Method, system and device for automatically calibrating origin information of multiple radars in expressway traffic scene and medium

Abstract

The invention relates to a multi-radar origin information automatic calibration method, system and device in an expressway traffic scene and a medium, and the method comprises the steps: obtaining the device information and trajectory data of a to-be-corrected radar, and judging the irradiation direction of the radar; acquiring and processing map data; judging whether the to-be-corrected road radar is a reference radar or not; if the radar is a reference radar without a connection area, adjusting the angle of the cleaned trajectory data, matching the cleaned trajectory data with a road range in map data, and determining an optimal correction angle according to the number of matched trajectories; and if the non-reference radar has the connection area, referring to the connected reference radar, determining an optimal matching track through cleaning and track matching, carrying out preliminary correction according to the difference between the matching tracks, and carrying out secondary correction according to the matching degree of track data in the radar irradiation direction before cleaning and the road range. Through the advanced track processing technology and the automatic calibration process, the problems in the radar calibration process are effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method, system, device and medium for automatically calibrating multi-radar origin information in an expressway traffic scenario. Background Art

[0002] Millimeter-wave radar has been widely used in modern transportation systems, especially in vehicle detection and monitoring. Due to its ability to work stably in various weather conditions and provide accurate object detection and speed measurement information, millimeter-wave radar has become a key sensor in intelligent transportation systems. However, the effectiveness of millimeter-wave radar is highly dependent on its installation position, angle, and precise alignment with the surrounding environment.

[0003] At present, the installation and calibration of millimeter-wave radars are mainly carried out manually. During the installation process, the initial installation position and angle of the radar are determined by means of stake numbers, high-precision positioning instruments, and traffic channelization calibration. Then, through a series of conversion processes, the physical location of the radar is mapped to a digital map to ensure that the radar data accurately corresponds to the actual geographic location. In addition, a certain number of millimeter-wave radar detection tracks need to be collected and compared with the corresponding channelization and visual detection images to achieve accurate calibration of the radar origin.

[0004] However, this manual calibration method has several problems. First, the steps are cumbersome, especially when carrying out large-scale projects such as expressways and tunnels. The large number of equipment involved leads to huge deployment and calibration costs. Secondly, due to the lack of unified calibration principles, subjective factors have a greater impact in the calibration process, which can easily cause inconsistent calibration accuracy, thus affecting the effect of the holographic trajectory. Finally, since external factors such as vibration and weather may cause the radar angle and position to shift, the radar origin needs to be regularly inspected and recalibrated, which brings about greater operation and maintenance costs.

[0005] In holographic projects with long range and large scale, such as expressways, highways, and tunnels, the calibration and debugging of the installation positions and angles of a large number of millimeter-wave radar devices becomes a challenge. If debugging is completely dependent on manual work, it is not only easy to cause subjective calibration deviations, but also lead to a large waste of resources, and ultimately affect the detection and connection effects. Therefore, there is an urgent need to improve the installation and calibration methods of millimeter-wave radars. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method, system, device and medium for automatic calibration of multi-radar origin information in a freeway traffic scenario, which solves the technical problems that the installation and calibration process of millimeter-wave radar is cumbersome and costly, and is easily affected by subjective factors and the external environment, resulting in inconsistent calibration accuracy and difficulty in meeting the needs of large-scale holographic projects.

[0008] (II) Technical solution

[0009] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0010] In a first aspect, an embodiment of the present invention provides a method for automatically calibrating multi-radar origin information in an expressway traffic scenario, comprising:

[0011] Obtain the equipment information and trajectory data of the radar at the intersection to be corrected, and determine the radar illumination direction;

[0012] Acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected;

[0013] Determine whether the intersection radar to be corrected is a reference radar;

[0014] If it is a baseline radar without a connection zone, the angle of the trajectory data after adaptive data cleaning is gradually adjusted and matched with the road section range in the map data. The optimal correction angle of the radar origin is determined according to the change in the number of matched trajectories.

[0015] If there is a non-reference radar in the connection area, the connected reference radar is used as a reference, and the optimal matching trajectory is determined through adaptive data cleaning and trajectory matching. The angle and longitude and latitude are initially corrected based on at least one data difference between the optimal matching trajectories. In addition, the angle and longitude and latitude of the radar origin are secondarily corrected based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

[0016] Optionally, obtaining device information and trajectory data of the radar at the intersection to be corrected, and determining the radar illumination direction includes:

[0017] Send an information acquisition command to the radar at the intersection to be corrected to request the device to return data;

[0018] Receive the device information and trajectory data returned by the radar at the intersection to be corrected, and parse them according to a predetermined data format to obtain the required device information and trajectory data;

[0019] Process and analyze the acquired trajectory data to extract information related to the target movement;

[0020] The radar's illumination range is determined based on information related to the target's motion and in combination with at least one of the radar's operating mode, scanning method, and antenna pointing information.

[0021] Optionally, if there is no reference radar in the connection zone, the angle of the trajectory data after adaptive data cleaning is gradually adjusted and matched with the road section range in the map data, and the optimal correction angle of the radar origin is determined according to the change in the number of matched trajectories, including:

[0022] By means of adaptive parameters, the target trajectory to be corrected whose x-coordinate is less than a preset threshold, is consistent with the driving direction of the road section and whose trajectory number is greater than a set value is selected from the trajectory data of the intersection radar to be corrected;

[0023] The target track angle is gradually adjusted within a preset first coarse adjustment angle range with a set first coarse adjustment step length, and the longitude and latitude of the track after the angle adjustment are calculated, and the longitude and latitude of the track after the angle adjustment are matched with the road section range in the map data until the number of tracks in the corrected road section is reduced, thereby determining the angle after the coarse adjustment correction;

[0024] Based on the coarse-adjusted angle, the target track angle is gradually adjusted within the preset first fine-adjustment angle range with the set first fine-adjustment step size, and the longitude and latitude of the track after the angle adjustment is calculated. The longitude and latitude of the track after the angle adjustment are matched with the road section range in the map data until the number of tracks in the corrected road section is reduced again, and finally the optimal correction angle of the radar origin is determined.

[0025] Optionally, if there is a non-reference radar in the connection area, the connected reference radar is used as a reference, and the optimal matching trajectory is determined through adaptive data cleaning and trajectory matching, and the angle and longitude and latitude are initially corrected according to at least one data difference between the optimal matching trajectories, and the angle and longitude and latitude of the radar origin are secondary corrected according to the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data, including:

[0026] A reference radar connected to the non-reference radar is selected as a reference radar, and adaptive data cleaning is performed on the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected;

[0027] The cleaned reference radar and the trajectory data of the non-reference radar to be corrected are time-stamped and aligned, and the optimal matching trajectory is determined by calculating the speed difference and direction angle difference at each time node;

[0028] Preliminarily correcting the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected according to at least one data difference between the best matching tracks;

[0029] The trajectory data of the non-reference radar to be corrected in the radar illumination direction before adaptive cleaning are screened and matched with the road section range of the map data. The angle and longitude and latitude of the radar origin of the non-reference radar to be corrected are corrected secondary according to the matching degree.

