Roadside millimeter wave radar calibration method and system based on Internet of Vehicles

Abstract

The invention provides a roadside millimeter-wave radar calibration method and system based on the Internet of Vehicles, and the method comprises the steps: obtaining vehicle driving track point data through an intelligent connected vehicle, obtaining the millimeter-wave radar measurement data of a current environment, obtaining a corresponding track point coordinate data set through time synchronization processing, and obtaining a corresponding track point coordinate data set; and millimeter wave radar calibration parameters are solved by using an optimization algorithm, and the problems of low working efficiency and low safety caused by road closing are solved. And a local high-precision map stored by the edge calculation unit is used for inspection and evaluation, so that the accuracy of a millimeter-wave radar calibration result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking, and in particular to a roadside millimeter-wave radar calibration method and system based on vehicle networking. Background Art

[0002] With the development of intelligent transportation systems, roadside perception systems play an important role in obtaining information such as the location, speed, and posture of traffic participants (including motor vehicles, non-motor vehicles, and pedestrians). Roadside millimeter-wave radar is one of the most important sensors in the roadside perception system. Roadside millimeter-wave radar can identify and track the movement trajectory of traffic targets, and can obtain the relative position of the target from the locator installed in the millimeter-wave radar. In order to obtain the World Geodical System-84 (WGS-84) coordinates of the target detected by the millimeter-wave radar, calibration is required.

[0003] At present, millimeter wave radar calibration methods include traditional offline calibration methods based on calibration objects and online calibration methods based on neural networks. The former relies on calibration objects with obvious features (such as using corner reflectors). The calibration process requires closing roads and manual operations, which will inevitably affect traffic flow during calibration and may even cause traffic accidents. The latter uses neural networks to calibrate sensors. In order to obtain the parameters of the neural network, a large computing cost and storage capacity are required, which may be difficult to meet in actual traffic engineering projects. Summary of the invention

[0004] The present invention provides a roadside millimeter-wave radar calibration method and system based on the Internet of Vehicles, which are used to solve the defects of millimeter-wave radar calibration in the prior art, such as strong dependence on the objective environment, large amount of calculation and difficulty in application and promotion.

[0005] In a first aspect, the present invention provides a roadside millimeter-wave radar calibration method based on vehicle networking, comprising: Step 1, collecting vehicle data within the millimeter wave radar detection area; Step 2, clustering the vehicle data based on the density clustering algorithm OPTICS to extract vehicle trajectory points; Step 3, using a speed matching method to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory point; Step 4, obtaining the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resampling the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using a linear interpolation method to obtain a resampled motion trajectory of the intelligent networked vehicle; Step 5, selecting any group of intelligent networked vehicle collection data from the vehicle data to execute steps 1 to 4, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters using a pseudo-inverse transformation method; Step 6: Use the calibration parameters to calibrate the remaining groups of intelligent connected vehicle collected data in the vehicle data, and verify the calibration results.

[0006] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, step 1 includes: Obtain target information of multiple vehicles traveling in different lanes, the target information including timestamp, relative position, speed and identity information, the multiple vehicles including intelligent networked vehicles and ordinary vehicles; The WGS-84 coordinates of the plurality of vehicles are collected.

[0007] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, step 2 includes: Obtaining the x-coordinate and y-coordinate of the vehicle trajectory point at any two moments in the millimeter wave radar coordinate system, as well as the velocity of the trajectory point at the said any two moments; Based on OPTICS, the x-coordinate and y-coordinate at any two moments and the speed of the track point at any two moments are calculated to obtain the travel distance between any two moments; Determine the accessibility distance, and extract the vehicle trajectory point according to the travel distance between the arbitrary two time points and the accessibility distance.

