A v2x vehicle classification method, apparatus and computer program product

Abstract

The application discloses a V2X vehicle classification method, device and computer program product, calculates the relative classification of the remote vehicle to the vehicle and the coordinate axis through the projection, considers the influence of the vehicle size, and accurately calculates the potential collision risk under the relative heading angle of different remote vehicles and the vehicle; meanwhile, the remote vehicle under any curve or even different road sections can be approximately processed as the scene under the same straight road with the vehicle to perform the projection classification calculation; finally, the vehicle position classification and the heading classification are independently calculated, the classification categories and the accuracy are effectively increased, the classification result is more comprehensive and easy to use, and the algorithm coupling degree is reduced. The application is suitable for any curvature curve and combined lane, and the classification change caused by the vehicle size and the relative heading angle is calculated, the vehicle classification accuracy can be greatly improved, the accuracy of the V2V and related ADAS scenes is further improved, and the scene missed trigger or false trigger is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving vehicle technology, specifically to a V2X vehicle classification method, device, and computer program product. Background Technology

[0002] The relative position classification of the remote vehicle to the local vehicle (hereinafter referred to as vehicle classification) is the premise and foundation of all vehicle-to-vehicle (V2V) warning and related advanced driver assistance system (ADAS) control algorithms. Vehicle classification directly determines the scenarios triggered by V2V or other ADAS-related applications. Incorrect vehicle classification will lead to missed or false triggering of scenarios. Therefore, only accurate vehicle classification can ensure the reliability and accuracy of V2V warning and control.

[0003] Cameras and radar sensors struggle to identify distant vehicles on curves or in other obstructed areas, a problem that can be effectively solved by Vehicle-to-Everything (V2X) systems. However, traditional V2X vehicle classification algorithms rely solely on the relative positions of the two vehicles' centers of gravity and the included heading angle, or on the lanes of the distant and the vehicle itself. The former struggles to correctly identify curved scenarios with varying curvature (especially S-shaped curves), while the latter is prone to misjudgment when either the vehicle or the distant vehicle is not traveling along the lane centerline or when the vehicle's heading angle deviates from the lane centerline.

[0004] Furthermore, traditional vehicle classification algorithms do not consider the impact of vehicle size and the deviation of the heading angle of distant vehicles, relying solely on the centroid coordinates of distant vehicles relative to the vehicle's position or lane. For example, if distant vehicles RV1 and RV2 are to the left and right front of the vehicle's HV (both vehicles are located in the adjacent lane ahead of the vehicle, and their heading angle deviations are within the same direction), traditional classification algorithms will assume there is no forward collision risk and will not trigger a forward collision scenario, even though there is actually a possibility of a forward collision between the vehicle and the distant vehicles. Simply increasing the classification threshold in traditional classification algorithms can easily include vehicles without relevant collision risks in the classification, thus triggering unnecessary scenarios and wasting computational resources. Summary of the Invention

[0005] The technical problem to be solved by the embodiments of the present invention is to provide a V2X vehicle classification method, device and computer program product to improve the accuracy and practicality of vehicle classification.

[0006] To address the aforementioned technical problems, this invention provides a V2X vehicle classification method, comprising the following steps:

[0007] Establish the vehicle's coordinate system based on the vehicle information and the received lane information;

[0008] Based on the remote vehicle information and the received lane information, establish a remote vehicle coordinate system; calculate the position information of the remote vehicle in the remote vehicle coordinate system, and transform the position information to the local vehicle coordinate system;

[0009] A forward projection coordinate system is established based on the heading angle of the vehicle itself; the position information of the distant vehicle in the coordinate system of the vehicle itself is projected onto the forward projection coordinate system to obtain the projection of the distant vehicle in the forward projection coordinate system.

[0010] Based on the projection of the distant vehicle into the forward projection coordinate system, the relative positional relationship between the distant vehicle and the vehicle itself is determined, and the relative positional classification result is obtained.

[0011] Based on the heading angles of the vehicle and the distant vehicle, as well as the angle between them and the direction of travel in the lane, the reverse driving coefficient is calculated, and the relative heading relationship between the two vehicles is determined based on the reverse driving coefficient to obtain the relative heading classification result.

[0012] The relative position classification result is combined with the relative heading classification result to obtain the final vehicle classification result.

[0013] Preferably, establishing the vehicle coordinate system based on the vehicle information and the acquired lane information specifically includes:

[0014] Based on the vehicle information and the lane information received from the roadside unit (RSU), determine the road segment to which the vehicle's lane belongs, and obtain the centerline coordinate matrix of the road segment where the vehicle's lane is located.

