CN116229713A - A spatial alignment method and system for vehicle-road coordination - Google Patents

Abstract

The invention relates to a space alignment method and a system for vehicle-road cooperation, wherein the method comprises the following steps: s1, respectively deploying intelligent network link side subsystems and intelligent network link vehicle-mounted subsystems at a road side and a vehicle end, wherein the subsystems are in road cooperation through V2X; s2, the intelligent network link side subsystem and the intelligent network link vehicle subsystem are used for carrying out space alignment among the subsystems by constructing a ground WGS84 coordinate system respectively; s3, constructing a northeast day coordinate system by the intelligent network link side subsystem to align the space inside the subsystem, and then converting the northeast day coordinate system into a ground WGS84 coordinate system; s4, the intelligent network vehicle-mounted subsystem builds a northeast day coordinate system to align the space inside the vehicle-mounted subsystem, and then the northeast day coordinate system is converted into a ground WGS84 coordinate system based on the vehicle-mounted GPS/Beidou. Compared with the prior art, the method ensures the spatial alignment of the vehicle-road cooperative system in the information fusion process, constructs a unified coordinate system for vehicle-road cooperative sensing, and improves the positioning and tracking precision of the sensing target object.

Description

A spatial alignment method and system for vehicle-road coordination

technical field

The invention relates to the technical field of intelligent transportation, in particular to a spatial alignment method and system for vehicle-road coordination.

Background technique

Vehicle-road coordination technology is an important supporting technology for road safety, and it is also a basic research issue for realizing intelligent transportation. At present, many automobile manufacturers are committed to the development of intelligent transportation system ITS (Intelligent Transportation Systems), among which the key technology of intelligent vehicle road coordination is the hot spot and cutting-edge technology of ITS research. At present, countries around the world are actively conducting research and experiments on the vehicle-road coordination system, and regard it as an important means to improve road traffic safety and efficiency. In foreign countries, European automobile companies have already applied the Internet of Vehicles technology to the intelligent management of fleets. At the same time, European passenger transport companies are also actively promoting the application of Internet of Vehicles technology. Systems such as IVHS in the United States and VICS in Japan also establish effective information communication between vehicles and roads, thereby realizing intelligent traffic management and information services. Although these information services have different focuses, they are all based on vehicle-road coordination technology.

The vehicle-road coordination system is supported by "advanced vehicles" and "smart roads", and realizes the integrated construction of vehicle-road cloud through the communication platform to meet various application requirements of vehicle-road coordination. "Advanced vehicles" refer to smart cars equipped with sensing equipment, but the perception accuracy and range of single-vehicle intelligence are limited, such as pedestrians blocked by buses; while "smart roads" Advanced sensors collect and share real-time information on roads and traffic participants. Due to the high installation position of roadside sensors, they can approximately provide a wide field of vision, thereby making up for the blind spots of vehicle-end sensing equipment; in addition, edge sensors can also be deployed on the roadside Cloud to reduce the perceived cost of the vehicle. Through vehicle-road collaborative sensing, the sensing range and sensing ability of vehicles can be greatly increased, which in turn helps to reduce traffic accidents, protect the personal safety and property safety of traffic participants on the highway, avoid road congestion, and improve the transportation environment. Before carrying out the vehicle-road collaborative perception fusion, it is necessary to align the vehicle-road space first, and integrate the information sensed by the vehicle and the roadside in a unified coordinate system. If there is no spatial alignment, errors will easily occur when associating the objects or trajectories perceived by the car and road, which will affect the fusion effect.

Contents of the invention

The purpose of the present invention is to provide a spatial alignment method and system for vehicle-road coordination in order to overcome the above-mentioned defects in the prior art, to ensure the spatial alignment of the vehicle-road coordination system during the information fusion process, and to provide a vehicle-road collaborative perception Construct a unified coordinate system to improve the positioning and tracking accuracy of the perceived target.

