An intelligent networked vehicle positioning method based on fusion of laser radar and GNSS

Abstract

The application relates to a kind of intelligent networked vehicle positioning methods based on laser radar and GNSS fusion, belong to the field of intelligent vehicle positioning, utilize the information of multiple sensors and the cooperative information of surrounding intelligent devices, give full play to the characteristics of intelligent networked vehicle, more easily obtain the latitude and longitude positioning information of detection target;Adopt robust fusion function under the constraint of positioning information, point cloud information, cooperative information, fitting correction function under multi-time information and weight adaptive function, so that coordinate transformation is more accurate and stable, so that the latitude and longitude positioning information of detection target is more accurate.The absolute positioning of the target detected by the intelligent vehicle avoids the problem that the positioning results of the target by each intelligent vehicle are inconsistent due to the different positions of the target relative to different intelligent networked vehicles, avoids the problem that the target positioning results of the intelligent vehicle are useless to other intelligent vehicles, improves the universality of the target positioning results in intelligent transportation, benefits the overall scheduling of the intelligent transportation system and reduces the overall system calculation amount.

Description

Technical Field

[0001] This invention belongs to the field of intelligent vehicle positioning, and relates to an intelligent connected vehicle positioning method based on the fusion of lidar and GNSS. Background Technology

[0002] In recent years, with the development of new energy, intelligent technology, and wireless communication technology, intelligent vehicles have become a modern automotive trend, with more and more intelligent devices and technologies being applied to automotive systems. Intelligent connected vehicles, relying on technologies such as computer vision, deep learning, and wireless network communication, are equipped with various sensors and possess functions such as environmental perception, path planning, and decision control, making them important participants in future intelligent transportation systems. Among these, positioning technology is the most fundamental and crucial technology for intelligent connected vehicles to achieve autonomous driving. The positioning of intelligent connected vehicles includes both its own positioning and its positioning of other traffic participants.

[0003] To obtain its own positioning information, intelligent connected vehicles are usually equipped with GNSS receivers to acquire satellite positioning signals. Furthermore, to improve positioning accuracy, inertial navigation and satellite navigation systems are coupled to form a combined navigation system, achieving a positioning system with high accuracy and robustness.

[0004] Regarding the acquisition of location information for other traffic participants, the intelligentization process of the entire transportation system is slow, and for a considerable period, the system will be a mix of intelligent connected vehicles, traditional vehicles, intelligent traffic signs, traditional signs, pedestrians, and other traffic participants. On the one hand, intelligent connected vehicles can relatively easily acquire the location information of other intelligent devices through wireless communication technology; on the other hand, acquiring the location information of other traffic participants without intelligent devices is difficult. Currently, target detection is commonly used to obtain the relative positions of these participants to the intelligent connected vehicles themselves. However, due to the inconsistent motion states of each traffic participant, their relative positions differ for different intelligent connected vehicles, rendering this relative position information ineffective for other intelligent connected vehicles and lacking universality. Therefore, acquiring the absolute location information such as latitude and longitude of the detected targets (other traffic participants) is of great significance. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a positioning method for intelligent connected vehicles based on the fusion of lidar and GNSS, so as to obtain the latitude and longitude of the target vehicle detected by the intelligent connected vehicle. This method utilizes target detection technology, satellite positioning technology, and wireless communication technology to calculate the latitude and longitude of the detected target. It also provides a coordinate transformation module to avoid large errors during multiple coordinate transformations. This method provides more detailed and universal target detection information for intelligent connected vehicles, enabling them to be more accurate and rational in path planning and real-time obstacle avoidance, and making the entire intelligent transportation system more unified.

[0006] In an intelligent transportation system, there are intelligent connected vehicles (CIV1 and CIV2) and a traditional vehicle (V1). At a certain moment, CIV2 cannot detect target V1, while CIV1 can detect target V1 and obtain its relative and absolute position information, and broadcasts the absolute position information to CIV2 via wireless communication. To achieve the above objective, the present invention provides the following technical solution:

[0007] A positioning method for intelligent connected vehicles based on the fusion of lidar and GNSS includes the following steps:

[0008] S1: Using GNSS positioning data based on the Mercator projection model, latitude and longitude are converted to geodetic coordinates to construct a geodetic coordinate system; a local coordinate system of the lidar is constructed, and the initial transformation parameters from the local coordinate system to the geodetic coordinate system are solved using the spatial transfer calibration method; the PointRCNN 3D detection algorithm is used to obtain the position (x, y) of V1 in the local coordinate system from the point cloud data. L ,y L );