[0015] Calculate the coordinates of point O, the centerline point of the road segment closest to the vehicle's center of gravity.

[0016] With point O as the origin, the tangent direction of the centerline at point O is the positive X-axis direction, and the normal direction of the tangent direction to the right is the positive Y-axis direction, thus establishing the vehicle coordinate system.

[0017] Preferably, establishing the remote vehicle coordinate system based on the remote vehicle information and the received lane information specifically includes:

[0018] Based on the remote vehicle information and the lane information received from the roadside unit (RSU), determine the road segment to which the remote vehicle's lane belongs, and obtain the centerline coordinate matrix of the road segment where the remote vehicle's lane is located.

[0019] Calculate the coordinates of the road segment centerline point O' that is closest to the centroid of the distant vehicle;

[0020] With point O' as the origin, the tangent direction of the centerline at point O' is the positive X-axis, and the normal direction of the tangent direction to the right is the positive Y-axis, the remote vehicle coordinate system is established; wherein, the tangent direction of the centerline at point O' is along the direction of curve O→O'.

[0021] Preferably, the position information of the remote vehicle in the remote vehicle coordinate system specifically refers to the coordinates of its centroid and vertex. Transforming the centroid and vertex coordinates of the remote vehicle to the local vehicle coordinate system specifically includes:

[0022] The centroid and vertex coordinates of the distant vehicle are translated along the positive X-axis of the vehicle coordinate system by a distance S, where S is the actual curve distance between point O and point O' on the center line of the road segment.

[0023] If the vector direction of the line connecting the nearest point O1 to point O in the curve O→O' is opposite to the direction of the tangent to the center line at point O, then the translated coordinates will be rotated 180° counterclockwise around point O.

[0024] If point O and point O' are not on the same road segment, the translation distance is determined by the set of points on the center line of the road segment covered by the connection between the two points.

[0025] Preferably, the forward projection coordinate system is established based on the vehicle's heading angle, specifically including: taking the midpoint of the vehicle's front boundary as the origin, the vehicle's heading angle as the positive X-axis direction, and the normal direction of the vehicle's heading angle to the right as the positive Y-axis direction, to establish the forward projection coordinate system.

[0026] Preferably, the step of projecting the position information of the remote vehicle in the local vehicle coordinate system onto the forward projection coordinate system to obtain the projection of the remote vehicle in the forward projection coordinate system specifically includes:

[0027] Represent the vehicle and the distant vehicle as rectangles, and calculate the projection point of the rectangle vertex of the distant vehicle onto the rectangle of the vehicle in the forward projection coordinate system.

[0028] If the rectangular vertex of the distant vehicle and the two rectangular boundaries of the vehicle itself both have projections, then only the rectangular boundary of the vehicle itself closest to the rectangular vertex of the distant vehicle is taken as the projection.

[0029] If there is no corresponding projection, calculate the projection points of the rectangular vertices of the distant vehicle on the X and Y axes;

[0030] If the line connecting the vertex of the remote vehicle's rectangle to its projection point along the X-axis or Y-axis first intersects the rectangular boundary of the remote vehicle itself, then the projection point is considered invalid.

[0031] Preferably, the method of determining the relative positional relationship between the distant vehicle and the main vehicle based on the projection of the distant vehicle in the forward projection coordinate system to obtain the relative position classification result includes at least one of the following:

[0032] Based on the projection points and projection line segments of the vehicle in the forward projection coordinate system, relative positional relationships are classified according to preset rules to obtain the relative positional classification results; specifically, the relative positional relationship classification according to preset rules is determined based on the existence of projection points on specific line segments and the comparison of the lengths of the projection line segments in the X and Y axis directions with preset thresholds; or

[0033] Calculate the overlap rate between the area of ​​the distant vehicle's projection within the safe area of ​​the vehicle and the actual projected area of ​​the distant vehicle; when the overlap rate exceeds a preset threshold, a collision risk is determined, and the relative position classification result is obtained based on the different overlap rates.

[0034] Preferably, when the distant vehicle and the vehicle are not on the same road segment, the step of determining the relative positional relationship between the distant vehicle and the vehicle to obtain the relative positional classification result specifically includes:

[0035] Place the intersection of the distant vehicle and the road segment where this vehicle is located in the coordinate system of this vehicle;

[0036] For curved road sections, an approximation of straight road is used, while preserving the included angle at the intersection of road sections, and using the curved distance to replace the straight distance;

[0037] Based on the centroids of the distant vehicle and the local vehicle, and the projections of the two vehicles onto the local vehicle's road segment, the relative positions of the distant vehicle and the local vehicle are determined and classified.