The purpose of the present invention can be achieved through the following technical solutions:

According to the first aspect of the present invention, a spatial alignment method for vehicle-road coordination is provided, the method includes the following steps:

Step S1. Deploy the intelligent networked roadside subsystem and the intelligent networked vehicle subsystem on the roadside and the vehicle respectively, wherein the vehicle-road coordination is carried out between the subsystems through V2X;

Step S2, the intelligent network-connected roadside subsystem and the intelligent network-connected vehicle-mounted subsystem respectively carry out the spatial alignment between the subsystems by constructing the earth WGS84 coordinate system;

Step S3, the intelligent network-connected roadside subsystem constructs the northeast sky coordinate system for spatial alignment inside the roadside subsystem, and then converts the northeast sky coordinate system to the earth WGS84 coordinate system;

Step S4. The intelligent networked vehicle subsystem constructs the northeast sky coordinate system for spatial alignment inside the vehicle subsystem, and then converts the northeast sky coordinate system to the earth WGS84 coordinate system based on the vehicle GPS/Beidou.

Preferably, the intelligent networked roadside subsystem includes integrated roadside sensing equipment, roadside computing equipment, and roadside communication equipment; the roadside sensing equipment includes cameras, laser radars, and millimeter-wave radars.

Preferably, the intelligent internet-connected vehicle subsystem includes an integrated GPS/Beidou module, a vehicle perception device, a vehicle computing device and a vehicle communication device; the vehicle perception device includes a camera, a laser radar and a millimeter wave radar.

Preferably, the spatial alignment processes corresponding to the millimeter-wave radar, ultrasonic radar, lidar and camera are respectively:

Millimeter-wave radar: convert from the millimeter-wave radar coordinate system to the northeast sky coordinate system, specifically: install the millimeter-wave radar on the roadside, and convert the millimeter-wave radar coordinate system to the northeast sky coordinate system through the calibration of the millimeter-wave radar external parameters;

Lidar: Convert from the Lidar coordinate system to the Northeast sky coordinate system, specifically: fix the Lidar on the roadside, make the compass direction point to the X-axis in the Lidar coordinate system and record the angle θ of the coordinate system, and use the minimum Calculate the transformation matrix from the three-dimensional point cloud coordinate system of the lidar to the northeast sky coordinate system by the square multiplication method;

Camera: first perform internal reference calibration on the camera, then convert from the pixel coordinate system to the lidar coordinate system through joint calibration, and then convert from the lidar coordinate system to the northeast sky coordinate system.

Preferably, the internal reference calibration of the camera is specifically:

1) Construct the camera imaging model, the mathematical expression is:

为相机的内参矩阵，其中，f为像距，dX、dY分别为X、Y方向上的一个像素在相机感光板上的物理长度，u_O、v_O分别为相机感光板中心在像素坐标系下的坐标，θ表示感光板的横边和纵边之间的角度；In the formula, (U, V, W) is the physical coordinate of a point in the world coordinate system, (u, v) is the pixel coordinate in the pixel coordinate system corresponding to (U, V, W), and Z is the scale factor ;

is the internal reference matrix of the camera, where f is the image distance, dX and dY are the physical lengths of a pixel on the photosensitive plate of the camera in the X and Y directions respectively, u _O and v _O are the center of the photosensitive plate of the camera in the pixel coordinate system Under the coordinates, θ represents the angle between the transverse and longitudinal sides of the photosensitive plate;

2) Collect the picture of the checkerboard calibration board of the camera to be calibrated, and use the camera internal reference calibration principle to calculate the camera internal reference matrix and distortion coefficient.

Preferably, said conversion from pixel coordinate system to lidar coordinate system through joint calibration includes the following sub-steps:

1) Place markers in the common viewing area of the camera and lidar, and collect calibration data:

2) Obtain the pixel coordinates of the marker in the camera and the coordinates in the lidar point cloud respectively;

3) Use the pose algorithm to solve the transformation matrix from the lidar coordinate system to the camera coordinate system;

4) Extract the point cloud road surface number, and partition the road surface point cloud, then fit each point cloud partition, upsample the road surface point cloud, and generate point cloud-pixel coordinate point pairs;

5) Use the generated point cloud-pixel coordinate point pairs as a data set, use artificial intelligence algorithms to perform multiple regression prediction of 3D point cloud coordinates, and convert pixel coordinates to point cloud coordinates;

6) Based on the transformation from the lidar coordinate system to the northeast sky coordinate system, the pixel coordinates are converted to the northeast sky coordinates.

Preferably, the step 3) specifically includes: solving the conversion matrix from the lidar coordinate system to the camera coordinate system by using the EPnP algorithm.

Preferably, the artificial intelligence method in step 5) includes decision tree and random forest.

