to obtain more accurate millimeter-wave radar data. At the same time, the calibration vehicle also collects information such as the time of the vehicle target, the latitude and longitude position of the point cloud, and the speed. The calibration vehicles of the traffic targets obtained by the roadside millimeter-wave radar include the calibration vehicles in the integrated navigation and the target vehicles traveling on the road. The calibration vehicle information collected by the roadside millimeter-wave radar includes the timestamp, the relative position of the vehicle detected by the radar, the radial velocity, the point cloud identification, and the radar scattering area.

[0031] Exemplarily, in the test, the calibration vehicle drives in N lanes in sequence. When the calibration vehicle enters the millimeter-wave radar detection area, data collection begins, and when it leaves the radar detection area, data collection stops. In this process, millimeter-wave radar data of the calibration vehicle is obtained.

[0032] In this embodiment, the roadside millimeter wave radar calculates the radar coordinates of the calibrated vehicle in a specific coordinate system centered on the radar by transmitting and receiving millimeter wave signals, combined with its own installation position and angle information; at the same time, the integrated navigation of the calibrated vehicle determines the latitude and longitude coordinates of the vehicle on the earth's surface, i.e., the second data, by acquiring RTK data. Thus, data for subsequent processing is obtained from different data sources.

[0033] S102: Associating the radar coordinates and the longitude and latitude coordinates based on the similarity between the first data and the second data.

[0034] In this embodiment, the first data (data acquired by the millimeter-wave radar) and the second data (data acquired by the calibrated vehicle navigation) need to be synchronized in time. Because the vehicle is in different positions at different times, if the data time is not synchronized, the coordinates in two different coordinate systems cannot be accurately associated. For example, ensure that the time when the millimeter-wave radar detects the vehicle is as close as possible to the time when the navigation system records the vehicle position, so that the two times correspond to the same vehicle state.

[0035] In this embodiment, millimeter-wave radar data and navigation data can be selected within the same time range according to the principle of time synchronization. Then, by comparing the spatial characteristics and movement trends of the vehicle, the possible corresponding vehicle target is found. For example, a certain threshold can be set. When the difference in parameters such as the position and speed of the vehicle in the two data is within the threshold range, it is considered that the two data describe the same vehicle. After determining the corresponding data points, the association between the radar coordinates and the longitude and latitude coordinates can be established. The radar coordinates and longitude and latitude coordinates at the same time can be recorded as a pair of data points to provide a basis for subsequent coordinate transformations. Through multiple data point pairs, the conversion relationship between the two coordinate systems can be more accurately determined.

[0036] S103: Transform the radar coordinates based on the longitude and latitude coordinates to obtain a calibration matrix.

[0037] In this embodiment, the radar coordinates are transformed by representing the longitude and latitude coordinates through matrix multiplication.

[0038] For example, assuming that the radar coordinate system is coordinate system A and the longitude and latitude coordinate system is coordinate system B, the transformation from A to B can be expressed as a matrix T, so that for any point P in the A coordinate system A , after transformation, we get point P in the B coordinate system B , satisfying P B =T•P A .

[0039] In this embodiment, the calibration matrix is ​​determined by the least square method.

[0040] For multiple known radar coordinates and corresponding longitude and latitude coordinate data point pairs, the least squares method is used to solve the matrix T that minimizes the error of all data points after transformation. The error can be defined as the difference between the transformed longitude and latitude coordinates and the actual longitude and latitude coordinates. For example, the Euclidean distance can be used to measure the error.

[0041] In the actual solution process, multiple iterations of optimization can be performed to continuously adjust the parameters of the matrix T until a certain convergence condition is met. The convergence condition can be that the error is less than a preset threshold, or the number of iterations reaches a certain value.

[0042] S104: Obtaining the target longitude and latitude coordinates of the target vehicle based on the calibration matrix.