[0030] Optionally, performing adaptive data cleaning on the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected includes:

[0031] For the trajectory data of the reference radar, the following adaptive data cleaning is performed:

[0032] Filter out reference trajectories whose x-coordinates are greater than the first connection comparison value obtained by the maximum x-coordinate value of the trajectory - the connection distance - the preset buffer value, to ensure that the selected trajectory data is within the connection range;

[0033] According to the direction of the road section, matching trajectory data is selected from the reference trajectory, and interference trajectories with inconsistent directions are excluded;

[0034] Filter out reference trajectories whose number of trajectories is greater than a preset number threshold;

[0035] Identify the trajectory segments in the reference trajectory whose direction angle is greater than a preset angle threshold, and use the adjacent point angles to calculate the direction angle to filter out the trajectory data that meets the preset floating feature requirements;

[0036] Calculate the difference between the first and last x-coordinates of the reference trajectory, and retain the trajectory data whose difference exceeds the preset buffer value;

[0037] For the trajectory data of the non-reference radar to be corrected, the following adaptive data cleaning is performed:

[0038] Filter out target tracks whose track x-coordinates are less than the second connection comparison value obtained by the minimum x-coordinate value + the connection distance + the preset buffer value, so as to ensure that the selected track data is within the range of connection with the reference radar;

[0039] According to the direction of the road section, select the matching trajectory data from the target trajectory and exclude the interference trajectory with inconsistent direction;

[0040] Filter out target trajectories whose number of trajectories is greater than a preset number threshold;

[0041] Calculate the x-coordinate difference between the first and last points of the target trajectory, and retain the trajectory data whose difference exceeds the preset buffer value.

[0042] Optionally, aligning the cleaned reference radar and the trajectory data of the non-reference radar to be corrected with timestamps, and determining the optimal matching trajectory by calculating the speed difference and the direction angle difference at each time node includes:

[0043] Align the timestamps of the cleaned reference trajectory and the target trajectory;

[0044] For each time node after alignment, select the time node and N nodes after the time node to calculate the speed of the reference trajectory and the target trajectory, and calculate the speed difference between the reference trajectory and the target trajectory according to the speed of the reference trajectory and the target trajectory;

[0045] For each time node after alignment, select the time node and N nodes after the time node to calculate the direction angle of the reference trajectory and the target trajectory, and calculate the direction angle difference between the reference trajectory and the target trajectory according to the direction angle of the reference trajectory and the target trajectory;

[0046] The track segments with a direction angle difference greater than 1 degree are read from the reference track and the target track, and the minimum value of the speed difference in the read track segments is used as the matching coefficient of the two targets to be matched;

[0047] According to the principle of minimum matching coefficient, the trajectory segment with the minimum matching coefficient is found from the read trajectory segments as the optimal matching trajectory.

[0048] Optionally, preliminarily correcting the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected according to at least one data difference between the best matching trajectories includes:

[0049] The reference trajectory and the target trajectory are respectively compared with the best matching trajectory, and the complete best matching reference trajectory and the complete best matching target trajectory containing the best matching trajectory segment are selected;

[0050] The slope is calculated according to the first and last coordinates of the complete best matching reference trajectory and the complete best matching target trajectory, and the heading angle is obtained by calculating the arc tangent value of the slope and converting it;

[0051] Perform preliminary angle correction based on the median of the heading angle difference between the complete best matching reference trajectory and the complete best matching target trajectory;

[0052] In the optimal matching trajectory, the longitude and latitude differences between each point are calculated, and from the longitude and latitude differences between each point, several groups of longitude and latitude differences with difference values ​​within a set range are selected to calculate the average longitude and latitude difference;

[0053] Using the latitude and longitude correction values, perform preliminary latitude and longitude correction on the origin latitude and longitude of the non-reference radar;

[0054] Determine whether the angle after the initial angle correction exceeds the correction angle threshold;

[0055] If the correction angle threshold is exceeded, the preliminary angle correction and preliminary longitude and latitude correction are repeatedly performed until the angle correction value is less than the correction angle threshold.

[0056] Optionally, the trajectory data of the non-reference radar to be corrected in the radar illumination direction before adaptive cleaning is screened and matched with the road section range of the map data, and the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected are corrected twice according to the matching degree, including:

[0057] From the trajectory data of the non-reference radar to be corrected within the radar illumination direction, select the trajectory data whose x coordinate is less than a preset correction judgment value;

[0058] Matching the selected trajectory data that is less than a preset correction judgment value with the road section range of the map data to determine whether the matching degree between the trajectory data and the road section range meets the preset standard;

[0059] If the matching degree reaches the preset standard, the angle is gradually adjusted within the preset second coarse adjustment angle range with the set second coarse adjustment step length, and the longitude and latitude of the trajectory after the angle adjustment is calculated, and the longitude and latitude of the trajectory after the angle adjustment is matched with the road section range in the map data until the number of trajectories in the corrected road section becomes smaller, and the angle after the coarse adjustment is determined;

[0060] Based on the coarsely adjusted angle, the angle is gradually adjusted within the preset second fine adjustment angle range with the set second fine adjustment step length, and the longitude and latitude of the track after the angle adjustment is calculated, and the longitude and latitude of the track after the angle adjustment is matched with the road section range in the map data until the number of tracks in the corrected road section becomes smaller again, and the finely adjusted angle is determined;

[0061] In the optimal matching trajectory, the longitude and latitude differences between each point are calculated, and from the longitude and latitude differences between each point, several groups of longitude and latitude differences with difference values ​​within a set range are selected to calculate the average longitude and latitude difference;

[0062] The latitude and longitude correction values ​​are used to perform secondary latitude and longitude correction on the origin latitude and longitude of the non-reference radar.

[0063] In a second aspect, an embodiment of the present invention provides a system for automatically calibrating multi-radar origin information in an expressway traffic scenario, comprising:

[0064] The illumination direction judgment module is used to obtain the device information and trajectory data of the radar at the intersection to be corrected, and to judge the radar illumination direction;

[0065] A map data acquisition module, used to acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected;

[0066] A reference radar judgment module is used to judge whether the intersection radar to be corrected is a reference radar;

[0067] The reference radar correction module is used to gradually adjust the angle of the trajectory data after adaptive data cleaning if there is no connection zone reference radar, and match it with the road section range in the map data, and determine the optimal correction angle of the radar origin according to the change in the number of matched trajectories;

[0068] The non-reference module correction module is used for determining the best matching trajectory through adaptive data cleaning and trajectory matching if there is a non-reference radar in a connection area, with the connected reference radar as a reference, and making a preliminary correction to the angle and longitude and latitude based on at least one data difference between the best matching trajectories, and making a secondary correction to the angle and longitude and latitude of the radar origin based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

[0069] In a third aspect, an embodiment of the present invention provides a device for connecting real-time radar queuing trajectories in an intersection traffic scenario, comprising: at least one database; and a memory communicatively connected to the at least one database; wherein the memory stores instructions executable by the at least one database, and the instructions are executed by the at least one database so that the at least one database can execute the method for automatic calibration of multi-radar origin information in an expressway traffic scenario as described above.