[0038] Preferably, the step of calculating the reverse driving coefficient based on the heading angles of the vehicle and the distant vehicle and the angle with the direction of travel of the lane, and determining the relative heading relationship between the two vehicles based on the reverse driving coefficient to obtain the relative heading classification result, specifically includes:

[0039] The reverse driving coefficient of this vehicle is calculated based on the angle between the heading angle of this vehicle and the driving direction of the lane in which this vehicle is located;

[0040] The reverse driving coefficient of the distant vehicle is calculated based on the angle between the heading angle of the distant vehicle and the prescribed driving direction of the lane where the vehicle is located.

[0041] Determine whether the reverse driving coefficient of this vehicle is equal to that of the distant vehicle. If they are equal, the two vehicles are traveling in the same direction; otherwise, they are traveling in opposite directions.

[0042] Preferably, the step of calculating the reverse driving coefficient based on the heading angles of the vehicle and the distant vehicle and the angle with the direction of travel of the lane, and determining the relative heading relationship between the two vehicles based on the reverse driving coefficient to obtain the relative heading classification result, specifically includes:

[0043] Calculate the reverse driving coefficient of this vehicle based on the angle between the heading angle of this vehicle and the direction of travel of the lane it is in;

[0044] Based on the heading angle of the distant vehicle and its relative relationship with the direction of travel in the lane where the main vehicle is located, and combined with the road segment information where the distant vehicle is located, the reverse driving coefficient of the distant vehicle is calculated.

[0045] Multiply the reverse driving coefficient of this vehicle by the reverse driving coefficient of the distant vehicle to obtain the reverse driving coefficient of the two vehicles;

[0046] The relative heading relationship between the two vehicles is determined based on the value of the reverse driving coefficient. If the value is 0, they are judged to be traveling in the same direction; if the value is 1, they are judged to be traveling in opposite directions.

[0047] Preferably, the step of combining the relative position classification result with the relative heading classification result to obtain the final vehicle classification result specifically involves performing a bitwise OR operation on the relative position classification result and the relative heading classification result to obtain the final vehicle classification result.

[0048] The present invention also provides a V2X vehicle classification device, comprising:

[0049] One or more processors;

[0050] Memory;

[0051] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to perform the V2X vehicle classification method.

[0052] The present invention also provides a computer program product, including computer instructions that instruct a computer device to perform an operation corresponding to the method.

[0053] Implementing this invention offers the following advantages: First, it not only achieves accurate classification of distant vehicles but also fully considers the impact of vehicle size and relative heading angle on potential collision risks. This method surpasses the traditional simple judgment relying solely on angle thresholds, significantly improving the accuracy and practicality of classification. Second, this invention effectively approximates distant vehicles on arbitrary curves and different road sections as if they were on the same straight road as the vehicle itself, simplifying the calculation process and improving classification efficiency. This greatly enhances the adaptability of this invention in complex road environments. Furthermore, this invention calculates vehicle position classification and heading classification independently, increasing the number of classification categories and accuracy while reducing the coupling between algorithms. The classification results are more comprehensive and easier to use, providing more accurate and reliable data support for subsequent V2V communication and related ADAS scenarios. This helps reduce missed or false triggers, thereby enhancing road safety and the driving experience. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart illustrating a V2X vehicle classification method according to an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the vehicle coordinate system established in an embodiment of the present invention.

[0057] Figure 3 This is a schematic diagram illustrating the transformation of the centroid and vertex coordinates of the distant vehicle to the coordinate system of the local vehicle in an embodiment of the present invention.

[0058] Figure 4 This is a schematic diagram of the forward projection coordinate system in an embodiment of the present invention.

[0059] Figure 5 This is a schematic diagram illustrating vehicle classification on different road sections in an embodiment of the present invention. Detailed Implementation

[0060] The following description of the embodiments is taken with reference to the accompanying drawings, which illustrate specific embodiments in which the invention can be implemented.

[0061] Please refer to Figure 1 As shown, Embodiment 1 of the present invention provides a V2X vehicle classification method, including the following steps:

[0062] Establish the vehicle's coordinate system based on the vehicle information and the acquired lane information;

[0063] Based on the information of the distant vehicle and the acquired lane information, a coordinate system for the distant vehicle is established; the position information of the distant vehicle in the coordinate system is calculated, and the position information is transformed to the coordinate system of the local vehicle.