[0043] In this embodiment, when it is necessary to determine the target longitude and latitude coordinates of the calibration vehicle, the radar coordinates of the currently detected calibration vehicle can be obtained from the roadside millimeter-wave radar. This radar coordinate is the vehicle position information expressed in a specific coordinate system centered on the radar, and may include parameters such as the relative distance and angle between the vehicle and the radar. Using the calibration matrix that has been obtained, the radar coordinates of the calibration vehicle are subjected to matrix multiplication. If the radar coordinate is a two-dimensional vector, the corresponding longitude and latitude coordinate vector can be obtained by multiplying it with the calibration matrix. After the matrix multiplication operation, the longitude and latitude coordinates obtained represent the position of the target vehicle on the surface of the earth, and the target vehicle is the vehicle detected in the actual application. The longitude value determines the position of the target vehicle in the east-west direction, and the latitude value determines the position of the target vehicle in the north-south direction.

[0044] It can be concluded from the above that this embodiment realizes the accurate acquisition of vehicle radar coordinates and longitude and latitude coordinates by combining millimeter wave radar and vehicle navigation data. On this basis, the similarity of data is used to associate the two, which effectively solves the problem of coordinate system mismatch. Furthermore, this embodiment accurately transforms the radar coordinates through longitude and latitude coordinates to obtain a reliable calibration matrix, which can reflect the conversion relationship between the radar coordinate system and the longitude and latitude coordinate system in real time. Finally, using this calibration matrix, the target longitude and latitude coordinates of the target vehicle can be quickly and accurately obtained, realizing accurate tracking and positioning of the vehicle position. Thereby improving the accuracy of radar positioning data.

[0045] In one embodiment of the present disclosure, before obtaining the radar coordinates of the calibrated vehicle based on the first data and obtaining the latitude and longitude coordinates of the calibrated vehicle based on the second data, the method further includes:

[0046] Performing point cloud preprocessing based on the third data to obtain fourth data;

[0047] The fourth data is clustered to obtain the first data.

[0048] In this embodiment, the third data is the original point cloud data directly detected by the millimeter wave radar. The millimeter wave radar can obtain information such as the distance and angle of objects in the surrounding environment to form a point cloud data set containing a large number of points. These raw data may have problems such as noise, clutter, and incomplete data. For example, the millimeter wave radar will be affected by factors such as reflections from other objects in the surrounding environment, multipath effects, and electromagnetic interference, resulting in erroneous or inaccurate points in the data.

[0049] In this embodiment, speed screening is performed based on the third data and a distribution histogram of the point cloud is established to obtain fourth data, including:

[0050] The third data is speed screened.

[0051] For example, speed screening: remove points with too low speed (remove m / s point cloud, this embodiment selects m / s), which can remove most of the noise.

[0052] Create a distribution histogram of the point cloud.

[0053] Exemplarily, a distribution histogram of the point cloud is established to reduce noise interference, while locating the road edge and dividing the tracking area.

[0054] In the point cloud data of the measured scenes, the two sides of the traffic roads are affected by trees, green belts, etc., the clutter interference within the radar field of view is strong, and there are many false points, which seriously affects the target tracking, data association and other links.

[0055] By counting the distribution histogram of the point cloud in the horizontal position and finding the peak value, the false track caused by clutter interference in the morning can be effectively reduced, and a more accurate road edge can be provided for subsequent target positioning and tracking, thereby dividing the tracking area. Figure 2 As shown in Figure 1, the lateral coordinate histogram of the radar target point cloud over a period of time. Based on the established histogram and the known number of lanes, the position of each lane can be located.

[0056] It can be concluded from the above that this embodiment effectively removes noise and clutter interference in the point cloud data by performing speed screening on the third data and establishing a point cloud distribution histogram, especially for the complex environment on both sides of the traffic road affected by trees, green belts, etc. in the actual measurement scene, and significantly improves the data quality.

[0057] In one embodiment of the present disclosure, clustering the fourth data to obtain the first data includes:

[0058] Performing a first clustering on the fourth data based on a dynamic neighborhood density algorithm to obtain fifth data;

[0059] The fifth data is clustered for the second time based on the azimuth-guided algorithm to obtain the first data.

[0060] In this embodiment, an azimuth-oriented dynamic density clustering iterative method is established to cluster radar point cloud data.

[0061] First, this embodiment can use the dynamic neighborhood density clustering method to perform the first clustering on the radar point cloud data. Figure 3 As shown in Figure 1, the shape of the dynamic neighborhood density cluster is set to an ellipse, whose major axis is parallel to the lane line direction and whose minor axis is perpendicular to the lane line direction. The attribute information of this clustering method includes radar distance and radial velocity.