[0070] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having computer executable instructions stored thereon, which, when executed by a processor, implements the method for automatic calibration of multi-radar origin information in a freeway traffic scenario as described above.

[0071] (III) Beneficial effects

[0072] The beneficial effects of the present invention are as follows: the present invention proposes a new automated calibration method to address the problems of insufficient accuracy and complex operation during radar calibration. The present invention is mainly used in expressway traffic scenarios, and achieves accurate calibration of radar origin information through advanced technologies such as trajectory clustering and trajectory data distribution fitting. This innovation not only unifies the debugging standards for radar origins, but also standardizes the entire calibration process, thereby greatly improving the accuracy and efficiency of calibration.

[0073] Specifically, the present invention first obtains the device information and trajectory data of the intersection radar to be corrected, and processes it in combination with the high-precision map data. During the calibration process, targeted calibration strategies are adopted for the benchmark radar without a connection area and the non-benchmark radar with a connection area. The trajectory data is cleaned by means of adaptive parameters to ensure that the selected trajectory has high confidence and representativeness. Then, using map matching and trajectory matching technology, the origin angle and / or longitude and latitude of the radar are adjusted at least once until the optimal calibration effect is achieved.

[0074] It is worth mentioning that this automatic calibration method also has the function of periodic triggering, which can provide continuous automatic operation and maintenance support for holographic digital road projects. This means that once the radar equipment deviates or its performance degrades, it can be discovered and automatically calibrated in time to ensure the accuracy and stability of the radar data. This feature is of great significance for the long-term operation and maintenance of large-scale holographic projects.

[0075] Therefore, the present invention effectively solves the technical problems in the radar calibration process through advanced trajectory processing technology and automated calibration process, and provides strong technical support for radar applications in expressway traffic scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 A flowchart of a method for automatically calibrating multi-radar origin information in an expressway traffic scenario proposed by an embodiment of the present invention;

[0077] Figure 2 A specific flow chart of step S1 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0078] Figure 3 A specific flow chart of step S4 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0079] Figure 4 A specific flow chart of step S5 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0080] Figure 5 A specific flow chart of step S51 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0081] Figure 6 A specific flow chart of step S52 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0082] Figure 7 A specific flow chart of step S53 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0083] Figure 8 A specific flow chart of step S54 of a method for automatic calibration of multi-radar origin information in a freeway traffic scenario proposed by an embodiment of the present invention;

[0084] Fig. 9The present invention is a schematic diagram of the overall process of a method for automatically calibrating multi-radar origin information in an expressway traffic scenario proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0085] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0086] like Figure 1 As shown, an embodiment of the present invention proposes a method for automatic calibration of multi-radar origin information in an expressway traffic scenario, including: obtaining device information and trajectory data of a radar at an intersection to be corrected, and determining the radar irradiation direction; obtaining and processing map data of a road section lane corresponding to the radar at the intersection to be corrected; determining whether the radar at the intersection to be corrected is a reference radar; if it is a reference radar without a connection area, gradually adjusting the angle of the trajectory data after adaptive data cleaning, and matching it with the road section range in the map data, and determining the best correction angle of the radar origin according to the change in the number of matched trajectories; if it is a non-reference radar with a connection area, using the connected reference radar as a reference, determining the best matching trajectory through adaptive data cleaning and trajectory matching, preliminarily correcting the angle and longitude and latitude according to at least one data difference between the best matching trajectories, and secondarily correcting the angle and longitude and latitude of the radar origin according to the matching degree of the trajectory data in the radar irradiation direction before adaptive cleaning and the road section range of the map data.

[0087] The present invention proposes a new automated calibration method to solve the problems of insufficient accuracy and complex operation in the radar calibration process. The present invention is mainly used in expressway traffic scenarios, and achieves accurate calibration of radar origin information through advanced technologies such as trajectory clustering and trajectory data distribution fitting. This innovation not only unifies the debugging standards of radar origin, but also standardizes the entire calibration process, thereby greatly improving the accuracy and efficiency of calibration.

[0088] Specifically, the present invention first obtains the device information and trajectory data of the intersection radar to be corrected, and processes it in combination with the high-precision map data. During the calibration process, targeted calibration strategies are adopted for the benchmark radar without a connection area and the non-benchmark radar with a connection area. The trajectory data is cleaned by means of adaptive parameters to ensure that the selected trajectory has high confidence and representativeness. Then, using map matching and trajectory matching technology, the origin angle and / or longitude and latitude of the radar are adjusted at least once until the optimal calibration effect is achieved.

[0089] It is worth mentioning that this automatic calibration method also has the function of periodic triggering, which can provide continuous automatic operation and maintenance support for holographic digital road projects. This means that once the radar equipment deviates or its performance degrades, it can be discovered and automatically calibrated in time to ensure the accuracy and stability of the radar data. This feature is of great significance for the long-term operation and maintenance of large-scale holographic projects.

[0090] Therefore, the present invention effectively solves the technical problems in the radar calibration process through advanced trajectory processing technology and automated calibration process, and provides strong technical support for radar applications in expressway traffic scenarios.

[0091] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0092] Specifically, an embodiment of the present invention provides a method for automatically calibrating multi-radar origin information in an expressway traffic scenario, which includes:

[0093] S1. Obtain the device information and trajectory data of the radar at the intersection to be corrected, and determine the radar illumination direction.

[0094] Furthermore, if Figure 2 As shown, step S1 includes:

[0095] S11. Send an information acquisition instruction to the radar at the intersection to be corrected to request the device to return data.

[0096] S12, receiving the device information and trajectory data returned by the radar at the intersection to be corrected, and parsing them according to a predetermined data format to obtain the required device information and trajectory data.

[0097] Once the data returned by the device is received, it is parsed according to the predetermined data format. This data includes the device's own information (such as device status, radar model, serial number, installation date, last maintenance or calibration time, etc.) and the trajectory data detected by the radar (such as the location of the target object, movement path, etc.).

[0098] S13: Process and analyze the acquired trajectory data to extract information related to the target movement.

[0099] The purpose of this step is to extract information directly related to the target movement from a large amount of raw data, such as the target's movement trajectory, speed changes, etc.

[0100] S14. Determine the radar's illumination range based on information related to the target's motion and in combination with at least one of the radar's operating mode, scanning mode, and antenna pointing information.

[0101] Based on the information extracted in the previous step, the radar's irradiation range is comprehensively determined by combining the radar's operating mode (such as continuous wave, pulse wave, etc.), scanning method (such as mechanical scanning, electronic scanning, etc.) and antenna pointing information (if available). In addition, if the radar is equipped with a directional antenna, the antenna pointing information can be directly used to determine the irradiation direction, thereby more accurately tracking and locating the target.