[0064] A forward projection coordinate system is established based on the heading angle of the vehicle itself; the position information of the distant vehicle in the coordinate system of the vehicle itself is projected onto the forward projection coordinate system to obtain the projection of the distant vehicle in the forward projection coordinate system.

[0065] Based on the projection of the distant vehicle into the forward projection coordinate system, the relative positional relationship between the distant vehicle and the vehicle itself is determined, and the relative positional classification result is obtained.

[0066] Based on the heading angles of the vehicle and the distant vehicle, as well as the angle between them and the direction of travel in the lane, the reverse driving coefficient is calculated, and the relative heading relationship between the two vehicles is determined based on the reverse driving coefficient to obtain the relative heading classification result.

[0067] The relative position classification result is combined with the relative heading classification result to obtain the final vehicle classification result.

[0068] As demonstrated by the above steps, this invention accurately calculates the projection of the distant vehicle onto the coordinate axes of the vehicle's own coordinate system, thereby precisely determining the relative position and heading relationship between the distant vehicle and the vehicle. This results in a more comprehensive vehicle classification, improving not only the accuracy of vehicle classification but also enhancing the reliability of V2V and related ADAS scenarios, effectively reducing false triggers and missed triggers. Furthermore, the method of this invention is applicable to curves of arbitrary curvature and combined lanes, exhibiting broad applicability.

[0069] Specifically, in this embodiment of the invention, relative position classification and relative heading classification are calculated independently, effectively increasing the number of classification categories and accuracy, making the classification results more comprehensive and easy to use, and reducing algorithm coupling. These will be described separately below.

[0070] (I) Classification by Relative Position

[0071] First, we need to establish the vehicle's coordinate system. The specific process is as follows:

[0072] 1. Based on the vehicle information and the lane information obtained from the Roadside Unit (RSU), determine the road segment to which the vehicle's lane belongs (i.e., the link in the V2X message, which contains multiple lanes), and obtain the centerline coordinate matrix of the road segment where the vehicle's lane is located (this information is contained in the MAP message set in the V2X message).

[0073] Calculate the coordinates of point O, the centerline point closest to the vehicle's centroid. Calculate the tangential vector of the road segment's centerline at point O (use the centerline point lattice to fit the local curve to calculate the tangent at this point, or directly use the two points before and after to calculate the slope). Note that the direction of the tangential vector at point O should have an angle ≤ 90° with the vehicle's heading angle (considering the possibility of the vehicle driving in the wrong direction, the lane-defined driving direction cannot be used).

[0074] Through the above steps, the lane where the vehicle is located is determined based on the vehicle information and the lane information in the RSU, and the center line coordinate point O of the road segment closest to the vehicle's center of gravity is found as the origin of the vehicle's coordinate system.

[0075] 2. Using point O from step 1 above as the origin of the vehicle's coordinate system, the tangent to the centerline at point O (in the direction described in step 1) is the positive X-axis direction, and the normal to the right is the positive Y-axis direction, establishing the following... Figure 2The coordinate system of the vehicle is shown. Further, based on the position of the vehicle's center of mass relative to point O, the coordinates of the vehicle's center of mass in this vehicle coordinate system are calculated. At the same time, the vehicle is approximated as a rectangle. Combining the angle between the vehicle's heading angle and the tangent at point O, as well as the length and width of the vehicle, the coordinates of the four rectangular vertices of the vehicle in this vehicle coordinate system are obtained.

[0076] After establishing the coordinate system of this vehicle, or simultaneously, establish the coordinate system of the remote vehicle in a similar manner. The specific process is as follows:

[0077] 3. Similar to steps 1 and 2 above, calculate the road segment where the distant vehicle is located, and select the point O' that is closest to the centroid of the distant vehicle on the centerline of the road segment. Calculate the tangent line of the centerline of the road segment at point O'. Note that the direction of the tangent line is along the curve O→O' (because the calculation is for the position classification of the distant vehicle relative to the vehicle itself, the orientation of the distant vehicle is not considered here). Then, establish the coordinate system of the distant vehicle with point O' as the origin, the direction of the tangent line at point O' as the positive direction of the X-axis, and the normal direction of the tangent line to the right as the positive direction of the Y-axis.