[0009] S2: Robustly fuse the current lidar point cloud data, GNSS positioning data, and surrounding intelligent connected vehicle positioning data received via wireless communication to calculate the transformation parameters from the current local coordinate system to the geodetic coordinate system at time k, i.e., the rotation matrix R′. k Translation matrix T′ k ;

[0010] S3: Fit the navigation information and transformation parameters at M time points to obtain the transformation parameters from the local coordinate system to the geodetic coordinate system at time k, i.e., the rotation matrix R″. k Translation matrix T″ k The fitted estimate;

[0011] S4: Transformation parameters R′ obtained based on the robust fusion function at the current time. k 、T′ k and the transformation parameter R″ estimated based on the multi-time fitting correction function k 、T″ kThe optimal transformation parameters R at time k are obtained. k T k Based on the determined transformation parameters, the local coordinates (x) will be... L ,y L Transform into geodetic coordinates (x) Ge ,y Ge Finally, the geodetic coordinates are converted into latitude and longitude to obtain the absolute position of the target vehicle, and then broadcast to the surrounding intelligent connected vehicles via wireless communication.

[0012] Furthermore, step S1 specifically includes the following steps:

[0013] S11: Install a lidar on the roof of the intelligent connected vehicle to collect surrounding point cloud data; install a GNSS receiver directly below the lidar to collect positioning data;

[0014] S12: Based on the Mercator projection model, latitude and longitude are converted to geodetic coordinates using WGS84 parameters to construct a geodetic coordinate system.

[0015]

[0016] Where X represents the horizontal rectangular coordinates, Y represents the horizontal vertical coordinates; K is the conversion coefficient; ln(·) represents the natural logarithm; B represents latitude, L represents longitude, in radians; B0 represents the projection reference latitude, L0 represents the projection reference longitude, the latitude and longitude obtained the first time after startup are taken as the reference latitude and longitude; a is the semi-major axis of the Earth ellipsoid, b is the semi-minor axis of the Earth ellipsoid; e represents the first eccentricity, e′ represents the second eccentricity;

[0017] S13: The local coordinate system of the lidar is unified with the geodetic coordinate system through a spatial transfer calibration method; the intelligent connected vehicle CIV1 first performs initialization correction motion to obtain its own coordinates in the geodetic coordinate system. Coordinates of CIV1 in the initial local coordinate system Finally, calculate the rotation matrix R0 and rotation angle θ from the initial local coordinate system to the geodetic coordinate system:

[0018]

[0019] S14: Input the preprocessed LiDAR point cloud data into the Point R-CNN algorithm network to obtain the target vehicle's state information s = [x, y, z, l, w, h] T Where x, y, z are the coordinates of the center point of the target 3D bounding box, and l, w, h are the length, width, and height of the 3D bounding box. (x, y) is taken as the position of target V1 in the local coordinate system. L ,y L ).

[0020] Furthermore, step S2 specifically includes the following steps:

[0021] S21: Using GNSS positioning data and the positioning information of surrounding intelligent vehicles, calculate the displacement distance from time k-1 to time k. The calculation is as follows:

[0022]

[0023] Where m represents the total number of intelligent vehicles that can communicate in the surrounding area, and ||·||2 represents the L2 norm operation; This represents the GNSS positioning information of the i-th vehicle at time k, where i = 0. This indicates the GNSS positioning information of CIV1 itself;

[0024] S22: Calculate the vehicle's own heading deflection angle using GNSS positioning information. The calculation formula is as follows:

[0025]

[0026] in, Indicates the direction of the origin of the geodetic coordinate system. When vectors are in the same time interval, <·> represents the dot product operation of vectors, and |·| represents the modulo operation;

[0027] S23: Utilizing lidar point cloud information, based on the nearest point residual iteration idea, to... and Using the positioning information of CIV1 itself and surrounding intelligent vehicles as initial values, a robust fusion function is proposed, and the coordinate transformation parameter R′ is calculated. k 、T′ k :

[0028]

[0029] Where r represents the system residual, N represents the total number of laser points in the point cloud, and p k,i This represents the i-th point in the point cloud of the k-th frame. These represent the rotations of the point cloud around the z-axis, y-axis, and x-axis, respectively, by θ. k The rotation matrix; θ k Indicates the rotation angle of the point cloud; t k v represents the point cloud translation vector; k It is the maximum permissible error of displacement, w k It is the maximum permissible error of rotation, which ensures t during the nearest point residual iteration process. k and v k Minimize the error value.