[0102] S2. Acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected.

[0103] Retrieve map data related to the lanes of the road section corresponding to the radar at a specific intersection from the map database or related services. Process the acquired map data. The processing process includes data cleaning to remove erroneous or redundant information; data formatting to meet the set data processing standards; and data enhancement, such as adding lane markings, to improve the availability and richness of the data.

[0104] S3. Determine whether the intersection radar to be corrected is a reference radar.

[0105] In radar systems, reference radars usually refer to those radar devices with fixed positions, stable performance and accurate calibration, whose data are used as references for calibration and correction of other radar devices. Therefore, accurately determining whether an intersection radar is a reference radar is crucial to ensure the accuracy and reliability of the entire radar system.

[0106] The judgment process first relies on the acquired equipment information. By comparing the equipment information with the known reference radar equipment list, possible reference radar candidates can be preliminarily screened out.

[0107] Next, the radar’s coordinate information is used for further verification. This data can be obtained through GPS or other positioning technologies. These coordinates are compared with the known reference radar coordinates to check whether they are located at the predetermined reference location.

[0108] At the same time, considering that in the expressway traffic scenario, the radar in the non-connection zone has no reference radar connected to it, so only its own radar can be defined as the benchmark radar, and the non-benchmark radar in the non-connection zone is not judged.

[0109] When a radar with a connection zone is to be calibrated, the reference radar connected to it is selected as the reference radar. The non-reference radar algorithm includes a high-precision map comparison algorithm and a trajectory matching algorithm. The calibration accuracy is higher than that of the reference radar calibration algorithm, so the radar with a connection zone is generally judged as a non-reference radar. Therefore, the reference radar without a connection zone and the non-reference radar with a connection zone are judged and corrected next.

[0110] S4. If it is a reference radar without a connection zone, the angle of the trajectory data after adaptive data cleaning is gradually adjusted and matched with the road section range in the map data, and the optimal correction angle of the radar origin is determined according to the change in the number of matched trajectories.

[0111] Furthermore, if Figure 3 As shown, step S4 includes:

[0112] S41, by means of adaptive parameters, selecting from the trajectory data of the intersection radar to be corrected the target trajectory to be corrected whose x coordinate is less than a preset threshold, is consistent with the driving direction of the road section, and whose trajectory number is greater than a set value.

[0113] S42, gradually adjusting the target track angle within a preset first coarse adjustment angle range with a set first coarse adjustment step size and calculating the track longitude and latitude after the angle adjustment, matching the track longitude and latitude after the angle adjustment with the road section range in the map data until the number of tracks in the corrected road section is reduced, thereby determining the angle after the coarse adjustment correction.

[0114] S43. Based on the angle after the coarse adjustment correction, the target track angle is gradually adjusted within the preset first fine adjustment angle range with the set first fine adjustment step size, and the longitude and latitude of the track after the angle adjustment is calculated, and the longitude and latitude of the track after the angle adjustment is matched with the road section range in the map data until the number of tracks in the corrected road section is reduced again, and finally the optimal correction angle of the radar origin is determined.

[0115] In one embodiment, the target trajectory data cleaning and angle correction process is as follows.

[0116] First, the target trajectory data is cleaned. This process is based on adaptive parameters, and its main purpose is to intercept trajectories with high confidence, corresponding to the driving direction of the road section, and long appearance time. The specific steps include: screening out trajectory data with x-coordinates less than 200, which means that its confidence is high; then, according to the driving direction of the road section, further selecting the corresponding target trajectory data; finally, screening out trajectories with more than 50 target trajectories from these trajectories to ensure that the selected trajectories are sufficiently representative and stable.

[0117] Next, the angle is corrected. This process aims to determine the best correction angle by gradually adjusting the track angle and matching it with the road range in the map until the number of corrected tracks is reduced. Angle correction is divided into two stages: coarse adjustment and fine adjustment.

[0118] In the coarse adjustment stage, the angle of the target track is gradually adjusted in the range of -10 to 10 and in steps of 1. After each adjustment, the new longitude and latitude of the track is calculated and matched with the road section range in the map data. This process continues until the number of tracks in the corrected road section decreases. At this point, the coarse adjustment process ends and the current angle before correction is taken as the angle after coarse adjustment correction.

[0119] In the fine tuning stage, based on the angle corrected by the coarse adjustment, a finer angle adjustment is performed with a range of -1 to 1 and a step size of 0.1. Similar to the coarse adjustment stage, the new longitude and latitude of the track is calculated after each adjustment and matched with the road section range. The fine tuning process continues until the number of tracks in the corrected road section decreases again. At this point, the fine tuning process ends and the current angle before correction is determined to be the optimal correction angle of the radar origin.

[0120] S5. If there is a non-reference radar in the connection area, the connected reference radar is used as a reference, and the optimal matching trajectory is determined through adaptive data cleaning and trajectory matching. The angle and longitude and latitude are initially corrected based on at least one data difference between the optimal matching trajectories. In addition, the angle and longitude and latitude of the radar origin are secondarily corrected based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

[0121] Furthermore, if Figure 4 As shown, step S5 includes:

[0122] S51, selecting a reference radar connected to the non-reference radar as a reference radar, and performing adaptive data cleaning on the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected.

[0123] Furthermore, if Figure 5 As shown, step S51 includes:

[0124] For the trajectory data of the reference radar, the following adaptive data cleaning is performed:

[0125] S511 , select reference trajectories whose trajectory x-coordinates are greater than a first connection comparison value obtained by subtracting the connection distance from the preset buffer value from the maximum trajectory x-coordinates to ensure that the selected trajectory data is within the connection range.

[0126] S512: According to the direction of the road section, matching trajectory data is selected from the reference trajectory, and interference trajectories with inconsistent directions are excluded.

[0127] S513: Filter out reference trajectories whose number of trajectories is greater than a preset number threshold.

[0128] S514, identifying a trajectory segment in the reference trajectory whose direction angle is greater than a preset angle threshold, and using adjacent point angles to calculate the direction angle to filter out trajectory data that meets the preset floating feature requirements;

[0129] S515, calculating the difference between the first and last x-coordinates of the reference trajectory, and retaining the trajectory data whose difference exceeds the preset buffer value.

[0130] And, for the trajectory data of the non-reference radar to be corrected, the following adaptive data cleaning is performed:

[0131] S516, select target trajectories whose trajectory x-coordinates are less than a second connection comparison value obtained by the x-coordinate minimum value+the connection distance+the preset buffer value, so as to ensure that the selected trajectory data is within the range of connection with the reference radar.

[0132] S517: According to the road section direction, matching trajectory data is selected from the target trajectory, and interference trajectories with inconsistent directions are excluded.

[0133] S518: Filter out target trajectories whose number of trajectories is greater than a preset number threshold.