[0062] The specific process is as follows: loop through each point cloud sample to determine whether the point cloud sample has been visited. If it has been visited, mark it as a noise point; if it has not been visited, mark the sample as visited. Next, find the neighborhood set of the dynamic density of the point cloud sample. For a density center point P0, its coordinates in the radar coordinate system are (x0, y0), and the radial velocity is v0. For any neighborhood point P (x, y) of the density center point P0, its radial velocity is v. If the distance D between points P0 and P is ≤ D0 (D0 is a constant), then point P is judged to be in the dynamic neighborhood density set of P0; otherwise, it is not. The distance D is calculated as follows:

[0063]

[0064] Define the length of the major axis of the ellipse as:

[0065]

[0066] Define the length of the minor axis of the ellipse as:

[0067]

[0068] Among them, a and b are the length of the major axis and minor axis of the dynamic density clustering ellipse, respectively, and their sizes depend on the radar cross-sectional area. Parameters a0 and b0 are the major and minor axis sizes of general cars, respectively. and According to the radar cross-section (RCS) values ​​of different radar point clouds and the size of common vehicles, this embodiment targets the current radar cross-section R P Three different thresholds are divided and set as M1, M2 and M3 respectively, and N is represented as a constant.

[0069] Then the number of neighborhood points of the point cloud sample is determined. If it is less than the minimum number of neighborhood points, the point is marked as a noise point; otherwise, if it is greater than the minimum number of neighborhood points, a new cluster is created and each neighborhood point is traversed to expand the cluster.

[0070] Secondly, an azimuth-guided secondary clustering algorithm is established to solve the problem of large target vehicle splitting and the existence of noise in the cluster. The established azimuth-guided dynamic density clustering iterative process is as follows: Figure 4 shown.

[0071] For the two point cloud clusters A and B (fifth data) after the first clustering, (several clusters adjacent to each other in a certain arrangement), find the two point cloud samples P1 and P2 with the closest Euclidean distance in cluster A and cluster B, and , . Compare the Euclidean distance between point cloud sample P1 and point cloud sample P2. If the Euclidean distance is greater than the set threshold T1, point cloud clusters A and B are not merged. Otherwise, if the Euclidean distance is less than the set threshold T1, the azimuths of point cloud sample P1 and point cloud sample P2 are further compared. If the difference between the azimuths of cloud sample P1 and cloud sample P2 is less than the threshold T2, cluster A and cluster B are merged. Otherwise, cluster A and cluster B are not merged.

[0072] It can be concluded from the above that this embodiment performs efficient and accurate clustering processing on radar point cloud data, effectively solves the problems of large target vehicle splitting and the presence of noise in the cluster, improves the accuracy and reliability of the clustering results, and provides strong data support for subsequent vehicle identification, tracking and decision-making.

[0073] In one embodiment of the present disclosure, obtaining the radar coordinates of the calibrated vehicle based on the first data includes: resampling the first data to obtain the radar coordinates of the calibrated vehicle.

[0074] In this embodiment, after obtaining the first data, the point cloud data corresponding to the calibrated vehicle is determined through clustering processing. These point cloud data are usually obtained by the millimeter wave radar detecting the vehicle at different time points and different angles, and there may be problems such as uneven data distribution and inconsistent density. In order to more accurately determine the radar coordinates of the calibrated vehicle, this embodiment needs to resample the first data.

[0075] In this embodiment, resampling can make the distribution of point cloud data in space more uniform. For example, if the original data is too dense in some areas and sparse in other areas, resampling can adjust the distribution of data so that the density of data is more balanced in the entire space. Resampling can reduce the impact of outliers and noise on coordinate calculation. If the original uneven data is directly used to calculate the radar coordinates, the coordinate calculation results may be inaccurate due to the existence of individual abnormal points. The resampled data is more stable, which helps to improve the accuracy of coordinate calculation.

[0076] In one embodiment of the present disclosure, resampling the first data to obtain radar coordinates of a calibrated vehicle includes:

[0077] Resampling the first data based on the first formula;

[0078] The first formula is:

[0079]

[0080] in, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinate, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates.

[0081] In this embodiment, a numerical interpolation time synchronization method is used to resample the first data (data acquired by the millimeter wave radar), and the radar point cloud data and the RTK data timestamps are matched one by one to obtain a data pair of the calibrated vehicle at the radar coordinates and the longitude and latitude coordinates.