[0030] Furthermore, step S3 specifically includes the following steps:

[0031] S31: Based on the fitting correction idea, using GNSS positioning data at M time points and the transformation parameter R′ solved in S2. k 、T′ k A fitting correction function J(ζ,ξ) is proposed:

[0032]

[0033] in, ζ and ξ are the rotation matrices R′ at two different time points. k Translation matrix T′ k The fitting coefficient; where k > M, when this condition is not met, R′ is used directly. k 、T′ k As the final transformation parameters;

[0034] S32: The fitting correction function J(ζ,ξ) must satisfy the following requirements:

[0035]

[0036]

[0037] Solving the above two formulas yields the fitting coefficients ζ and ξ, which provide the highest degree of fit between the GNSS position and the lidar-derived position. This leads to the transformation parameter R″. k =ζR′ k-1 and T″ k =ξT′ k-1 .

[0038] Furthermore, step S4 specifically includes the following steps:

[0039] S41: Utilizing the weighted adaptive function, combined with R′ k 、T′ k 、R″ k 、T″ k Calculate the optimal transformation parameters R k T k The formula is as follows:

[0040]

[0041]

[0042] Where α1 and α2 are the adaptive weights of the rotation matrix based on the current time-time transformation and multi-time-time fitting, respectively, and β1 and β2 are the adaptive weights of the translation matrix based on the current time-time transformation and multi-time-time fitting, respectively; R ref The reference rotation matrix is ​​determined by the heading angle of the intelligent vehicle, T.ref This represents the reference translation vector, determined by GNSS information;

[0043] S42: Obtain the latitude and longitude coordinates of the detected target using coordinate transformation formulas and geodetic coordinate system formulas; determine the coordinates of the target in the local coordinate system of the lidar at time k. Transform to geodetic coordinates The calculation formula is as follows:

[0044]

[0045] Finally, what will be obtained The absolute position latitude and longitude of the target vehicle are obtained by substituting them into the geodetic coordinate system formula, and then broadcast to the surrounding intelligent connected vehicles via wireless communication.

[0046] The beneficial effects of this invention are as follows:

[0047] 1. This invention utilizes multi-sensor information and collaborative information from surrounding intelligent devices to fully leverage the characteristics of intelligent connected vehicles, making it easier to obtain the latitude and longitude positioning information of the target.

[0048] 2. Compared with existing target detection and localization methods, this invention adopts a robust fusion function constrained by localization information, point cloud information, and collaborative information, a fitting correction function under multi-time information, and a weighted adaptive function, thereby making the coordinate transformation more accurate and stable, and thus making the latitude and longitude positioning information of the detected target more precise.

[0049] 3. The target detected by the intelligent vehicle is absolutely located to obtain latitude and longitude positioning information. This avoids the problem of inconsistent target positioning results among intelligent vehicles due to different relative positions of the target to different intelligent connected vehicles. It also avoids the problem that the target positioning results of one intelligent vehicle are useless to other intelligent vehicles. This improves the universality of target positioning results in intelligent transportation, which is conducive to the overall scheduling of intelligent transportation system and reduces the overall system computation.

[0050] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0051] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0052] Figure 1 This is the main flowchart of the target localization method described in the embodiments of the present invention;

[0053] Figure 2 This is a schematic diagram of a scenario described in an embodiment of the invention. Detailed Implementation

[0054] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0055] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0056] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0057] Please see Figures 1-2 First, assume that in an intelligent transportation system there are two connected and intelligent vehicles (CIV1 and CIV2), and a traditional vehicle V1 (which cannot communicate with other vehicles). At a certain moment, CIV2 cannot detect the target V1 due to distance, occlusion, etc., while CIV1 can detect V1 and obtain its relative position information; however, this relative position information is invalid for CIV2. To improve the universality of target positioning information, CIV1 needs to obtain the absolute position information of V1, such as latitude and longitude, and then broadcast it to CIV2 via wireless communication. This includes the following steps:

[0058] S1: Using GNSS positioning data based on the Mercator projection model, latitude and longitude are converted to geodetic coordinates using WGS84 parameters to construct a geodetic coordinate system; then, a local coordinate system of the lidar is constructed, and a spatial transfer calibration method is proposed to solve for the initial transformation parameters of rotating the local coordinate system to the geodetic coordinate system; the PointRCNN 3D detection algorithm is used to obtain the position (x, y) of V1 in the local coordinate system from the point cloud data. L ,y L ).