Preferably, in the step S3 or S4, the northeast sky coordinate system is converted to the earth WGS84 coordinate system, and the conversion expression is:

In the formula, (x, y) is the coordinate of any point in the northeast sky coordinate system, (lat ₁ , lon ₁ ) is the longitude and latitude of (x, y) in the WGS84 coordinate system, (lat ₀ , lon ₀ ) is the northeast sky coordinate The latitude and longitude of the origin, R is the radius of the earth.

According to the second aspect of the present invention, a space alignment system for vehicle-road coordination is provided, the system includes:

The intelligent networked roadside subsystem installed on the roadside, the intelligent networked vehicle subsystem installed on the vehicle, and the intelligent networked vehicle subsystem and the intelligent networked roadside subsystem perform vehicle-road coordination through V2X;

Wherein, the method described in any one is used for spatial alignment in the vehicle-road coordination process.

Compared with the prior art, the present invention has the following advantages:

1) Designed for the need of time-space synchronization for vehicle-road coordination, through the spatial alignment mechanism, even for different subsystems, such as vehicle-mounted subsystems or roadside subsystems, the coordinates of each target object in a unified coordinate system can be obtained The location state creates favorable conditions for the vehicle-road collaborative fusion based on the location point, ensures the spatial alignment of the vehicle-road collaborative system in the process of information fusion, and can build a unified coordinate system for the vehicle-road collaborative perception, thereby helping to improve Perceive the positioning and tracking accuracy of the target;

2) The present invention establishes the northeast sky coordinate system respectively for different sensing devices, and improves the accuracy of position coordinate conversion through the joint calibration method;

3) The present invention adopts artificial intelligence algorithm to realize the mapping of three-dimensional point cloud to image pixel, so that the accuracy of spatial alignment is higher;

4) The present invention can be further extended to an integrated vehicle, road, and cloud system, which has strong scalability and high practicability.

Description of drawings

FIG. 1 is a structural diagram of a spatial alignment method and system for vehicle-road coordination according to the present invention;

FIG. 2 is a structural diagram of the intelligent network roadside subsystem of the spatial alignment method and system for vehicle-road coordination according to the present invention;

Fig. 3 is the upsampling process of the laser radar point cloud according to the spatial alignment method and system for vehicle-road coordination according to the present invention;

4 is a schematic diagram of the conversion from the radar coordinate system to the northeast sky coordinate system according to the spatial alignment method and system for vehicle-road coordination according to the present invention;

Fig. 5 is a schematic diagram of checkerboard acquisition according to the spatial alignment method and system for vehicle-road coordination according to the present invention; wherein, Fig. 5a and Fig. 5b are respectively different angle acquisition diagrams.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

Example

Referring to Figures 1-5, this embodiment presents a spatial alignment system for vehicle-road coordination, which includes: an intelligent networked roadside subsystem installed on the roadside and an intelligent networked vehicle-mounted subsystem installed on the vehicle subsystem, and the vehicle-road coordination system that integrates the intelligent network roadside subsystem and the intelligent network vehicle subsystem through V2X. As a preferred technical solution, the vehicle-road coordination system can be further extended to an integrated vehicle, road, and cloud system.

The intelligent network roadside subsystem includes roadside sensing equipment, roadside computing equipment and roadside communication equipment, among which: roadside computing equipment includes MEC server, industrial computer and computer, roadside sensing equipment includes cameras, laser radar, millimeter Wave radar or other sensors such as microwave radar, roadside communication equipment including 5G/4G base stations, V2X RSU;

Roadside sensing devices, roadside computing devices, and roadside communication devices are integrated through routers, switches, or point-to-point networks. The spatial alignment process is shown in Figure 2. For millimeter-wave radar, from the millimeter-wave radar coordinate system Convert to the northeast sky coordinate system, the microwave radar is the same as the millimeter-wave radar; for the lidar, convert from the lidar coordinate system to the northeast sky coordinate system; for the camera, first perform internal reference calibration on the camera, and then convert from the pixel coordinate system through joint calibration To the lidar coordinate system, and then convert from the lidar coordinate system to the northeast sky coordinate system.

The intelligent networked vehicle subsystem includes a GNSS module compatible with Beidou/GPS, a combined navigation and positioning system integrating RTK and IMU, vehicle-mounted perception equipment, vehicle-mounted computing equipment, and vehicle-mounted communication equipment; GPS/Beidou includes RTK high-precision positioning.