[0134] S519, calculating the difference between the first and last x-coordinates of the target trajectory, and retaining the trajectory data whose difference exceeds the preset buffer value.

[0135] In one embodiment, the process of adaptive cleaning of the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected will be described in detail:

[0136] First, the trajectory data of the reference radar is cleaned. This process aims to intercept reference trajectories that meet specific conditions based on adaptive parameters, including connection range, corresponding road section driving direction, long appearance time, obvious floating, and large moving distance. The specific steps are: first, select trajectory data with x-coordinates greater than (x-coordinate maximum value-connection distance-20) to ensure that the selected trajectory is within the appropriate connection range; then, according to the driving direction of the road section, further select the corresponding target trajectory data; then, select the trajectories with more than 20 target trajectories from these trajectories to reflect their long appearance time; in addition, it is necessary to select the trajectories with a direction turning angle greater than 1 degree in the target trajectory. In this step, the 5 points before and after are used to calculate the direction turning angle to ensure that the floating of the trajectory is obvious; finally, select the trajectories with a difference of more than 20 x-coordinates at the beginning and end of the target to reflect its large moving distance.

[0137] Next, the trajectory data of the non-reference radar to be corrected is cleaned. Similar to the cleaning of the reference trajectory data, this process is also based on the adaptive parameter method, but the focus is on intercepting the trajectory to be corrected that meets the connection range, the corresponding road section driving direction, the long appearance time and the large moving distance. The specific steps include: first, filter out the trajectory data whose x-coordinate is less than (x-coordinate minimum value + connection distance + 20) to determine that it is within the appropriate connection range; second, select the target trajectory data corresponding to the driving direction according to the road section direction; then, select the trajectory with a target trajectory number greater than 20 from these trajectories to reflect the persistence of its appearance time; finally, filter out the trajectory with a target head and tail x-coordinate difference of more than 20 to ensure that it has a large moving distance.

[0138] S52, aligning the timestamps of the cleaned reference radar and the trajectory data of the non-reference radar to be corrected, and determining the optimal matching trajectory by calculating the speed difference and the direction angle difference at each time node.

[0139] Furthermore, if Figure 6 As shown, step S52 includes:

[0140] S521 , aligning the timestamps of the cleaned reference trajectory and the target trajectory.

[0141] S522: For each aligned time node, select the time node and N nodes after the time node to calculate the speeds of the reference trajectory and the target trajectory, and calculate the speed difference between the reference trajectory and the target trajectory according to the speeds of the reference trajectory and the target trajectory.

[0142] S523: For each aligned time node, select the time node and N nodes after the time node to calculate the direction angles of the reference trajectory and the target trajectory, and calculate the direction angle difference between the reference trajectory and the target trajectory according to the direction angles of the reference trajectory and the target trajectory.

[0143] S524, reading out the trajectory segments having a direction angle difference greater than 1 degree in the reference trajectory and the target trajectory, and taking the minimum value of the speed difference in the read trajectory segments as the matching coefficient of the two targets to be matched.

[0144] S525 . According to the principle of minimum matching coefficient, find the trajectory segment with the minimum matching coefficient from the read trajectory segments as the optimal matching trajectory.

[0145] In another specific embodiment, the target trajectory matching process is described in detail. The process aims to align the reference trajectory with the trajectory to be corrected by time stamp, and calculate the speed and vector angle at each time node, so as to select the reference trajectory with the smallest speed vector difference from the trajectory to be corrected in the same time period as the optimal matching trajectory.

[0146] Specifically, first, the pre-step of target trajectory matching is performed, that is, the reference trajectory data is timestamped with the trajectory to be corrected. This is to ensure that the data points of the two trajectories correspond in subsequent calculations, so that they can be accurately compared and analyzed.

[0147] Next, for each aligned time node, two key calculations are performed. The first is to select the time node and the 5 points after the time node to calculate the speed of the reference trajectory and the target trajectory. The speed calculation is based on the distance and time difference between each point and the 5th point after it. In this way, the speed value of each time node can be obtained, and then the absolute value of the speed difference between the reference trajectory and the target trajectory can be calculated.

[0148] Second, similarly, for each time node, select the node and the fifth point after it to calculate the direction angle of the reference trajectory and the target trajectory. The calculation of the direction angle takes into account the positional relationship between each point and the fifth point before and after it. After calculating the direction angle, we can get the absolute value of the direction angle difference between the reference trajectory and the target trajectory.

[0149] After completing the above calculations, the data is further processed to read the trajectory segments with a direction angle difference greater than 1 degree, and the minimum value of the average speed difference is used as the matching coefficient of the two targets; according to the principle of minimum matching coefficient, the target matching pair is found.

[0150] S53: Preliminarily correct the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected according to at least one data difference between the best matching trajectories.

[0151] Furthermore, if Figure 7 As shown, step S53 includes:

[0152] S531 , respectively comparing the reference trajectory and the target trajectory with the best matching trajectory, and selecting a complete best matching reference trajectory and a complete best matching target trajectory including the best matching trajectory segment.

[0153] S532, calculating the slope according to the first and last coordinates of the complete best matching reference trajectory and the complete best matching target trajectory, and obtaining the heading angle by calculating the arc tangent value of the slope and converting it.

[0154] The heading angle is calculated from the slope, and the inverse tangent of the slope is calculated to obtain an angle in radians. Then, this angle is converted from radians to degrees (180 / π) to obtain the heading angle. The conversion formula is: heading angle = (180 / π) × atan (slope).

[0155] S533: Perform preliminary angle correction based on the median of the heading angle difference between the complete best matching reference trajectory and the complete best matching target trajectory.

[0156] S534. In the optimal matching trajectory, the longitude and latitude differences between the points are calculated, and from the longitude and latitude differences between the points, several groups of longitude and latitude differences whose difference values ​​are within a set range are selected to calculate the average longitude and latitude differences.

[0157] S535. Perform preliminary latitude and longitude correction on the origin longitude and latitude of the non-reference radar using the latitude and longitude correction values.

[0158] S536: Determine whether the angle after the initial angle correction exceeds the correction angle threshold.

[0159] S537: If the correction angle threshold is exceeded, the preliminary angle correction and preliminary longitude and latitude correction are repeatedly performed until the angle correction value is less than the correction angle threshold.

[0160] In one embodiment, first, an angle correction step is performed. For each target, the slope and heading angle are calculated based on its first and last coordinates within the matching range. Here, the heading angle represents the direction in which the target is moving. Next, the heading angle difference between the two sets of radar tracks is used to calculate the correction angle, specifically by taking the median of these differences as the correction value. The purpose of this step is to correct possible angle deviations and make the track data more accurate.

[0161] Next, the longitude and latitude correction is performed. The longitude and latitude differences between each point are calculated between matching pairs. In order to obtain more reliable correction data, several groups of data with similar longitude and latitude differences are selected and their average longitude and latitude differences are calculated. This average is then used for longitude and latitude correction to further improve the geolocation accuracy of the trajectory.