[0008] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, step 3 includes: Get the first moment detection speed of the millimeter wave radar and the second moment reporting speed of the intelligent connected vehicle; Determine a preset time constraint range so that the difference between the first moment and the second moment is within the preset time constraint range; Acquire the total number of trajectory point data within the preset time constraint range, and obtain a speed similarity value of any trajectory point within the preset time constraint range based on the total number of trajectory point data, the detection speed at the first moment, and the reported speed at the second moment; The motion trajectory of the intelligent networked vehicle is constructed using trajectory points whose speed similarity values ​​are within a similarity threshold.

[0009] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, step 4 includes: Obtaining a plurality of first moment coordinates and a plurality of next moment coordinates of the first moment of the motion trajectory of the intelligent networked vehicle in the millimeter wave radar coordinate system; Performing linear interpolation calculation on the plurality of first moment coordinates and the plurality of first moment coordinates next to the plurality of first moment coordinates to obtain a plurality of second moment coordinates; The resampled motion trajectory of the intelligent connected vehicle is constructed by integrating all the second-time coordinates to obtain the millimeter-wave radar data with the same timestamp and the corresponding WGS-84 coordinates.

[0010] According to a roadside millimeter-wave radar calibration method based on the Internet of Vehicles provided by the present invention, the step 5 of randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle for coordinate system transformation includes: Convert WGS-84 coordinate system to ECEF Earth-centered rectangular coordinate system:

[0011]

[0012] in, are the major and minor radii of the reference ellipsoid, is the square of the eccentricity of the ellipsoid, is the radius of curvature of the reference ellipsoid, are longitude, latitude and altitude respectively. , , They are the three-dimensional coordinates of the Earth-centered, Earth-fixed rectangular coordinate system ECEF; Convert the Earth-centered rectangular coordinate system ECEF to the local Cartesian coordinate system ENU:

[0013] in, , , They are the three-dimensional coordinates of the local Cartesian coordinate system ENU, They are the 3D coordinates of the local reference point and the WGS-84 coordinates with the millimeter-wave radar position as the reference point. what you get; Convert WGS-84 coordinates Convert to local Cartesian coordinates (ENU) .

[0014] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, the step 5 of solving the calibration parameters by using the pseudo-inverse transformation method includes:

[0015]

[0016] in, are the millimeter wave radar coordinates, , is a two-dimensional change matrix; from Select For data, , the roadside calibration problem is formulated as an optimization problem of minimizing the projection error on the local coordinates: in, yes The distance between The parameters of are solved using the pseudo-inverse transformation method.

[0017] According to a roadside millimeter-wave radar calibration method based on vehicle networking provided by the present invention, step 6 includes: Obtain the standard WGS-84 coordinates of each lane centerline from the preset high-precision map; The calibrated mapped WGS-84 coordinates are obtained based on the collected data of the remaining groups of intelligent networked vehicles, and the average distance error between the calibrated mapped WGS-84 coordinates and the standard WGS-84 coordinates is obtained by calculating the foot point of the lane centerline and the Vincenty distance formula:

[0018] in, Indicates the foot point, Indicates the WGS-84 coordinates mapped after calibration. Indicates the center line of the lane; If it is determined that the average distance error is less than the preset error value, the calibration is qualified, otherwise recalibration is performed.

[0019] In a second aspect, the present invention further provides a roadside millimeter-wave radar calibration system based on Internet of Vehicles, comprising: The acquisition module is used to collect vehicle data within the detection area of ​​the millimeter-wave radar; A clustering module, used for clustering the vehicle data based on the density clustering algorithm OPTICS, and extracting vehicle trajectory points; An identification module, used to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory point by using a speed matching method; A resampling module, used to obtain the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resample the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using a linear interpolation method to obtain a resampled motion trajectory of the intelligent networked vehicle; A calibration module, used for selecting any group of intelligent networked vehicle collected data from the vehicle data to perform clustering, identification and resampling steps, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters using a pseudo-inverse transformation method; The verification module is used to calibrate the remaining groups of intelligent networked vehicle collected data in the vehicle data using the calibration parameters and verify the calibration results.