[0082] The trajectory of the calibration vehicle is resampled in radar coordinates. Interpolation is performed between two adjacent points, with the interval between them being the radar cycle time. In such a short time, the velocity of the calibration vehicle is assumed to be constant, because the velocity of the vehicle can only change slightly. Therefore, the trajectory of the calibration vehicle can be resampled in radar coordinates using linear interpolation. The first data is at t k and t k+1 The x coordinates at the time are x k and x k+1 , and its coordinates are known, t can be calculated according to the first formula m The x coordinate at time t can be obtained by the same method. m The y coordinate of the moment.

[0083] After resampling the trajectory of the calibration vehicle, a set of latitude and longitude coordinates of the calibration vehicle with the same time stamp is obtained. and its corresponding radar coordinates .

[0084] in, Represented as the longitude coordinate of the i-th point, Represented as the latitude coordinate of the i-th point, It is expressed as the horizontal coordinate of the i-th point in the radar coordinate system, It is expressed as the ordinate of the i-th point in the radar coordinate system, Expressed as the number of coordinate points, Represented as the transpose of a matrix.

[0085] From the above, it can be concluded that the use of numerical interpolation time synchronization method to resample the radar data of the calibrated vehicle can accurately match the radar point cloud data with the RTK data timestamp, thereby reconstructing the vehicle's trajectory in the radar coordinates through linear interpolation while ensuring that the vehicle speed changes slightly, effectively improving the accuracy and efficiency of data processing.

[0086] In one embodiment of the present disclosure, associating radar coordinates with longitude and latitude coordinates based on similarity between first data and second data includes:

[0087] Calculating the similarity between the first data and the second data based on a second formula;

[0088] The second formula is:

[0089]

[0090] in, It is represented by the similarity of the calibrated vehicle speed in the first data and the second data, Expressed as The speed of the calibrated vehicle obtained by the millimeter-wave radar at the moment, Expressed as The speed of the calibrated vehicle obtained by navigation at all times, Expressed as the number of data points.

[0091] In this embodiment, a speed similarity matching method is established to associate and calibrate the vehicle radar coordinates and the longitude and latitude coordinates.

[0092] In this embodiment, , where T0 is a set time threshold, ms represents millimeter, here T0=100ms is set, within T0 time, it is assumed that the speed of the calibrated vehicle is a constant, and N0 is the total number of data point pairs of clustered vehicle trajectories that meet this time range condition.

[0093] From the above, it can be concluded that by calculating the similarity between the calibrated vehicle speed obtained by the millimeter-wave radar and the navigation system, the radar coordinates and the longitude and latitude coordinates can be accurately associated, effectively measuring the consistency between the two different data sources.

[0094] In one embodiment of the present disclosure, the radar coordinates are transformed based on the longitude and latitude coordinates to obtain a calibration matrix, including:

[0095] Transformation of radar coordinates to longitude and latitude coordinates based on affine transformation;

[0096] The calibration matrix for transforming radar coordinates to longitude and latitude coordinates is obtained based on the least squares method.

[0097] In this embodiment, affine transformation is applied to realize the transformation from the radar coordinate system to the latitude and longitude coordinate system. The least square method is used to solve the transformation matrix parameters from the radar coordinate system to the latitude and longitude coordinate system of the calibration vehicle.

[0098]

[0099] in, Expressed as transformation matrix parameters, the latitude and longitude coordinates of the calibrated vehicle are (x1, y1),…, (x n ,y n ), whose arrangement matrix is ​​represented by W, and the corresponding position coordinates of the target vehicle in the radar coordinates are ( ),…,( ), the permutation matrix is ​​represented by A, and the coordinate system transformation matrix is ​​represented by Z, then:

[0100]

[0101] Represented as the transpose of matrix A.