Vehicle-mounted computing equipment includes embedded controllers and industrial computers, and vehicle-mounted communication equipment includes 5G/4G CPE and V2XOBU;

Vehicle-mounted perception equipment includes cameras, lidars, millimeter-wave radars, or other sensors such as ultrasonic radars; the spatial alignment process is: for millimeter-wave radars, transform from the millimeter-wave radar coordinate system to the northeast sky coordinate system; for ultrasonic radars, from ultrasonic radar to Convert the coordinate system to the northeast sky coordinate system; for the lidar, convert from the lidar coordinate system to the northeast sky coordinate system; for the camera, first perform internal reference calibration on the camera, and then convert from the pixel coordinate system to the lidar coordinate system through joint calibration, Then convert from the lidar coordinate system to the northeast sky coordinate system.

Transform from the lidar coordinate system to the northeast sky coordinate system, including the following steps:

The conversion between the lidar coordinate system and the northeast sky coordinate system is a rigid transformation of the three-dimensional Cartesian coordinate system, and the conversion formula is as follows:

Among them, X _ENU , Y _ENU , and Z _ENU are the coordinates of the northeast sky, and X _Lidar , Y _Lidar , and Z _Lidar are the coordinates of the lidar. By calibrating the lidar, when the corresponding coordinates of several points in the two coordinate systems are known, the transformation matrix can be obtained by using the least square method.

Calibrate the internal reference of the camera, including the following steps:

The camera imaging model can be described by the following formula:

Among them, (U, V, W) are the physical coordinates of a certain point in the world coordinate system, (u, v) are the pixel coordinates corresponding to the point in the pixel coordinate system, and Z is the scale factor.

matrix:

It is called the internal reference matrix of the camera, wherein, f is the image distance, dX and dY are respectively the physical length of a pixel on the photosensitive plate of the camera in the X and Y directions (that is, how many millimeters a pixel is on the photosensitive plate), u _O , v _O are the coordinates of the center of the photosensitive plate of the camera in the pixel coordinate system, and θ represents the angle between the horizontal and vertical sides of the photosensitive plate (90° means no error).

First make a checkerboard calibration board, measure the size of a grid in the checkerboard;

Put the checkerboard in the camera field of view, move the checkerboard continuously, and ensure that the checkerboard appears in the top, bottom, left, right, front, back, and front of the camera field of view, and has inclination angles in different directions. Take about 30 checkerboard images, as shown in Figure 5;

Use the Python language to process the collected checkerboard image using Zhang Zhengyou’s calibration method, calculate the camera’s internal parameter matrix and distortion coefficient, and use the obtained distortion coefficient to perform de-distortion processing on the image, compare the changes before and after de-distortion of the image, and display the de-distortion of the image. Distorted effect.

Converting from the pixel coordinate system to the lidar coordinate system through joint calibration includes the following steps:

(1) Calibration data collection. Markers are placed in the common viewing area of the camera and the lidar. In order to clearly distinguish the position of the marker in the point cloud, a high-reflectivity reflective strip can be attached to the marker, such as a hand-held device with a high-reflectivity reflective strip. GPS calibration rod. When ROS is used to record lidar point cloud data and image data at the same time, when using the EPnP algorithm to solve the joint calibration parameters, theoretically only 4 feature points are needed to solve the calibration parameters. Due to the error introduced by the reason, more than 4 feature points can be collected to solve the calibration parameters of the lidar and camera;

(2) Obtain the pixel coordinates of the marker in the camera. Since the recorded camera data is in ROS format, the Python programming language is first used to extract pictures from the camera data in ROS format in order to select the pixels in the picture. After obtaining the picture, use the C++ programming language to extract the pixel coordinates from the picture. When using it, pay attention to accurately select the position of the reflective strip attached to the marker to ensure that the extracted pixels represent the real position of the marker;

(3) Obtain the coordinates of the marker in the lidar point cloud. Use the ROS tool to play back the collected lidar calibration data, and visualize the lidar point cloud data in Rviz. Due to the high reflectivity of the reflective strip on the marker, its color will be significantly different from the surrounding color in Rviz. Use the Publish Point tool Select the position of the marker in the point cloud, and extract the coordinates of the marker in the lidar point cloud;

(4) After obtaining the coordinates of multiple feature points in the lidar coordinate system and the camera coordinate system, use the C++ programming language to iteratively optimize the EPnP algorithm to solve the conversion matrix from the lidar coordinate system to the camera coordinate system, and calculate the pixel reprojection Error, to verify the accuracy of the calibration results.