[0020] In a third aspect, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles as described above is implemented.

[0021] The method and system for calibrating roadside millimeter-wave radar based on the Internet of Vehicles provided by the present invention, based on high-reliability and low-latency Internet of Vehicles communication technology, uses intelligent connected vehicles to obtain vehicle driving trajectory point data, obtains millimeter-wave radar measurement data of the current environment, obtains the corresponding trajectory point coordinate data set through time synchronization processing, and uses an optimization algorithm to solve the millimeter-wave radar calibration parameters, which solves the problems of low work efficiency and safety caused by closed roads, and uses local high-precision maps stored in edge computing units for inspection and evaluation, thereby improving the accuracy of millimeter-wave radar calibration results. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 This is one of the flow charts of the roadside millimeter-wave radar calibration method based on the Internet of Vehicles provided by the present invention; Figure 2 This is the second flow chart of the roadside millimeter-wave radar calibration method based on the Internet of Vehicles provided by the present invention; Figure 3 It is a schematic diagram of a high-precision map lane provided by the present invention; Figure 4 It is a structural schematic diagram of a roadside millimeter-wave radar calibration system based on Internet of Vehicles provided by the present invention; Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] In view of the limitations of the existing technology, the present invention proposes a roadside millimeter-wave radar calibration method based on vehicle networking technology, and uses a local high-precision roadside map to verify the calibration results, which can more conveniently and accurately realize the rapid calibration of the roadside millimeter-wave radar.

[0026] Figure 1 This is one of the flow charts of the roadside millimeter wave radar calibration method based on the Internet of Vehicles provided by the present invention, such as Figure 1 As shown, including: Step 1, collecting vehicle data within the millimeter wave radar detection area; Step 2, clustering the vehicle data based on the density clustering algorithm OPTICS to extract vehicle trajectory points; Step 3, using a speed matching method to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory point; Step 4, obtaining the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resampling the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using a linear interpolation method to obtain a resampled motion trajectory of the intelligent networked vehicle; Step 5, selecting any group of intelligent networked vehicle collection data from the vehicle data to execute steps 1 to 4, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters using a pseudo-inverse transformation method; Step 6: Use the calibration parameters to calibrate the remaining groups of intelligent connected vehicle collected data in the vehicle data, and verify the calibration results.

[0027] Specifically, the purpose of calibrating the roadside millimeter wave radar in the embodiment of the present invention is to obtain the corresponding WGS-84 coordinates of the detected target. The detailed technical route is as follows: Figure 2 shown.

[0028] Step 1: Data collection. Obtain vehicle data (including smart connected vehicles and ordinary vehicles) within the millimeter wave radar detection area, use vehicle to everything (V2X) communication technology to obtain the basic safety message (BSM) information of smart connected vehicles, and use two smart connected vehicles to collect two sets of data.

[0029] Step 2: Clustering of vehicle trajectory points. Using the density clustering algorithm (Ordering Points To Identify the Clustering Structure, OPTICS) clustering method, we can distinguish vehicles traveling in the same lane at different times, extract the trajectories of connected vehicles, and achieve vehicle trajectory clustering.

[0030] Step 3: Identify the track of the intelligent connected vehicle. Using the automatic speed matching method, the precise speed of the intelligent connected vehicle is obtained through V2X communication technology, and matched with the speed of all vehicles in the detection area measured by the millimeter wave radar, and the corresponding track data of the intelligent connected vehicle is automatically and quickly found.

[0031] Step 4: Data resampling. The frequency of millimeter-wave radar detection data is inconsistent with the frequency of the BSM message sent by the on-board unit (OBU). In order to synchronize the millimeter-wave radar data and the WGS-84 coordinates of the connected vehicle received by the road side unit (RSU), the linear interpolation method is used to resample the trajectory of the intelligent connected vehicle in the millimeter-wave radar coordinate system.