[0102] In this embodiment, reference Figure 5 , Figure 5 A flow chart of a dynamic calibration method for a roadside millimeter wave radar provided in another embodiment of the present disclosure. The method includes:

[0103] Read the PTK data file of the integrated navigation and obtain the PTK data of the calibrated vehicle in each lane (equivalent to the second data);

[0104] Read the traffic target point cloud data (equivalent to the third data) file captured by the roadside millimeter wave radar. Perform point cloud preprocessing on the target point cloud data (obtain the fourth data after preprocessing), and establish an orientation-oriented dynamic density clustering method to cluster the point cloud (obtain the first data after clustering);

[0105] The radar data is resampled using the data interpolation time synchronization method, and the radar point cloud data and RTK data timestamps are matched one by one to obtain the data pair of the connected vehicle in the millimeter wave radar coordinate system (equivalent to the radar coordinates) and the calibrated RTK system (equivalent to the longitude and latitude coordinates);

[0106] Establishing a speed similarity matching method to associate the vehicle trajectory points in the radar coordinate system with the trajectory point coordinates of the calibration vehicle;

[0107] The least square method is used to solve the conversion matrix between radar coordinates and longitude and latitude coordinates to obtain the calibration matrix;

[0108] The calibration matrix is ​​used to calculate the longitude and latitude coordinates of the vehicle under GPS.

[0109] In this embodiment, dynamic calibration reduces errors caused by installation errors, environmental changes, etc. by real-time calibration of the conversion parameters between the radar coordinate system and the ground coordinate system, thereby ensuring that the data collected by the radar is highly accurate. This provides a reliable data basis for subsequent accurate detection and tracking of traffic vehicles.

[0110] The road environment is complex and changeable, including weather changes, road construction and other factors. Dynamic calibration can adjust radar system parameters in real time to adapt to these changes and maintain high detection accuracy. This adaptability is crucial to achieve accurate detection and tracking of traffic vehicles.

[0111] Through regular dynamic calibration, potential problems in the radar system can be discovered and corrected in a timely manner, such as sensor aging, signal interference, etc. This helps to improve the long-term stability and reliability of the radar system and provide continuous support for accurate detection and tracking of traffic vehicles.

[0112] Corresponding to the dynamic calibration method for roadside millimeter wave radar in the above embodiment, Figure 6 This is a structural block diagram of a dynamic calibration device for a roadside millimeter wave radar provided in an embodiment of the present disclosure. For ease of description, only the parts related to the embodiment of the present disclosure are shown. Figure 6 The dynamic calibration device 20 for the roadside millimeter-wave radar includes: a data acquisition module 21, a data association module 22, a data transformation module 23, and a data calibration module 24.

[0113] The data acquisition module 21 is used to obtain the radar coordinates of the calibration vehicle based on the first data, and obtain the longitude and latitude coordinates of the calibration vehicle based on the second data, wherein the first data is the data obtained by the millimeter wave radar, and the second data is the data obtained by the navigation of the calibration vehicle;

[0114] A data association module 22, configured to associate radar coordinates with latitude and longitude coordinates based on similarity between the first data and the second data;

[0115] A data conversion module 23 is used to transform the radar coordinates based on the longitude and latitude coordinates to obtain a calibration matrix;

[0116] The data calibration module 24 is used to obtain the target longitude and latitude coordinates of the target vehicle based on the calibration matrix.

[0117] In one embodiment of the present disclosure, the dynamic calibration device 20 for the roadside millimeter wave radar further includes: a data preprocessing module; the data preprocessing module is specifically used for:

[0118] Performing point cloud preprocessing based on the third data to obtain fourth data;

[0119] The fourth data is clustered to obtain the first data.

[0120] In one embodiment of the present disclosure, the data preprocessing module is further used to:

[0121] Performing a first clustering on the fourth data based on a dynamic neighborhood density algorithm to obtain fifth data;

[0122] The fifth data is clustered for the second time based on the azimuth-guided algorithm to obtain the first data.

[0123] In one embodiment of the present disclosure, the data acquisition module 21 is specifically used for:

[0124] The first data is resampled to obtain radar coordinates of the calibrated vehicle.

[0125] In one embodiment of the present disclosure, the data acquisition module 21 is further used to:

[0126] Resampling the first data based on the first formula;

[0127] The first formula is:

[0128]

[0129] in, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinate, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates.

[0130] In one embodiment of the present disclosure, the data association module 22 is specifically used to:

[0131] Calculating the similarity between the first data and the second data based on a second formula;

[0132] The second formula is:

[0133]

[0134] in, It is represented by the similarity of the calibrated vehicle speed in the first data and the second data, Expressed as The speed of the calibrated vehicle obtained by the millimeter-wave radar at the moment, Expressed as The speed of the calibrated vehicle obtained by navigation at all times, Expressed as the number of time points.