[0078] After establishing the remote vehicle coordinate system, the position information of the remote vehicle in the remote vehicle coordinate system is calculated, and the position information is transformed into the local vehicle coordinate system. The specific process is as follows:

[0079] 4. Calculate the position information of the remote vehicle in the remote vehicle coordinate system according to the method in step 2. In this embodiment, the position information is specifically the coordinates of the centroid and the vertex.

[0080] 5. Transform the centroid and vertex coordinates of the distant vehicle calculated in step 4 into the coordinate system of the local vehicle in step 2.

[0081] like Figure 3 As shown, the coordinates are converted to the coordinates of the far-end vertex and centroid by a translation distance along the positive X-axis. If the vector direction of the line connecting the nearest point O1 to point O in curve OO' is opposite to the tangent direction in step 1, then the translated coordinates need to be rotated 180° counterclockwise around the origin; note that distance S is the distance of curve OO', not the straight-line distance between points O and O' (i.e., the actual curve distance between points O and O' on the center line of the road segment), (x i ,y i This represents the set of coordinates of the center points of the road segment between point O and point O' (which can be obtained from the MAP message set in the V2X message), arranged sequentially from point O to point O', where (x0, y0) are the coordinates of point O, and (x... n+1 ,y n+1 Let O' be the coordinates of point O. If point O and point O' are not on the same road segment, then the centerline points of the road segment covered by connecting the two points are taken. It should be noted that if the road segments where points O and O' are located cannot be connected, then the distant vehicle is classified as NONE, indicating that the road segment where this vehicle is located cannot be connected to the road segment where the distant vehicle is located, and there is no risk of collision.

[0082] Next, the position information of the remote vehicle in the local coordinate system needs to be projected onto the forward projection coordinate system to obtain the projection points and projection line segments of the remote vehicle. The specific process is as follows:

[0083] 6. Figure 3 Transformation of the remote vehicle coordinate system into, for example Figure 4 The forward projection coordinate system is shown below. Specifically, the forward projection coordinate system is established with the heading angle of the vehicle (represented by rectangle ABCD, where A, B, C, and D are the vertices of the rectangle) as the positive X-axis, the midpoint of the vehicle's front boundary (i.e., line segment CD) as the origin, and the normal direction of the vehicle's heading angle to the right as the positive Y-axis. Then, the projection points of the vertices (R, Q, S, T) of the distant vehicle (represented by rectangle RQST, where R, Q, S, and T are the vertices of the rectangle) onto the vehicle's rectangle ABCD are calculated. (If there is a projection with both boundaries of the vehicle's rectangle, only the boundary closest to the distant vehicle's rectangle vertex is used for projection). If there is no corresponding projection, the projection points of the distant vehicle's rectangle vertex on the X and Y axes are calculated. Figure 4 In this context, Ty represents the projection point of the distant vehicle falling on line segment CD. It should be noted that if the line connecting the vertex of the distant vehicle's rectangle to its projection point along the X-axis or Y-axis first intersects the rectangular boundary of the distant vehicle itself, then that projection point is considered invalid and will not participate in the relative position classification judgment described later (i.e., it is not considered). Figure 4 The projection of S on the Y-axis is obscured by the distant vehicle itself.

[0084] Based on the projection point and projection line segment of the distant vehicle, the relative positional relationship between the distant vehicle and the vehicle itself is determined, and the relative position classification result is obtained. The specific method is as follows:

[0085] (1) If there is a projection point of a distant vehicle on line segment CD, it is classified as directly in front;

[0086] (2) If there is a projection point of the distant vehicle on line segment AB, it is classified as directly behind;

[0087] (3) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is ≥ Smax, and the length of the projection line segment in the positive Y-axis direction is ≥ W... l If so, it is classified as the right front;

[0088] (4) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is ≥Smax, and the length of the projection line segment in the positive Y-axis direction is ≥2W. l Then it is classified as far right front;

[0089] (5) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is ≥ Smax, and the length of the projection line segment in the negative Y-axis direction is ≥ W... l If so, it is classified as the left front;

[0090] (6) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is ≥Smax, and the length of the projection line segment in the negative Y-axis direction is ≥2W. l Then it is classified as far left front;

[0091] (7) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is < Smin, and the length of the projection line segment in the negative Y-axis direction is ≥ W l If so, it is classified as the right rear;

[0092] (8) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is <Smin, and the length of the projection line segment in the negative Y-axis direction is ≥2W l Then it is classified as far right rear;

[0093] (9) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is < Smin, and the length of the projection line segment in the negative Y-axis direction is ≥ W l If so, it is classified as the left rear;

[0094] (10) If there are no projection points of the distant vehicle on line segments AB and CD, and the length of the projection line segment of the distant vehicle in the positive X-axis direction is <Smin, and the length of the projection line segment in the negative Y-axis direction is ≥2W l Then it is classified as far left rear;

[0095] Where Smax is the preset maximum threshold for the length of the projected line segment in the X-axis direction, Smin is the preset minimum threshold for the length of the projected line segment in the X-axis direction, and W l This is a preset basic threshold for the length of the projected line segment along the Y-axis.