[0032] Step 5: Solve the calibration parameters. Select the first group of intelligent connected vehicle collection data to perform steps 2-4, then randomly select N training points, perform coordinate system transformation, and use the pseudo-inverse transformation method to solve the calibration parameters.

[0033] Step 6: Verification and evaluation. Use the calibration results to calibrate the second set of millimeter-wave radar data, calculate the distance between the WGS-84 coordinates mapped after calibration and the lane centerline of the local high-precision map, and determine the size of the average distance difference and the preset error value. If the average distance difference is not greater than the preset error, the millimeter-wave radar calibration is qualified; otherwise, return to step 1 and recalibrate.

[0034] The present invention uses V2X communication technology to calibrate the roadside millimeter-wave radar, and uses the local high-precision map of the roadside to verify the calibration results. There is no need to close the road during the calibration process, which solves the traffic risks caused by the exposure of personnel or the closure of roads in traditional calibration methods, and improves the efficiency and safety of the calibration work. Compared with the use of neural network methods to calibrate millimeter-wave radars, the method of the present invention has a smaller computational cost and storage capacity, and can be deployed in the roadside edge computing unit to realize the rapid and accurate calibration of the roadside millimeter-wave radar.

[0035] Based on the above embodiment, step 1 includes: Obtain target information of multiple vehicles traveling in different lanes, the target information including timestamp, relative position, speed and identity information, the multiple vehicles including intelligent networked vehicles and ordinary vehicles; The WGS-84 coordinates of the plurality of vehicles are collected.

[0036] Specifically, the embodiment of the present invention uses two intelligent connected vehicles, which travel in the left lane for the first time and in the right lane for the second time. The vehicles try to maintain a constant speed in the middle of the lane, and two sets of data are obtained in this process.

[0037] When a smart connected vehicle passes through the detection area of ​​the roadside millimeter-wave radar, the millimeter-wave radar will obtain information about the vehicle, including smart connected vehicles and ordinary vehicles. The target information obtained by the millimeter-wave radar, including timestamp, relative position, speed and identity information, will be sent to the edge computing unit (Mobile Edge Computing, MEC) through the optical fiber network. At the same time, the WGS-84 coordinates of the smart connected vehicle will also be sent from the RSU to the MEC. The RSU receives the vehicle WGS-84 coordinates sent by the OBU carried by the smart connected vehicle through V2X communication.

[0038] Based on the above embodiment, step 2 includes: Obtaining the x-coordinate and y-coordinate of the vehicle trajectory point at any two moments in the millimeter wave radar coordinate system, as well as the velocity of the trajectory point at the said any two moments; Based on OPTICS, the x-coordinate and y-coordinate at any two moments and the speed of the track point at any two moments are calculated to obtain the travel distance between any two moments; Determine the accessibility distance, and extract the vehicle trajectory point according to the travel distance between the arbitrary two time points and the accessibility distance.

[0039] Specifically, since the tracking object of the millimeter-wave radar in actual scenarios is a dynamic target, there is a lot of noise in the acquisition of millimeter-wave radar data, so it is impossible to simply use the ID to extract the trajectory of the connected vehicle.

[0040] The embodiment of the present invention uses the OPTICS clustering method to cluster vehicle trajectories. Compared with the traditional K-means algorithm, OPTICS does not need to pre-specify the number of clusters and can identify cluster structures of different densities. In the OPTICS clustering method, a neighborhood radius is first determined, which is the minimum cluster radius that the algorithm can identify; for each data point, its neighborhood-based reachability distance is calculated, and the reachability distance of each data point is recorded; the data points are sorted according to their reachability distances to form a sorted list; clusters are identified and extracted through the sorted list.