[0135] In one embodiment of the present disclosure, the data transformation module 23 is specifically used for:

[0136] Transformation of radar coordinates to longitude and latitude coordinates based on affine transformation;

[0137] The calibration matrix for transforming radar coordinates to longitude and latitude coordinates is obtained based on the least squares method.

[0138] See also Figure 7 , Figure 7 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 7 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 6 The functions of modules 21 to 24 are shown.

[0139] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0140] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0141] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0142] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the dynamic calibration method for roadside millimeter-wave radar provided in the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.

[0143] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0144] The computer-readable storage medium may be an internal storage unit of the electronic device of any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0145] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0146] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0147] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0148] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0149] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0150] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A dynamic calibration method for roadside millimeter wave radar, characterized in that: include: Obtaining radar coordinates of the calibrated vehicle based on first data, and obtaining longitude and latitude coordinates of the calibrated vehicle based on second data, wherein the first data is data acquired by a millimeter-wave radar, and the second data is data acquired by a navigation system of the calibrated vehicle; associating the radar coordinates with the latitude and longitude coordinates based on similarity between the first data and the second data; Transforming the radar coordinates based on the latitude and longitude coordinates to obtain a calibration matrix; Obtaining target latitude and longitude coordinates of the target vehicle based on the calibration matrix; Before obtaining the radar coordinates of the calibrated vehicle based on the first data and obtaining the latitude and longitude coordinates of the calibrated vehicle based on the second data, the method further includes: Performing point cloud preprocessing based on the third data to obtain fourth data; Clustering the fourth data to obtain the first data; The step of performing point cloud preprocessing based on the third data to obtain fourth data includes: Performing speed screening and establishing a distribution histogram of the point cloud based on the third data to obtain fourth data; The clustering the fourth data to obtain the first data includes: Performing a first clustering on the fourth data based on a dynamic neighborhood density algorithm to obtain fifth data; Specifically, the shape of the dynamic neighborhood density cluster is set to an ellipse, with its major axis parallel to the lane line direction and its minor axis perpendicular to the lane line direction; Loop through each point cloud sample to determine whether the point cloud sample has been visited. If it has been visited, mark it as a noise point; if it has not been visited, mark the sample as visited. Finding a neighborhood set of dynamic density of the point cloud sample; If the distance from the density center point to any neighboring point in the neighborhood set of dynamic density is less than or equal to the preset value, then any neighboring point is judged to be in the neighborhood set of dynamic density. The distance calculation formula from the density center point to any neighboring point in the neighborhood set of dynamic density is: Where D represents the distance from the density center point to any neighboring point in the neighborhood set. For a density center point P0, its coordinates in the radar coordinate system are (x0, y0), and the radial velocity is v0. For any neighborhood point P(x, y) of the density center point P0, its radial velocity is v. a is the length of the major axis of the dynamic density clustering ellipse, b is the length of the minor axis of the dynamic density clustering ellipse, a0 is the major axis parameter of the car, and b0 is the minor axis parameter of the car. and is the dynamic change of the setting, R P is the radar cross-section, N is represented as a constant, and M1, M2, and M3 are different thresholds; Determine the number of neighborhood points of the point cloud sample. If it is less than the minimum number of neighborhood points, mark the point as a noise point. If it is greater than the minimum number of neighborhood points, create a new point cloud cluster and traverse each neighborhood point to expand the point cloud cluster. The fifth data is clustered for the second time based on the azimuth guidance algorithm to obtain the first data; and the associating the radar coordinates and the longitude and latitude coordinates based on the similarity between the first data and the second data comprises: Calculating the similarity between the first data and the second data based on a second formula; The second formula is: in, It is represented by the similarity of the calibrated vehicle speed in the first data and the second data, Expressed as The speed of the calibrated vehicle obtained by the millimeter-wave radar at the moment, Expressed as The speed of the calibrated vehicle obtained by the navigation at the time, Expressed as the number of data points.