[0162] After completing the above two correction steps, check whether the angle correction value exceeds 0.1 degrees. If it exceeds, it means that the correction has not yet reached the required accuracy, so it is necessary to repeat the steps of origin angle correction and origin latitude and longitude correction. By continuously iterating these two correction processes until the angle correction value is less than 0.1 degrees, it is ensured that the target's trajectory is accurately calibrated in terms of angle and position. This iterative correction method helps to improve the accuracy and reliability of trajectory correction.

[0163] S54, screening the trajectory data of the non-reference radar to be corrected in the radar illumination direction before adaptive cleaning, matching it with the road section range of the map data, and performing secondary correction on the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected according to the matching degree.

[0164] Furthermore, if Figure 8 As shown, step S54 includes:

[0165] S541 . Select trajectory data whose x-coordinate is less than a preset correction judgment value from the trajectory data of the non-reference radar to be corrected within the radar illumination direction.

[0166] S542: Match the selected trajectory data that is less than a preset correction judgment value with the road section range of the map data to determine whether the matching degree between the trajectory data and the road section range meets a preset standard.

[0167] S543. If the matching degree reaches the preset standard, the angle is gradually adjusted within the preset second coarse adjustment angle range with the set second coarse adjustment step size, and the longitude and latitude of the trajectory after the angle adjustment is calculated. The longitude and latitude of the trajectory after the angle adjustment are matched with the road section range in the map data until the number of trajectories in the corrected road section becomes smaller, and the coarsely adjusted angle is determined.

[0168] S544. Based on the coarsely adjusted angle, the angle is gradually adjusted with the set second fine-tuning step length within the preset second fine-tuning angle range, and the longitude and latitude of the trajectory after the angle adjustment is calculated. The longitude and latitude of the trajectory after the angle adjustment are matched with the road section range in the map data until the number of trajectories in the corrected road section becomes smaller again, and the angle after fine-tuning is determined.

[0169] S545. In the optimal matching trajectory, the longitude and latitude differences between the points are calculated, and from the longitude and latitude differences between the points, several groups of longitude and latitude differences whose difference values ​​are within a set range are selected to calculate the average longitude and latitude differences.

[0170] S546. Perform a secondary latitude and longitude correction on the origin longitude and latitude of the non-reference radar using the latitude and longitude correction values.

[0171] In a specific embodiment, first, the trajectory data is screened and range matched. From the target trajectory data before radar cleaning, the trajectory data that is within the irradiation direction and has an x-coordinate less than 100 is screened. Then, the screened trajectory data is range matched with the road section. If more than 90% of the data falls within the road section range, it means that the data match is high and the next correction operation can be continued; otherwise, the process ends directly and prompts the information of pairing failure and correction failure.

[0172] After confirming that the data match is successful, enter the secondary correction. The first is the angle coarse adjustment stage, with a step size of 1, adjust the angle from -5 to 5 in sequence, and calculate the longitude and latitude of the track after each adjustment. After each adjustment, match it with the road section range to observe whether the number of tracks in the road section becomes smaller. When the number of tracks starts to decrease, it means that the current angle is no longer the optimal correction angle, so the coarse adjustment process ends and the angle before the current correction is taken as the corrected angle.

[0173] Next is the angle fine-tuning stage, where the angle is fine-tuned in a smaller range (-1 to 1) with a step size of 0.1. Similarly, after each adjustment, the longitude and latitude of the track are calculated and matched with the road segment range until a critical point is found that reduces the number of tracks in the road segment. After the fine-tuning is completed, the angle before the current correction is taken as the final correction angle.

[0174] Finally, the latitude and longitude of the origin are corrected. The latitude and longitude differences of each point between the matching pairs are calculated, and several groups of data with similar longitude and latitude differences are selected, and their average values ​​are calculated. This average value is used to correct the latitude and longitude of the origin to improve the accuracy of the trajectory data.

[0175] After completing the secondary correction, check whether all radars have been calibrated. If there are still uncalibrated radars, continue to calibrate the next connected radar. Through this cyclic iterative process, ensure that each radar can be accurately calibrated and corrected.

[0176] In addition, an embodiment of the present invention provides a system for automatically calibrating multi-radar origin information in an expressway traffic scenario, comprising:

[0177] The illumination direction judgment module is used to obtain the device information and trajectory data of the radar at the intersection to be corrected, and to judge the radar illumination direction;

[0178] A map data acquisition module, used to acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected;

[0179] A reference radar judgment module is used to judge whether the intersection radar to be corrected is a reference radar;

[0180] The reference radar correction module is used to gradually adjust the angle of the trajectory data after adaptive data cleaning if there is no connection zone reference radar, and match it with the road section range in the map data, and determine the optimal correction angle of the radar origin according to the change in the number of matched trajectories;

[0181] The non-reference module correction module is used for determining the best matching trajectory through adaptive data cleaning and trajectory matching if there is a non-reference radar in a connection area, with the connected reference radar as a reference, and making a preliminary correction to the angle and longitude and latitude based on at least one data difference between the best matching trajectories, and making a secondary correction to the angle and longitude and latitude of the radar origin based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

[0182] Furthermore, an embodiment of the present invention provides a device for connecting real-time radar queuing trajectories in an intersection traffic scenario, comprising: at least one database; and a memory communicatively connected to the at least one database; wherein the memory stores instructions that can be executed by the at least one database, and the instructions are executed by the at least one database so that the at least one database can execute the method for automatic calibration of multi-radar origin information in the expressway traffic scenario as described above.

[0183] When these instructions are executed by the database, they enable the database to perform a series of specific operations, which are exactly the method of automatic calibration of multi-radar origin information in expressway traffic scenarios proposed by the present invention. This method aims to improve the accuracy and consistency of radar trajectory data by automatically calibrating the origin information of multiple radars, thereby optimizing the monitoring and management of intersection traffic.

[0184] In short, the device provided by the embodiment of the present invention combines the functions of a database and a memory, and by executing specific instructions, realizes accurate calibration of multi-radar origin information in complex traffic scenarios, bringing convenience and benefits to traffic monitoring and management.

[0185] Furthermore, an embodiment of the present invention provides a computer-readable medium having computer executable instructions stored thereon, and when the executable instructions are executed by a processor, the method for automatically calibrating multi-radar origin information in a freeway traffic scenario as described above is implemented.

[0186] When these instructions are executed by the processor, they can trigger and implement a specific method, namely the method of automatic calibration of multi-radar origin information in expressway traffic scenarios described in this invention. This method accurately calibrates the origin information of multiple radars in expressway traffic scenarios in an automated manner, thereby ensuring that the trajectory data generated by each radar is optimized in terms of accuracy and consistency. This not only improves the overall performance of the traffic monitoring system, but also provides a more reliable data foundation for subsequent advanced applications such as traffic flow analysis and congestion prediction.