[0041] Since there are fewer vehicles traveling in the same lane, the trajectories of these vehicles will overlap in the coordinate system of the millimeter wave radar. Therefore, in the OPTICS method, using Euclidean distance to determine the reachability distance is not suitable for clustering vehicle trajectories. In this paper, the timestamp and speed are used to calculate the reachability distance in the OPTICS method. The distance calculation formula is as follows: (1) In the formula, and are the x and y coordinate values ​​of the vehicle trajectory point in the millimeter wave radar coordinate system, is the track point timestamp, is the velocity of the trajectory point. It can be understood that this method can distinguish vehicles traveling in the same lane at different times.

[0042] Based on the above embodiment, step 3 includes: Get the first moment detection speed of the millimeter wave radar and the second moment reporting speed of the intelligent connected vehicle; Determine a preset time constraint range so that the difference between the first moment and the second moment is within the preset time constraint range; Acquire the total number of trajectory point data within the preset time constraint range, and obtain a speed similarity value of any trajectory point within the preset time constraint range based on the total number of trajectory point data, the detection speed at the first moment, and the reported speed at the second moment; The motion trajectory of the intelligent networked vehicle is constructed using trajectory points whose speed similarity values ​​are within a similarity threshold.

[0043] Specifically, after clustering the vehicle trajectories, the embodiment of the present invention uses a speed matching method to quickly identify the motion trajectory of the intelligent connected vehicle. The millimeter wave radar can measure the speed of all vehicles in the detection area, and the RSU can also obtain the precise speed of the intelligent connected vehicle through V2X communication technology. By measuring the speed similarity between each clustered vehicle and the intelligent connected vehicle, the trajectory of the intelligent connected vehicle can be automatically and quickly identified. The calculation formula is as follows: (2) In the formula, Millimeter wave radar The speed detected at any moment, Is the smart connected car in The speed of reporting at all times, Indicates that the time constraint The total number of track points. Assuming that the speed of the intelligent connected vehicle remains constant within 100ms, the calculation result , it can be determined that the clustered vehicle trajectories are the trajectories of intelligent connected vehicles.

[0044] Based on the above embodiment, step 4 includes: Obtaining a plurality of first moment coordinates and a plurality of next moment coordinates of the first moment of the motion trajectory of the intelligent networked vehicle in the millimeter wave radar coordinate system; Performing linear interpolation calculation on the plurality of first moment coordinates and the plurality of first moment coordinates next to the plurality of first moment coordinates to obtain a plurality of second moment coordinates; The resampled motion trajectory of the intelligent connected vehicle is constructed by integrating all the second-time coordinates to obtain the millimeter-wave radar data with the same timestamp and the corresponding WGS-84 coordinates.

[0045] Specifically, the embodiment of the present invention takes the Continental millimeter-wave radar as an example, and outputs the detection results of the object every 72ms in the millimeter-wave radar coordinate system. The origin of the millimeter-wave radar coordinate system is the position of the roadside millimeter-wave radar, and the positive directions of the x-axis and y-axis of the coordinate are the front and connector side of the millimeter-wave radar, respectively. The roadside RSU and OBU are both taken as Hikvision Intelligent Connection equipment, and the OBU broadcasts the vehicle positioning information to the RSU every 100ms. In order to synchronize the millimeter-wave radar data and the vehicle positioning information received by the RSU, the linear interpolation method is used to resample the trajectory points in the millimeter-wave radar coordinate system. Assuming that the speed of the intelligent connected vehicle is constant during the millimeter-wave radar detection cycle, it is known that the millimeter-wave radar data is at time and The x-coordinate of The x-coordinate of the moment, The y coordinate at the moment is obtained using the same method, and the interpolation formula is as follows: (3) After resampling the trajectory point data, a set of millimeter wave radar data with the same timestamp is obtained. and its corresponding WGS-84 coordinates .

[0046] Based on the above embodiment, step 5 includes: The embodiment of the present invention randomly selects 10 points from the time-synchronized trajectory data set as training data to solve the calibration parameters. First, the WGS-84 coordinate data of the intelligent networked vehicle is transformed, and the coordinate data in the WGS-84 coordinate system is transformed into the local Cartesian coordinate system, and then a transformation relationship is established with the laser radar coordinate system, and finally the PI pseudo-inverse transformation method is used to solve the coordinate transformation matrix.