[0096] It is understood that the above method of determining the relative positional relationship between the distant vehicle and the vehicle itself based on the projection point and projection line segment of the distant vehicle is only an example. The actual classification method is not limited to this. For example, classification can also be performed based on the calculation of projection and overlap rate.

[0097] Specifically, the position information of the remote vehicle in the vehicle's coordinate system is projected onto the forward projection coordinate system. After obtaining the projection of the remote vehicle in the forward projection coordinate system, the projected information can be simplified into a two-dimensional graphic. By analyzing these two-dimensional graphics, the occupancy of the remote vehicle relative to the front, side, and other directions of the vehicle can be intuitively understood.

[0098] To determine whether a potential collision is imminent between the vehicle and a distant vehicle, the overlap rate can be calculated as the ratio of the area of ​​the distant vehicle projected onto the vehicle's safe zone (such as the forward warning zone or lateral buffer zone) to the actual projected area of ​​the distant vehicle. When the overlap rate exceeds a certain preset threshold, it indicates that the distant vehicle has entered the vehicle's safe boundary, potentially posing a collision risk. Based on the different overlap rates, the relative positional relationship between the distant and distant vehicles can be categorized into different types, such as "safe distance," "close approach," "dangerous approach," or "collision possible."

[0099] In addition, the rectangular length and width of this vehicle can be appropriately increased to make them larger than the actual size of the vehicle, in order to improve safety.

[0100] Understandably, it's only necessary to further calculate the relative position classification of the distant vehicle to the current vehicle, including details such as projection points, projection lines, and their overlap rates, when the distant vehicle and the current vehicle are on intersecting road segments. This is to accurately assess whether there is a collision risk between the two vehicles or whether appropriate driving strategies need to be adopted. If the road segment where the distant vehicle is located does not intersect with the current vehicle, there is no need for subsequent detailed relative position classification calculations, as their relative positions in physical space have no direct impact on the current vehicle's safety decisions. In this embodiment, the method for determining whether road segments intersect is as follows: based on the link information in the V2X message (link information represents the connection relationship between road segments, including different lanes and the connection between these lanes), traverse all road segments connected to all lanes in the current vehicle's road segment and determine whether the road segment where the distant vehicle is located exists. If it does, it means that the road segment where the distant vehicle is located directly intersects with the current vehicle's road segment; otherwise, it means that the road segment where the distant vehicle is located does not directly intersect with the current vehicle's road segment.

[0101] Please refer to again Figure 5 As shown, when the vehicle and the distant vehicle are on different road segments, especially when the vehicles are at intersections, this embodiment of the invention can still classify the relative positions of the vehicles. Taking an intersection as an example, at an intersection, the vehicle and the distant vehicle may come from different road segments and be traveling in different directions. Therefore, an effective method is needed to classify their relative positions. Specifically:

[0102] (1) First, the intersection of the road segment is placed in the coordinate system of the vehicle. The purpose is to unify the coordinate system so that the position information of all vehicles is based on the same reference system, which is convenient for subsequent position calculation and classification.

[0103] (2) Then, the curved road sections are approximated as straight sections to reduce computational complexity, allowing for the use of simpler geometric relationships to determine the relative positions of vehicles. Of course, this approximation introduces some error, especially on sections with significant curvature. Therefore, it is necessary to retain the angle information at the intersections of road sections (i.e.,...). Figure 5 The θ shown is used for correction in subsequent calculations.

[0104] (3) Use curved distance instead of straight distance

[0105] Due to the curvature of road sections, the actual distance traveled between two vehicles may be greater than their straight-line distance. Therefore, curved distance is used instead of straight-line distance to more accurately reflect the actual distance between the two vehicles. Curved distance can be calculated based on the shape of the road section and the angle information at the intersection of the road sections.