[0059] S11: To obtain better point cloud data, the lidar is installed in the center of the roof of the intelligent connected vehicle, 1.99 meters above the ground, and collects surrounding point cloud data at a rate of 10 frames per second. To ensure that the origin of the geodetic coordinate system is consistent with the origin of the local two-dimensional coordinate system, the GNSS receiver is installed directly below the lidar, 1.6 meters above the ground, and collects positioning data at a rate of 10 frames per second.

[0060] S12: Based on the Mercator projection model, latitude and longitude are converted to geodetic coordinates using WGS84 parameters to construct a geodetic coordinate system. The calculation formula is as follows:

[0061]

[0062] Where X represents the horizontal rectangular coordinate, Y represents the horizontal vertical coordinate, and the unit is meters; K is the conversion factor; ln(·) represents the natural logarithm; B represents latitude, L represents longitude, and the unit is radians; B0 represents the projection reference latitude, L0 represents the projection reference longitude, and the unit is radians. The latitude and longitude obtained on the first boot are taken as the reference latitude and longitude; a is the semi-major axis of the Earth ellipsoid, with a value of 6378137.0000, b is the semi-minor axis of the Earth ellipsoid, with a value of 6356752.3142, and the unit is meters; e represents the first eccentricity, with a value of 0.08181979, and e′ represents the second eccentricity, with a value of 0.08209504.

[0063] S13: To unify the local coordinate system of the lidar with the geodetic coordinate system, a spatial transfer calibration method is proposed. The intelligent vehicle first performs initialization correction motion to obtain its own coordinates in the geodetic coordinate system. Coordinates of CIV1 in the initial local coordinate system Finally, the rotation matrix R0 and rotation angle θ from the initial local coordinate system to the geodetic coordinate system are calculated as follows:

[0064]

[0065] S14: Input the preprocessed LiDAR point cloud data into the Point R-CNN algorithm network to obtain the target vehicle's state information s = [x, y, z, l, w, h] TWhere x, y, z are the coordinates of the center point of the target 3D bounding box, and l, w, h are the length, width, and height of the 3D bounding box. (x, y) is taken as the position of target V1 in the local coordinate system. L ,y L ).

[0066] S2: A robust fusion function is proposed to robustly fuse the current lidar point cloud data, GNSS positioning data, and surrounding intelligent connected vehicle positioning data received via wireless communication, thereby calculating the transformation parameters from the current local coordinate system to the geodetic coordinate system at time k, i.e., the rotation matrix R′. k Translation matrix T′ k .

[0067] S21: Using GNSS positioning data and the positioning information of surrounding intelligent vehicles, calculate the displacement distance from time k-1 to time k. The calculation is as follows:

[0068]

[0069] Where m represents the total number of intelligent vehicles that can communicate in the vicinity, and ||·||2 represents the L2 norm operation. The denoted represents the GNSS positioning information of the i-th vehicle at time k, and specifically, when i = 0, it represents its own GNSS positioning information.

[0070] S22: Calculate the vehicle's own heading deflection angle using GNSS positioning information. The calculation formula is as follows:

[0071]

[0072] in, Indicates the direction of the origin of the geodetic coordinate system. The vectors are defined by the following expressions: <·> represents the inner product operation of vectors, and |·| represents the modulo operation.

[0073] S23: Utilizing lidar point cloud information, based on the nearest point residual iteration idea, to... and Using the positioning information of itself and surrounding intelligent vehicles as initial values, and taking these as constraints, a robust fusion function is proposed to derive the accurate and robust coordinate transformation parameter R′. k 、T′ k :

[0074]

[0075] Where r represents the system residual, N represents the total number of laser points in the point cloud, and p k,i This represents the i-th point in the point cloud of the k-th frame. These represent the rotations of the point cloud around the z-axis, y-axis, and x-axis, respectively, by θ. k The rotation matrix; θ k Indicates the rotation angle of the point cloud; t k v represents the point cloud translation vector; k It is the maximum permissible error of displacement, w k This is the maximum permissible error of rotation. During the nearest point residual iteration process, t should be guaranteed. k and v k Minimize the error value.