[0047] (1) Coordinate system conversion First, convert the WGS-84 coordinates of the intelligent connected vehicle into local Cartesian coordinates. The WGS-84 coordinate system is a geodetic coordinate system. Before converting to the local Cartesian coordinate system, first convert the WGS-84 coordinate system into the Earth-centered Earth-fixed rectangular coordinate system (ECEF). The conversion formula is as follows: (4)

[0048] In the formula, are the major and minor radii of the reference ellipsoid, is the square of the eccentricity of the ellipsoid, is the radius of curvature of the reference ellipsoid, are longitude, latitude and altitude respectively. The altitude of the target is not considered in the present invention. Set to 0.

[0049] Then convert the Earth-centered Earth-fixed rectangular coordinate system (ECEF) to the local Cartesian coordinate system (ENU) using the following conversion formula: (5) It should be noted that this process requires a local reference point , which can use the WGS-84 coordinates of the reference point By using formula (4), the reference point can be selected as the location of the millimeter wave radar.

[0050] According to the above process, the WGS-84 coordinates can be Convert to local Cartesian coordinates (ENU) .

[0051] (2) Solving the calibration parameters The calibration parameters are solved using the pseudo inverse (PI) method. The calibration formula is as follows: (6)

[0052] In the formula, are the millimeter wave radar coordinates, , is a two-dimensional change matrix. Select For the data, the roadside calibration problem is formulated as an optimization problem of minimizing the projection error on the local coordinates: (7) In the formula, yes The distance between The parameters can be solved using PI.

[0053] Based on the above embodiment, step 6 includes: The embodiment of the present invention selects the collected data of the first vehicle, performs vehicle trajectory point clustering, trajectory line recognition, data resampling, and solves calibration parameters, and uses the millimeter-wave radar calibration parameters to calibrate the collected data of the second vehicle for verification and evaluation.

[0054] (1) Extracting lane centerline High-precision maps using the ASAM OpenDRIVE standard, which uses concepts such as sections, lanes, lane segments, and intersections to define road features, such as Figure 3 As shown, the precise WGS-84 coordinates of the centerline of each lane are obtained.

[0055] (2) Calculation error The distance between the calibrated mapped WGS-84 coordinates and the high-precision map lane centerline WGS-84 coordinates is calculated using formula (8) to calculate the foot point on the lane centerline, and then the Vincenty formula is used to determine the distance. This distance is used to measure the millimeter wave radar calibration accuracy. (8) In the formula, Indicates the foot point, Indicates the WGS-84 coordinates mapped after calibration. Indicates the center line of the lane. Vehicle positioning needs to meet the lane-level positioning accuracy in V2X applications, and the absolute positioning accuracy must be within 1.5m.

[0056] The predicted error value is 1.5m. The average distance error and the preset error value are determined. If the average distance difference is not greater than the preset error, the millimeter wave radar calibration is qualified; otherwise, recalibrate.

[0057] The roadside millimeter-wave radar calibration system based on the Internet of Vehicles provided by the present invention is described below. The roadside millimeter-wave radar calibration system based on the Internet of Vehicles described below and the roadside millimeter-wave radar calibration method based on the Internet of Vehicles described above can be referenced to each other.