[0106] (4) Classification based on the relative position determined by the centroid and the projections of the two vehicles

[0107] The center of gravity is an important reference point for the positions of two vehicles. By comparing the positions of their centers of gravity, their relative positions can be preliminarily determined. Additionally, the projected positions of the two vehicles in their respective road segment coordinate systems can be considered. Projected positions more intuitively reflect the relative positions of the two vehicles on the same road segment. Combining information from the center of gravity and projected positions allows for a more accurate classification of the relative positions of the two vehicles.

[0108] When two vehicles traverse too many road sections, the relative distance between them can become very large. In this case, it can be assumed that there is no direct risk of collision between them. Therefore, this embodiment of the invention does not consider the classification of their relative positions in this situation, in order to reduce unnecessary calculations and improve vehicle classification efficiency.

[0109] It should be noted that classifying relative positions based on centroids and the projections of the two vehicles still follows a similar approach, namely, establishing a forward projection coordinate system based on the vehicle's heading angle: projecting the distant vehicle's position information from the vehicle's own coordinate system onto the forward projection coordinate system to obtain the distant vehicle's two-dimensional position information in the forward projection coordinate system, and then determining the relative position relationship based on the distant vehicle's projection in the forward projection coordinate system. When the two vehicles are not on the same road segment, relying solely on the projected position is insufficient to accurately determine the relative position; the offset between the centroids of the two vehicles needs to be considered. For example, if the centroid of the distant vehicle is significantly offset laterally relative to the centroid of the vehicle on the road segment, this may mean that the distant vehicle is approaching or moving away from the side of the vehicle. Combining the projected position and the centroid offset allows for a comprehensive judgment of the relative position between the distant and the vehicle. For example, the relative position can be classified into the following categories: front, rear, left, right, left front, right front, left rear, right rear, etc. These classifications can be further refined according to actual needs and scenarios.

[0110] For example, suppose this car is traveling straight, and a distant car approaches from the right side of this car via a curve. In this car's coordinate system, the projected position of the distant car might be in front of or to the side of this car. However, because the distant car is approaching from a curve, its center of gravity will have a significant lateral shift relative to this car's center of gravity. Combining this information, we can determine that the distant car is located to the right front of this car.

[0111] The above process can further improve and enhance the accuracy and efficiency of relative position classification in the embodiments of the present invention. It provides more effective and reliable support, especially when dealing with complex road sections and intersections.

[0112] (II) Relative heading classification

[0113] The process for classifying the headings of this vehicle and a distant vehicle on the same road segment is as follows:

[0114] The reverse driving coefficient Kh of this vehicle is calculated based on the angle between its heading angle and the direction of travel in its lane (Kh = 1 indicates reverse driving when the angle is greater than 90°). Similarly, the reverse driving coefficient Kr of the distant vehicle is calculated based on the angle between its heading angle and the direction of travel in its lane. It should be noted that the lane where the distant vehicle is located is not used here to define the direction of travel, because different lanes within the same road segment may have different defined directions of travel; therefore, the direction of travel in the lane where this vehicle is located is used.

[0115] Determine if Kh and Kr are equal. If they are equal, the two vehicles are traveling in the same direction; otherwise, they are traveling in opposite directions. At this point, "same direction" and "opposite direction" do not represent the difference in the actual heading angles of the two vehicles. The influence of the heading angles of the two vehicles has already been reflected in the projection method calculation described above.

[0116] More specifically, the specific process of the relative heading classification method is as follows:

[0117] (1) Determine whether the vehicle is traveling in the wrong direction based on the deviation between the heading angle of the vehicle and the driving direction of the lane it is in (obtained from the MAP message). If it is traveling in the wrong direction, the vehicle's reverse driving coefficient Kh = 1; otherwise, Kh = 0.

[0118] (2) Based on the lane information, obtain the road segment information connected to the lane where the vehicle is located. Taking the driving direction of the lane where the vehicle is located as the reference, the road segment connected upstream of the lane is referred to as the upstream road segment, and the road segment connected downstream of the lane is referred to as the downstream road segment.

[0119] (3) If the road segment to which the distant vehicle belongs is in the upstream road segment and there is only one upstream road segment (i.e. there is no fork), then determine whether the distant vehicle is heading towards the road segment where the vehicle is located. If so, the reverse driving coefficient Kr = 0. If the distant vehicle is leaving the road segment where the vehicle is located, then Kr = 1.

[0120] (4) Calculate the reverse driving coefficient K = Kh * Kr. If K = 0, it means that the two vehicles are traveling in the same direction. If K = 1, it means that the two vehicles are traveling in opposite directions. The specific scenarios are as follows:

[0121] (5) If the direction of the distant vehicle is towards the lane where this vehicle is located (Kr=0) and this vehicle is traveling in the direction of the lane (Kh=0), then the distant vehicle is determined to be a vehicle traveling in the same direction.