2. The dynamic calibration method for roadside millimeter wave radar according to claim 1, characterized in that: The step of obtaining the radar coordinates of the calibrated vehicle based on the first data includes: The first data is resampled to obtain radar coordinates of the calibrated vehicle.

3. The dynamic calibration method for roadside millimeter wave radar according to claim 2, characterized in that: The step of resampling the first data to obtain radar coordinates of the calibrated vehicle includes: Resampling the first data based on a first formula; The first formula is: in, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinates at the moment, Expressed as The horizontal coordinate of the radar coordinate, Expressed as The vertical coordinate of the radar coordinates at the moment, Expressed as The vertical coordinate of the radar coordinates.

4. The dynamic calibration method for roadside millimeter wave radar according to claim 1, characterized in that: The step of transforming the radar coordinates based on the longitude and latitude coordinates to obtain a calibration matrix includes: Transformation of radar coordinates to longitude and latitude coordinates based on affine transformation; The calibration matrix for transforming radar coordinates to longitude and latitude coordinates is obtained based on the least squares method.

5. A dynamic calibration device for roadside millimeter wave radar, characterized in that: include: A data acquisition module, used to obtain the radar coordinates of the calibration vehicle based on first data, and obtain the longitude and latitude coordinates of the calibration vehicle based on second data, wherein the first data is data obtained by the millimeter wave radar, and the second data is data obtained by the navigation of the calibration vehicle; A data association module, configured to associate the radar coordinates with the latitude and longitude coordinates based on similarity between the first data and the second data; A data conversion module, used for transforming the radar coordinates based on the latitude and longitude coordinates to obtain a calibration matrix; A data calibration module, used to obtain the target latitude and longitude coordinates of the target vehicle based on the calibration matrix; Before obtaining the radar coordinates of the calibrated vehicle based on the first data and obtaining the latitude and longitude coordinates of the calibrated vehicle based on the second data, the method further includes: Performing point cloud preprocessing based on the third data to obtain fourth data; Clustering the fourth data to obtain the first data; The step of performing point cloud preprocessing based on the third data to obtain fourth data includes: Performing speed screening and establishing a distribution histogram of the point cloud based on the third data to obtain fourth data; The clustering the fourth data to obtain the first data includes: Performing a first clustering on the fourth data based on a dynamic neighborhood density algorithm to obtain fifth data; Specifically, the shape of the dynamic neighborhood density cluster is set to an ellipse, with its major axis parallel to the lane line direction and its minor axis perpendicular to the lane line direction; Loop through each point cloud sample to determine whether the point cloud sample has been visited. If it has been visited, mark it as a noise point; if it has not been visited, mark the sample as visited; Finding a neighborhood set of dynamic density of the point cloud sample; If the distance from the density center point to any neighboring point in the neighborhood set of dynamic density is less than or equal to the preset value, then any neighboring point is judged to be in the neighborhood set of dynamic density. The distance calculation formula from the density center point to any neighboring point in the neighborhood set of dynamic density is: Where D represents the distance from the density center point to any neighboring point in the neighborhood set. For a density center point P0, its coordinates in the radar coordinate system are (x0, y0), and the radial velocity is v0. For any neighborhood point P(x, y) of the density center point P0, its radial velocity is v. a is the length of the major axis of the dynamic density clustering ellipse, b is the length of the minor axis of the dynamic density clustering ellipse, a0 is the major axis parameter of the car, and b0 is the minor axis parameter of the car. and is the dynamic change of the setting, R P is the radar cross-section, N is represented as a constant, and M1, M2, and M3 are different thresholds; Determine the number of neighborhood points of the point cloud sample. If it is less than the minimum number of neighborhood points, mark the point as a noise point. If it is greater than the minimum number of neighborhood points, create a new point cloud cluster and traverse each neighborhood point to expand the point cloud cluster. Performing a second clustering on the fifth data based on an azimuth-guided algorithm to obtain the first data; The associating the radar coordinates with the longitude and latitude coordinates based on the similarity between the first data and the second data comprises: Calculating the similarity between the first data and the second data based on a second formula; The second formula is: in, It is represented by the similarity of the calibrated vehicle speed in the first data and the second data, Expressed as The speed of the calibrated vehicle obtained by the millimeter-wave radar at the moment, Expressed as The speed of the calibrated vehicle obtained by the navigation at the time, Expressed as the number of data points.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images