[0058] Figure 4 is a schematic diagram of the structure of a roadside millimeter-wave radar calibration system based on the Internet of Vehicles provided by an embodiment of the present invention, such as Figure 4 As shown, it includes: a collection module 41, a clustering module 42, a recognition module 43, a resampling module 44, a calibration module 45 and a verification module 46, wherein: The acquisition module 41 is used to collect vehicle data within the millimeter-wave radar detection area; the clustering module 42 is used to cluster the vehicle data based on the density clustering algorithm OPTICS to extract vehicle trajectory points; the identification module 43 is used to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory points by using the speed matching method; the resampling module 44 is used to obtain the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resample the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using the linear interpolation method to obtain the resampled motion trajectory of the intelligent networked vehicle; the calibration module 45 is used to select any group of intelligent networked vehicle acquisition data in the vehicle data to perform clustering, identification and resampling steps, randomly select a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and use the pseudo-inverse transformation method to solve the calibration parameters; the verification module 46 is used to use the calibration parameters to calibrate the remaining groups of intelligent networked vehicle acquisition data in the vehicle data, and verify the calibration results.

[0059] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute the roadside millimeter-wave radar calibration method based on the Internet of Vehicles, which includes: step 1, collecting vehicle data within the millimeter-wave radar detection area; step 2, clustering the vehicle data based on the density clustering algorithm OPTICS to extract vehicle trajectory points; step 3, using the speed matching method to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory points; step 4, obtaining the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resampling the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by linear interpolation to obtain the resampled motion trajectory of the intelligent networked vehicle; step 5, selecting any group of intelligent networked vehicle collection data in the vehicle data to execute steps 1 to 4, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters by pseudo-inverse transformation method; step 6, using the calibration parameters to calibrate the remaining groups of intelligent networked vehicle collection data in the vehicle data, and verifying the calibration results.

[0060] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0061] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0062] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A roadside millimeter-wave radar calibration method based on vehicle networking, characterized in that: include: Step 1, collecting vehicle data within the millimeter wave radar detection area; Step 2, clustering the vehicle data based on the density clustering algorithm OPTICS to extract vehicle trajectory points; Step 3, using a speed matching method to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory point; Step 4, obtaining the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resampling the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using a linear interpolation method to obtain a resampled motion trajectory of the intelligent networked vehicle; Step 5, selecting any group of intelligent networked vehicle collection data from the vehicle data to execute steps 1 to 4, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters using a pseudo-inverse transformation method; Step 6: Use the calibration parameters to calibrate the remaining groups of intelligent connected vehicle collected data in the vehicle data, and verify the calibration results.

2. The roadside millimeter-wave radar calibration method based on the Internet of Vehicles according to claim 1 is characterized in that: Step 1 includes: Obtain target information of multiple vehicles traveling in different lanes, the target information including timestamp, relative position, speed and identity information, the multiple vehicles including intelligent networked vehicles and ordinary vehicles; The WGS-84 coordinates of the plurality of vehicles are collected.

3. The roadside millimeter-wave radar calibration method based on the Internet of Vehicles according to claim 1 is characterized in that: Step 2 includes: Obtaining the x-coordinate and y-coordinate of the vehicle trajectory point at any two moments in the millimeter wave radar coordinate system, as well as the velocity of the trajectory point at the any two moments; Based on OPTICS, the x-coordinate and y-coordinate at any two moments and the speed of the track point at any two moments are calculated to obtain the travel distance between any two moments; Determine the accessibility distance, and extract the vehicle trajectory point according to the travel distance between the arbitrary two time points and the accessibility distance.

4. The method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles according to claim 1, characterized in that: Step 3 includes: Get the first moment detection speed of the millimeter wave radar and the second moment reporting speed of the intelligent connected vehicle; Determine a preset time constraint range so that the difference between the first moment and the second moment is within the preset time constraint range; Acquire the total number of trajectory point data within the preset time constraint range, and obtain a speed similarity value of any trajectory point within the preset time constraint range based on the total number of trajectory point data, the detection speed at the first moment, and the reported speed at the second moment; The motion trajectory of the intelligent networked vehicle is constructed using trajectory points whose speed similarity values ​​are within a similarity threshold.