[0187] Therefore, this computer-readable medium and the instructions stored thereon constitute an indispensable part of the embodiments of the present invention, and provide an efficient and convenient solution for the origin information calibration of multiple radar systems in expressway traffic scenarios.

[0188] In summary, the embodiments of the present invention provide a method, system, device and medium for automatically calibrating multi-radar origin information in a freeway traffic scenario, such as Fig. 9As shown, the present invention uses adaptive parameter trajectory data cleaning technology to comprehensively and accurately clean and screen the collected trajectory data. This step comprehensively considers multiple dimensions such as the confidence level of the trajectory data, the length of time of occurrence, the moving distance, and the consistency of direction, thereby ensuring the reliability and accuracy of the retained data. Secondly, the present invention adopts an automatic angle correction technology based on trajectory data and high-precision maps. By gradually adjusting the longitude and latitude of the trajectory and accurately matching it with the high-precision map, the automatic and high-precision correction of the radar origin angle is achieved. During the calibration process, the present invention also uses an advanced trajectory matching algorithm. The algorithm can accurately align and match the trajectory data of the reference radar with the trajectory data of the radar to be corrected. By calculating the velocity vector and vector angle difference at each time node, the reference trajectory with the smallest velocity vector difference can be accurately found to ensure the best match of the two sets of trajectory data, and the angle calibration is performed based on the high-precision map. In addition, the present invention also adopts a method for origin calibration based on the position difference between the reference radar trajectory and the matching trajectory in the radar to be calibrated. By calculating the slope and heading angle of the first and last coordinates within the matching range, and using the median of the heading angle difference between the two sets of radar tracks for angle correction, combined with the screening and average calculation of the longitude and latitude differences, accurate correction of longitude and latitude is achieved.

[0189] The innovation of this invention is reflected in many aspects: by integrating trajectory data clustering and fitting technology, and with the assistance of high-precision maps, this invention has achieved breakthroughs in accuracy and application scope, providing more innovative solutions for complex traffic environments such as expressways. Its high accuracy and adaptability to complex scenarios are due to the perfect combination of trajectory data fitting and high-precision maps, which enables multi-radar automatic calibration to still perform well under complex conditions.

[0190] At the same time, the automatic calibration method of the present invention is designed to be extremely convenient and can be automatically triggered regularly, which significantly reduces the frequency and workload of manual intervention and improves the overall efficiency of system maintenance. Compared with the existing technology that relies on the periodic appearance of external vehicles, the present invention is undoubtedly more advantageous.

[0191] Finally, it is worth mentioning that the present invention is not only applicable to single radar systems, but can also be widely adapted to multi-radar systems, realizing the unification of radar origin calibration standards and the standardization of calibration processes, demonstrating strong versatility and expansion potential.

[0192] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.

[0193] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may 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.

[0194] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.

[0195] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0196] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0197] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0198] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for automatic calibration of multi-radar origin information in a freeway traffic scene, characterized in that: include: Obtain the equipment information and trajectory data of the radar at the intersection to be corrected, and determine the radar illumination direction; Acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected; Determine whether the intersection radar to be corrected is a reference radar; If it is a baseline radar without a connection zone, the angle of the trajectory data after adaptive data cleaning is gradually adjusted and matched with the road section range in the map data. The optimal correction angle of the radar origin is determined according to the change in the number of matched trajectories. If there is a non-reference radar in the connection area, the connected reference radar is used as a reference, and the optimal matching trajectory is determined through adaptive data cleaning and trajectory matching. The angle and longitude and latitude are initially corrected based on at least one data difference between the optimal matching trajectories. In addition, the angle and longitude and latitude of the radar origin are secondarily corrected based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

2. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 1, characterized in that: Obtain the device information and trajectory data of the intersection radar to be corrected, and determine the radar illumination direction including: Send an information acquisition command to the radar at the intersection to be corrected to request the device to return data; Receive the device information and trajectory data returned by the radar at the intersection to be corrected, and parse them according to a predetermined data format to obtain the required device information and trajectory data; Process and analyze the acquired trajectory data to extract information related to the target movement; The radar's illumination range is determined based on information related to the target's motion and in combination with at least one of the radar's operating mode, scanning method, and antenna pointing information.

3. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 1, characterized in that: If there is no baseline radar in the connection zone, the angle of the trajectory data after adaptive data cleaning is gradually adjusted and matched with the road section range in the map data. The optimal correction angle of the radar origin is determined according to the change in the number of matched trajectories, including: By means of adaptive parameters, the target trajectory to be corrected whose x-coordinate is less than a preset threshold, is consistent with the driving direction of the road section and whose trajectory number is greater than a set value is selected from the trajectory data of the intersection radar to be corrected; The target track angle is gradually adjusted within a preset first coarse adjustment angle range with a set first coarse adjustment step length, and the longitude and latitude of the track after the angle adjustment are calculated, and the longitude and latitude of the track after the angle adjustment are matched with the road section range in the map data until the number of tracks in the corrected road section is reduced, thereby determining the angle after the coarse adjustment correction; Based on the coarse-adjusted angle, the target track angle is gradually adjusted within the preset first fine-adjustment angle range with the set first fine-adjustment step size, and the longitude and latitude of the track after the angle adjustment is calculated. The longitude and latitude of the track after the angle adjustment are matched with the road section range in the map data until the number of tracks in the corrected road section is reduced again, and finally the optimal correction angle of the radar origin is determined.

4. The method for automatic calibration of multi-radar origin information in a freeway traffic scene according to any one of claims 1 to 3, characterized in that: If there is a non-reference radar in the connection area, the connected reference radar is used as a reference, and the best matching trajectory is determined through adaptive data cleaning and trajectory matching. According to at least one data difference between the best matching trajectories, the angle and longitude and latitude are initially corrected, and according to the matching degree of the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data, the angle and longitude and latitude of the radar origin are secondary corrected, including: A reference radar connected to the non-reference radar is selected as a reference radar, and adaptive data cleaning is performed on the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected; The cleaned reference radar and the trajectory data of the non-reference radar to be corrected are time-stamped and aligned, and the optimal matching trajectory is determined by calculating the speed difference and direction angle difference at each time node; Preliminarily correcting the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected according to at least one data difference between the best matching tracks; The trajectory data of the non-reference radar to be corrected in the radar illumination direction before adaptive cleaning are screened and matched with the road section range of the map data. The angle and longitude and latitude of the radar origin of the non-reference radar to be corrected are corrected twice according to the matching degree.

5. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 4, characterized in that: Adaptive data cleaning of the trajectory data of the reference radar and the trajectory data of the non-reference radar to be corrected includes: For the trajectory data of the reference radar, the following adaptive data cleaning is performed: Filter out reference trajectories whose x-coordinates are greater than the first connection comparison value obtained by the maximum x-coordinate value of the trajectory - the connection distance - the preset buffer value, to ensure that the selected trajectory data is within the connection range; According to the direction of the road section, matching trajectory data is selected from the reference trajectory, and interference trajectories with inconsistent directions are excluded; Filter out reference trajectories whose number of trajectories is greater than a preset number threshold; Identify the trajectory segments in the reference trajectory whose direction angle is greater than a preset angle threshold, and use the adjacent point angles to calculate the direction angle to filter out the trajectory data that meets the preset floating feature requirements; Calculate the difference between the first and last x-coordinates of the reference trajectory, and retain the trajectory data whose difference exceeds the preset buffer value; For the trajectory data of the non-reference radar to be corrected, the following adaptive data cleaning is performed: Filter out target tracks whose track x-coordinates are less than the second connection comparison value obtained by the minimum x-coordinate value + the connection distance + the preset buffer value, so as to ensure that the selected track data is within the range of connection with the reference radar; According to the direction of the road section, select the matching trajectory data from the target trajectory and exclude the interference trajectory with inconsistent direction; Filter out target trajectories whose number of trajectories is greater than a preset number threshold; Calculate the x-coordinate difference between the first and last points of the target trajectory, and retain the trajectory data whose difference exceeds the preset buffer value.

6. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 5, characterized in that: The cleaned reference radar and the trajectory data of the non-reference radar to be corrected are time-stamped and aligned. The speed difference and direction angle difference at each time node are calculated to determine the optimal matching trajectory, including: Align the timestamps of the cleaned reference trajectory and the target trajectory; For each time node after alignment, select the time node and N nodes after the time node to calculate the speed of the reference trajectory and the target trajectory, and calculate the speed difference between the reference trajectory and the target trajectory according to the speed of the reference trajectory and the target trajectory; For each time node after alignment, select the time node and N nodes after the time node to calculate the direction angle of the reference trajectory and the target trajectory, and calculate the direction angle difference between the reference trajectory and the target trajectory according to the direction angle of the reference trajectory and the target trajectory; The track segments with a direction angle difference greater than 1 degree are read from the reference track and the target track, and the minimum value of the speed difference in the read track segments is used as the matching coefficient of the two targets to be matched; According to the principle of minimum matching coefficient, the trajectory segment with the minimum matching coefficient is found from the read trajectory segments as the optimal matching trajectory.

7. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 6, characterized in that: According to at least one data difference between the best matching tracks, performing a preliminary correction on the angle and longitude and latitude of the radar origin of the non-reference radar to be corrected comprises: The reference trajectory and the target trajectory are respectively compared with the best matching trajectory, and the complete best matching reference trajectory and the complete best matching target trajectory containing the best matching trajectory segment are selected; The slope is calculated according to the first and last coordinates of the complete best matching reference trajectory and the complete best matching target trajectory, and the heading angle is obtained by calculating the arc tangent value of the slope and converting it; Perform preliminary angle correction based on the median of the heading angle difference between the complete best matching reference trajectory and the complete best matching target trajectory; In the optimal matching trajectory, the longitude and latitude differences between each point are calculated, and from the longitude and latitude differences between each point, several groups of longitude and latitude differences with difference values ​​within a set range are selected to calculate the average longitude and latitude difference; Using the latitude and longitude correction values, perform preliminary latitude and longitude correction on the origin latitude and longitude of the non-reference radar; Determine whether the angle after the initial angle correction exceeds the correction angle threshold; If the correction angle threshold is exceeded, the preliminary angle correction and preliminary longitude and latitude correction are repeatedly performed until the angle correction value is less than the correction angle threshold.

8. The method for automatic calibration of multi-radar origin information in a freeway traffic scene as claimed in claim 7, characterized in that: The trajectory data of the non-reference radar to be corrected in the radar illumination direction before adaptive cleaning is screened and matched with the road section range of the map data. The angle and longitude and latitude of the radar origin of the non-reference radar to be corrected are corrected twice according to the matching degree, including: From the trajectory data of the non-reference radar to be corrected within the radar illumination direction, select the trajectory data whose x coordinate is less than a preset correction judgment value; Matching the selected trajectory data that is less than a preset correction judgment value with the road section range of the map data to determine whether the matching degree between the trajectory data and the road section range meets the preset standard; If the matching degree reaches the preset standard, the angle is gradually adjusted within the preset second coarse adjustment angle range with the set second coarse adjustment step length, and the longitude and latitude of the trajectory after the angle adjustment is calculated, and the longitude and latitude of the trajectory after the angle adjustment is matched with the road section range in the map data until the number of trajectories in the corrected road section becomes smaller, and the angle after the coarse adjustment is determined; Based on the coarsely adjusted angle, the angle is gradually adjusted within the preset second fine adjustment angle range with the set second fine adjustment step length, and the longitude and latitude of the track after the angle adjustment is calculated, and the longitude and latitude of the track after the angle adjustment is matched with the road section range in the map data until the number of tracks in the corrected road section becomes smaller again, and the finely adjusted angle is determined; In the optimal matching trajectory, the longitude and latitude differences between each point are calculated, and from the longitude and latitude differences between each point, several groups of longitude and latitude differences with difference values ​​within a set range are selected to calculate the average longitude and latitude difference; The latitude and longitude correction values ​​are used to perform secondary latitude and longitude correction on the origin latitude and longitude of the non-reference radar.

9. A system for automatic calibration of multi-radar origin information in expressway traffic scenarios, characterized in that: include: The illumination direction judgment module is used to obtain the device information and trajectory data of the radar at the intersection to be corrected, and to judge the radar illumination direction; A map data acquisition module, used to acquire and process the map data of the road section lane corresponding to the intersection radar to be corrected; A reference radar judgment module is used to judge whether the intersection radar to be corrected is a reference radar; The reference radar correction module is used to gradually adjust the angle of the trajectory data after adaptive data cleaning if there is no reference radar in the connection area, and match it with the road section range in the map data, and determine the optimal correction angle of the radar origin according to the change in the number of matched trajectories; The non-reference module correction module is used for determining the best matching trajectory through adaptive data cleaning and trajectory matching if there is a non-reference radar in a connection area, with the connected reference radar as a reference, and making a preliminary correction to the angle and longitude and latitude based on at least one data difference between the best matching trajectories, and making a secondary correction to the angle and longitude and latitude of the radar origin based on the matching degree between the trajectory data in the radar illumination direction before adaptive cleaning and the road section range of the map data.

10. A device for connecting the real-time queuing track of radar vision in a traffic scene at an intersection, characterized in that: include: at least one database; And a memory communicatively connected to at least one database; wherein the memory stores instructions executable by at least one database, and the instructions are executed by at least one database so that at least one database can execute the method for automatic calibration of multi-radar origin information in an expressway traffic scenario as described in any one of claims 1-8.

11. A computer-readable medium having computer-executable instructions stored thereon, characterized in that: When the executable instructions are executed by the processor, the method for automatically calibrating multi-radar origin information in a freeway traffic scenario as described in any one of claims 1-8 is implemented.

Images