(5) Extract point cloud road surface data. Visualize the recorded laser radar point cloud calibration data in Rviz, use the Publish Point tool to select the point cloud of the road surface, record the coordinates on the road surface in the point cloud, and obtain the road surface point cloud data;

(6) Partition the road surface point cloud. Visualize the road surface point cloud data in Rviz, adjust the point cloud perspective, partition the point cloud according to the cross section of the road surface point cloud and the coordinates of the road surface point cloud, and divide the point cloud area approximately in a plane into a point cloud partition, Divide the entire road surface point cloud into multiple point cloud partitions, and use the Publish Point tool to save the corner points of each point cloud partition in a file;

(7) Fit each point cloud partition and upsample the road surface point cloud. Use the Python programming language to implement the least squares method to fit the plane algorithm, use the plane to fit each point cloud partition, and after obtaining the plane equation of each point cloud partition, generate a point at an interval of 1cm in the x-axis and y-axis directions, and add points Cloud density, to achieve upsampling of point clouds;

(8) Generate point cloud-pixel coordinate point pairs. Use the C++ programming language to realize the conversion from point cloud coordinates to pixel coordinates, and map the point cloud coordinates to pixel coordinates according to the results of lidar-camera joint calibration to generate point cloud-pixel coordinate point pairs;

(9) Convert pixel coordinates to point cloud coordinates. Use the Python programming language to implement decision tree regression or random forest regression algorithms, use the generated point cloud-pixel coordinate point pairs as a data set, realize multiple regression for predicting 3D point cloud coordinates, and realize the conversion from pixel coordinates to point cloud coordinates;

(10) Using the previous conversion from the lidar coordinate system to the northeast sky coordinate system, the conversion from pixel coordinates to northeast sky coordinates can be further realized.

Note that the embodiments of the present invention are not limited to joint calibration based on EPnP, and joint calibration can also be achieved by using methods such as P3P, direct linear transformation (DLT), and BA (Bundle Adjustment) to estimate poses of three pairs of points.

The spatial alignment between the intelligent networked roadside subsystem and the intelligent networked vehicle subsystem is achieved by converting the intelligent networked roadside subsystem and the intelligent networked vehicle subsystem from the northeast sky coordinate system to the WGS84 coordinate system, including The following steps:

As shown in Figure 4, suppose the latitude and longitude of the origin O of the northeast sky coordinate system are lat _O and lon _O , and the coordinates of a point in the northeast sky coordinate system are (x′, y′), then the WGS84 coordinates of this point are

Among them, R is the radius of the earth, and lat ₁ and lon ₁ are WGS84 coordinates.

The WGS84 coordinates of this point in the lidar coordinate system can be obtained through the above formula.

The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technical field can easily think of various equivalents within the technical scope disclosed in the present invention. Modifications or replacements shall all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. A spatial alignment method for vehicle-road coordination, characterized in that the method comprises the following steps: Step S1. Deploy the intelligent networked roadside subsystem and the intelligent networked vehicle subsystem on the roadside and the vehicle respectively, wherein the vehicle-road coordination is carried out between the subsystems through V2X; Step S2, the intelligent network-connected roadside subsystem and the intelligent network-connected vehicle-mounted subsystem respectively carry out the spatial alignment between the subsystems by constructing the earth WGS84 coordinate system; Step S3, the intelligent network-connected roadside subsystem constructs the northeast sky coordinate system for spatial alignment inside the roadside subsystem, and then converts the northeast sky coordinate system to the earth WGS84 coordinate system; Step S4. The intelligent networked vehicle subsystem constructs the northeast sky coordinate system for spatial alignment inside the vehicle subsystem, and then converts the northeast sky coordinate system to the earth WGS84 coordinate system based on the vehicle GPS/Beidou. 2. A spatial alignment method for vehicle-road coordination according to claim 1, wherein the intelligent network-connected roadside subsystem includes integrated roadside sensing equipment, roadside computing equipment, and roadside side communication equipment; the roadside perception equipment includes a camera, laser radar and millimeter wave radar. 3. A spatial alignment method for vehicle-road coordination according to claim 1, characterized in that, the intelligent network-linked vehicle subsystem includes an integrated GPS/Beidou module, a vehicle-mounted sensing device, and a vehicle-mounted computing device and vehicle-mounted communication equipment; the vehicle-mounted perception equipment includes cameras, laser radars and millimeter-wave radars. 4. A spatial alignment method for vehicle-road coordination according to claim 2 or 3, wherein the spatial alignment processes corresponding to the millimeter-wave radar, lidar, and camera are respectively: Millimeter-wave radar: convert from the millimeter-wave radar coordinate system to the northeast sky coordinate system, specifically: install the millimeter-wave radar on the roadside, and convert the millimeter-wave radar coordinate system to the northeast sky coordinate system through the calibration of the millimeter-wave radar external parameters; Lidar: Convert from the Lidar coordinate system to the Northeast sky coordinate system, specifically: install the Lidar fixedly, make the direction of the compass point to the X-axis in the Lidar coordinate system and record the angle θ of the coordinate system, and use the least square method to obtain it Transformation matrix from the LiDAR 3D point cloud coordinate system to the northeast sky coordinate system; Camera: first perform internal reference calibration on the camera, then convert from the pixel coordinate system to the lidar coordinate system through joint calibration, and then convert from the lidar coordinate system to the northeast sky coordinate system. 5. A spatial alignment method for vehicle-road coordination according to claim 4, characterized in that the internal reference calibration of the camera is specifically: 1) Construct the camera imaging model, the mathematical expression is:

In the formula, (U, V, W) is the physical coordinate of a point in the world coordinate system, (u, v) is the pixel coordinate in the pixel coordinate system corresponding to (U, V, W), and Z is the scale factor;

is the internal reference matrix of the camera, where f is the image distance, dX and dY are the physical lengths of a pixel on the photosensitive plate of the camera in the X and Y directions respectively, u _O and v _O are the center of the photosensitive plate of the camera in the pixel coordinate system Under the coordinates, θ represents the angle between the transverse and longitudinal sides of the photosensitive plate; 2) Collect the picture of the checkerboard calibration board of the camera to be calibrated, and use the camera internal reference calibration principle to calculate the camera internal reference matrix and distortion coefficient. 6. A spatial alignment method for vehicle-road coordination according to claim 5, wherein said conversion from the pixel coordinate system to the lidar coordinate system through joint calibration comprises the following sub-steps: 1) Place markers in the common viewing area of the camera and lidar, and collect calibration data: 2) Obtain the pixel coordinates of the marker in the camera and the coordinates in the lidar point cloud respectively; 3) Use the pose algorithm to solve the transformation matrix from the lidar coordinate system to the camera coordinate system; 4) Extract the point cloud road surface number, and partition the road surface point cloud, then fit each point cloud partition, upsample the road surface point cloud, and generate point cloud-pixel coordinate point pairs; 5) Use the generated point cloud-pixel coordinate point pairs as a data set, use artificial intelligence algorithms to perform multiple regression prediction of three-dimensional point cloud coordinates, and convert pixel coordinates to point cloud coordinates; 6) Based on the transformation from the lidar coordinate system to the northeast sky coordinate system, the pixel coordinates are converted to the northeast sky coordinates. 7. A spatial alignment method for vehicle-road coordination according to claim 6, characterized in that the step 3) is specifically: using the EPnP algorithm to solve the transformation matrix from the lidar coordinate system to the camera coordinate system. 8. A spatial alignment method for vehicle-road coordination according to claim 6, characterized in that the artificial intelligence method in step 5) includes decision trees and random forests. 9. A spatial alignment method for vehicle-road coordination according to claim 1, characterized in that, in the step S3 or S4, the northeast sky coordinate system is converted to the earth WGS84 coordinate system, and the conversion expression is:

In the formula, (x, y) is the coordinate of any point in the northeast sky coordinate system, (lat ₁ , lon ₁ ) is the longitude and latitude of (x, y) in the WGS84 coordinate system, (lat ₀ , lon ₀ ) is the northeast sky coordinate The latitude and longitude of the origin, R is the radius of the earth. 10. A spatial alignment system for vehicle-road coordination, characterized in that the system comprises: The intelligent networked roadside subsystem installed on the roadside, the intelligent networked vehicle subsystem installed on the vehicle, and the intelligent networked vehicle subsystem and the intelligent networked roadside subsystem perform vehicle-road coordination through V2X; Wherein, the method described in any one of claims 1 to 9 is used for spatial alignment during the vehicle-road coordination process.

Images