5. The method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles according to claim 4, characterized in that: Step 4 includes: Obtaining a plurality of first moment coordinates and a plurality of next moment coordinates of the first moment of the motion trajectory of the intelligent networked vehicle in the millimeter wave radar coordinate system; Performing linear interpolation calculation on the plurality of first moment coordinates and the plurality of first moment coordinates next to the plurality of first moment coordinates to obtain a plurality of second moment coordinates; The resampled motion trajectory of the intelligent connected vehicle is constructed by integrating all the second-time coordinates to obtain the millimeter-wave radar data with the same timestamp and the corresponding WGS-84 coordinates.

6. The method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles according to claim 1, characterized in that: In step 5, a plurality of training points are randomly selected from the resampled motion trajectory of the intelligent networked vehicle for coordinate system transformation, including: Convert WGS-84 coordinate system to ECEF Earth-centered rectangular coordinate system: in, are the major and minor radii of the reference ellipsoid, is the square of the eccentricity of the ellipsoid, is the radius of curvature of the reference ellipsoid, are longitude, latitude and altitude respectively. , , They are the three-dimensional coordinates of the Earth-centered, Earth-fixed rectangular coordinate system ECEF; Convert the Earth-centered rectangular coordinate system ECEF to the local Cartesian coordinate system ENU: in, , , They are the three-dimensional coordinates of the local Cartesian coordinate system ENU, They are the 3D coordinates of the local reference point and the WGS-84 coordinates with the millimeter-wave radar position as the reference point. what you get; Convert WGS-84 coordinates Convert to local Cartesian coordinates ENU .

7. The method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles according to claim 6, characterized in that: The pseudo-inverse transformation method is used in step 5 to solve the calibration parameters, including: in, are the millimeter wave radar coordinates, , is a two-dimensional change matrix; from Select For data, , the roadside calibration problem is formulated as an optimization problem of minimizing the projection error on the local coordinates: in, yes The distance between The parameters of are solved using the pseudo-inverse transformation method.

8. The method for calibrating a roadside millimeter-wave radar based on the Internet of Vehicles according to claim 1, characterized in that: Step 6 includes: Obtain the standard WGS-84 coordinates of each lane centerline from the preset high-precision map; The calibrated mapped WGS-84 coordinates are obtained based on the collected data of the remaining groups of intelligent networked vehicles, and the average distance error between the calibrated mapped WGS-84 coordinates and the standard WGS-84 coordinates is obtained by calculating the foot point of the lane centerline and the Vincenty distance formula: in, Indicates the foot point, Indicates the WGS-84 coordinates mapped after calibration. Indicates the center line of the lane; If it is determined that the average distance error is less than the preset error value, the calibration is qualified, otherwise recalibration is performed.

9. A roadside millimeter-wave radar calibration system based on vehicle networking, characterized in that: include: The acquisition module is used to collect vehicle data within the detection area of ​​the millimeter-wave radar; A clustering module, used for clustering the vehicle data based on the density clustering algorithm OPTICS, and extracting vehicle trajectory points; An identification module, used to identify and obtain the motion trajectory of the intelligent networked vehicle in the vehicle trajectory point by using a speed matching method; A resampling module, used to obtain the millimeter-wave radar coordinate system of the motion trajectory of the intelligent networked vehicle, and resample the motion trajectory of the intelligent networked vehicle in the millimeter-wave radar coordinate system by using a linear interpolation method to obtain a resampled motion trajectory of the intelligent networked vehicle; A calibration module, used for selecting any group of intelligent networked vehicle collected data from the vehicle data to perform clustering, identification and resampling steps, randomly selecting a number of training points from the resampled motion trajectory of the intelligent networked vehicle to perform coordinate system transformation, and solving calibration parameters using a pseudo-inverse transformation method; The verification module is used to calibrate the remaining groups of intelligent networked vehicle collected data in the vehicle data using the calibration parameters and verify the calibration results.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, it implements the roadside millimeter-wave radar calibration method based on the Internet of Vehicles as described in any one of claims 1 to 8.

Images