[0076] S3: Given the consistency of the intelligent vehicle's motion model over a short period and considering the trend changes in transformation parameters, a fitting correction function is proposed to fit the navigation information and transformation parameters at M time points, thereby obtaining the transformation parameters from the local coordinate system to the geodetic coordinate system at time k, i.e., the rotation matrix R″. k Translation matrix T″ k The fitted estimate.

[0077] S31: Based on the fitting correction idea, using GNSS positioning data at M time points and the transformation parameter R′ solved in S2. k 、T′ k A fitting correction function J(ζ,ξ) is proposed:

[0078]

[0079] in, ζ and ξ are the rotation matrices R′ at two different time points. k Translation matrix T′ k The fitting coefficient. Where k > M, when this condition is not met, R′ is used directly. k 、T′ k As the final transformation parameters.

[0080] S32: To ensure GNSS positioning Solving for position R′ with lidar k X k +T′ k For the highest fit, the fitting correction function J(ζ,ξ) must satisfy the following requirements.

[0081]

[0082]

[0083] Solving equations (7) and (8) yields the fitting coefficients ζ and ξ, which provide the highest degree of fit between the GNSS position and the lidar-solved position. This leads to the transformation parameter R″. k =ζR′ k-1 and T″k =ξT′ k-1 .

[0084] S4: An adaptive weight fusion method is proposed to fuse the transformation parameters R′ obtained based on the robust fusion function at the current time. k 、T′ k and the transformation parameter R″ estimated based on the multi-time fitting correction function k 、T″ k The optimal transformation parameters at time k are obtained, i.e., the rotation matrix R. k Translation matrix T k Then, based on the determined transformation parameters, the local coordinates (x... L ,y L Transform into geodetic coordinates (x) Ge ,y Ge Finally, the geodetic coordinates are converted into latitude and longitude to obtain the absolute position of the target vehicle, and then broadcast to the surrounding intelligent connected vehicles via wireless communication.

[0085] S41: Utilizing the weighted adaptive function, combined with R′ k 、T′ k 、R″ k 、T″ k Calculate the optimal transformation parameters R k T k The formula is as follows:

[0086]

[0087]

[0088] Where α1 and α2 are the adaptive weights of the rotation matrix based on the current time-time transformation and multi-time-time fitting, respectively, and β1 and β2 are the adaptive weights of the translation matrix based on the current time-time transformation and multi-time-time fitting, respectively; R ref The reference rotation matrix is ​​determined by the heading angle of the intelligent vehicle, T. ref This represents the reference translation vector, determined by GNSS information.

[0089] S42: Obtain the latitude and longitude coordinates of the detected target using the coordinate transformation formula and formula (1). Then, determine the coordinates of the target at time k in the local coordinate system of the lidar. Transform to geodetic coordinates The calculation formula is as follows:

[0090]

[0091] Finally, what will be obtained Substitute the coordinates of the target vehicle into formula (1) to obtain its absolute position latitude and longitude, and broadcast it to surrounding intelligent connected vehicles via wireless communication.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent connected vehicle positioning based on fusion of laser radar and GNSS, characterized in that: Includes the following steps: S1: based on the GNSS positioning data, convert the longitude and latitude into geodetic coordinates based on the Mercator projection model, construct a geodetic coordinate system; construct a local coordinate system of the laser radar, and solve the initial transformation parameters of the rotation of the local coordinate system to the geodetic coordinate system through a space transfer calibration method; use the PointRCNN 3D detection algorithm to obtain the position of V1 in the local coordinate system from the point cloud data ; S2: Robustly fuse the current lidar point cloud data, GNSS positioning data, and surrounding intelligent connected vehicle positioning data received via wireless communication to calculate the... k The transformation parameters from the local coordinate system to the geodetic coordinate system at any given time, i.e., the rotation matrix. Translation matrix ; S3: Yes The navigation information and transformation parameters at each time point are fitted together to obtain... k The transformation parameters from the local coordinate system to the geodetic coordinate system at any given time, i.e., the rotation matrix. Translation matrix The fitted estimate; S4: Transformation parameters obtained based on the robust fusion function at the current time. , Transform parameters estimated based on multi-time fitting correction function , ,get k Optimal transformation parameters at time , Based on the determined transformation parameters, the local coordinates will be... Transform to geodetic coordinates Finally, the geodetic coordinates are converted into latitude and longitude to obtain the absolute position of the target vehicle, and then broadcast to the surrounding intelligent connected vehicles via wireless communication. Step S2 specifically includes the following steps: S21: Utilizing GNSS positioning data, combined with the positioning information of surrounding intelligent vehicles, calculate... Time's up Displacement distance at time The calculation is as follows: in, This indicates the total number of intelligent vehicles in the vicinity that can communicate with each other. Represents the L2 norm operation; express Time of the first GNSS positioning information of the vehicle, when hour This indicates the GNSS positioning information of CIV1 itself; S22: Calculate the vehicle's own heading deflection angle using GNSS positioning information. The calculation formula is as follows: in, Indicates the direction of the origin of the geodetic coordinate system. Vector of time, This represents the dot product operation of vectors. This represents the modulo operation; S23: Utilizing lidar point cloud information, based on the nearest point residual iteration idea, to... and Using the positioning information of CIV1 itself and surrounding intelligent vehicles as initial values, a robust fusion function is proposed to determine the coordinate transformation parameters, constrained by the positioning information of CIV1 and surrounding intelligent vehicles. , : in, r Represents the system residual. This represents the total number of laser points in the point cloud. Indicates the first The first frame of point cloud One point, , , These represent the rotations of the point cloud around the z-axis, y-axis, and x-axis, respectively. The rotation matrix; Indicates the rotation angle of the point cloud; Represents the point cloud translation vector; It is the maximum permissible error of displacement. This is the maximum permissible error of rotation, guaranteed during the nearest point residual iteration process. and Minimize the error value; Step S3 specifically includes the following steps: S31: Based on the idea of ​​fitting correction, utilizing GNSS positioning data at time 1 and the transformation parameters solved in S2 , A fitting correction function is proposed. : in, , and These are the rotation matrices at two different time points. Translation matrix The fitting coefficients; where If this condition is not met, use directly. , As the final transformation parameters; S32: Fitting correction function The following requirements must be met: Solving the above two formulas yields the fitting coefficients that achieve the highest degree of agreement between the GNSS position and the lidar-derived position. and Thus, the transformation parameters are obtained. and .

2. The intelligent connected vehicle positioning method based on the fusion of lidar and GNSS according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11: Install a lidar on the roof of the intelligent connected vehicle to collect surrounding point cloud data; install a GNSS receiver directly below the lidar to collect positioning data; S12: Based on the Mercator projection model, latitude and longitude are converted to geodetic coordinates using WGS84 parameters to construct a geodetic coordinate system. in Represented as horizontal rectangular coordinates, Represented as horizontal and vertical coordinates; These are the conversion factors; Represents the natural logarithm; Indicates latitude, Longitude is expressed in radians. Indicates the reference latitude for projection. This indicates the reference longitude for projection, which is the latitude and longitude obtained the first time the device is powered on. The semi-major axis of the Earth's ellipsoid. Indicates the minor semi-axis of the Earth's ellipsoid; Indicates the first eccentricity. Indicates the second eccentricity; S13: The local coordinate system of the lidar is unified with the geodetic coordinate system through a spatial transfer calibration method; the intelligent connected vehicle CIV1 first performs initialization correction motion to obtain its own coordinates in the geodetic coordinate system. Coordinates of CIV1 in the initial local coordinate system Finally, calculate the rotation matrix from the initial local coordinate system to the geodetic coordinate system. and rotation angle : S14: Input the preprocessed LiDAR point cloud data into the Point R-CNN algorithm network to obtain the target vehicle's state information. ,in The coordinates of the center point of the target 3D bounding box. For the length, width, and height information of the 3D bounding box, take... Let V1 be the position of the target V1 in the local coordinate system. .

3. The intelligent connected vehicle positioning method based on the fusion of lidar and GNSS according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41: Utilizing the weighted adaptive function, combined with , , , Calculate the optimal transformation parameters , The formula is as follows: in, and These are the adaptive weights of the rotation matrix based on the current-time transformation and the multi-time-time fitting, respectively. and These are the adaptive weights of the translation matrix based on the current time-time transformation and the multi-time-time fitting, respectively; The reference rotation matrix is ​​determined by the heading angle of the intelligent vehicle. This represents the reference translation vector, determined by GNSS information; S42: Obtain the latitude and longitude coordinates of the target using coordinate transformation formulas and geodetic coordinate system formulas; k The coordinates of the target in the local coordinate system of the lidar at any given time Transformed into geodetic coordinates The calculation formula is as follows: Finally, what will be obtained The absolute position latitude and longitude of the target vehicle are obtained by substituting them into the geodetic coordinate system formula, and then broadcast to the surrounding intelligent connected vehicles via wireless communication.

Images