[0122] (6) If the distant vehicle is traveling in the direction of the lane where this vehicle is located, and this vehicle is traveling in the opposite direction, the distant vehicle is judged to be a vehicle traveling in the opposite direction.

[0123] (7) If the distant vehicle is traveling in a direction that leaves the lane where this vehicle is located, and this vehicle is traveling in the direction of the lane, the distant vehicle is judged to be a vehicle traveling in the opposite direction.

[0124] (8) If the distant vehicle is traveling in a direction that leaves the lane where this vehicle is located, and this vehicle is traveling in the opposite direction, it is determined that the distant vehicle is a vehicle traveling in the same direction.

[0125] (9) If the road segment to which the distant vehicle belongs is a downstream road segment and there is only one downstream road segment (i.e. there is no fork), determine whether the distant vehicle is heading towards the road segment where the vehicle is located. If so, the reverse driving coefficient of the distant vehicle is Kr = 1. If the distant vehicle is leaving the road segment where the vehicle is located, then Kr = 0. Then calculate K = Kh × Kr. If K = 0, it means that the two vehicles are in the same direction. If K = 1, it means that the two vehicles are traveling in opposite directions.

[0126] (10) If the road segment to which the vehicle belongs is in the upstream road segment and there is more than one upstream road segment (i.e., the road segment to which the vehicle belongs is connected to an intersection or ramp), the judgment method is the same as above, but the final result is to add the judgment of "intersection" or "ramp" (the ramp itself has a special identifier in the V2X message).

[0127] It should be noted that in current traditional classification methods, relative heading angle and vehicle position classification are highly coupled. In this embodiment of the invention, relative heading classification (i.e., the determination of whether a distant vehicle is approaching from the same direction, in the opposite direction, or laterally relative to the vehicle itself) is calculated separately and combined with the aforementioned relative position classification using a bitwise OR method to obtain a more comprehensive and accurate vehicle classification. For example, if the position classification result of a distant vehicle relative to the vehicle itself is right front, the corresponding identifier value is 0x04, and the heading classification is oncoming vehicle, the corresponding identifier value is 0x40, then the final vehicle classification result is 0x04|0x40=0x44, indicating an oncoming vehicle to the right front.

[0128] It should also be noted that all V2X scenarios rely on the accuracy of high-precision positioning, and V2X judgment should not be entered when the positioning is unreliable; therefore, this invention assumes that the positioning result is reliable; the interpretation of V2X related signals has been specified in national standards.

[0129] Corresponding to the V2X vehicle classification method described in Embodiment 1 of the present invention, Embodiment 2 of the present invention also provides a V2X vehicle classification device, comprising:

[0130] One or more processors;

[0131] Memory;

[0132] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to perform the V2X vehicle classification method.

[0133] Corresponding to the V2X vehicle classification method described in Embodiment 1 of the present invention, Embodiment 3 of the present invention also provides a computer program product, including computer instructions, which instruct a computer device to perform the operation corresponding to the method.

[0134] Preferably, the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can be any conventional processor. The processor is the control center of the device, connecting various parts of the device through various interfaces and lines.

[0135] The memory mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., while the data storage area can store related data, etc. Furthermore, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard drive, a SmartMedia Card (SMC), a Secure Digital (SD) card, and a Flash Card, or other volatile solid-state storage devices.

[0136] It should be noted that the above-mentioned devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art.

[0137] For the working principle and process of the above embodiments, please refer to the description of Embodiment 1 of the present invention, which will not be repeated here.

[0138] As explained above, compared with existing technologies, the advantages of this invention are as follows: First, it not only achieves accurate classification of distant vehicles but also fully considers the impact of vehicle size and relative heading angle on potential collision risks. This method surpasses the traditional simple judgment relying solely on angle thresholds, significantly improving the accuracy and practicality of classification. Second, this invention can effectively approximate distant vehicles on arbitrary curves and different road sections as if they were on the same straight road as the vehicle itself, thereby simplifying the calculation process and improving classification efficiency, greatly enhancing its adaptability in complex road environments. Furthermore, this invention calculates vehicle position classification and heading classification independently, increasing the number of classification categories and accuracy while reducing the coupling between algorithms. The classification results are more comprehensive and easier to use, providing more accurate and reliable data support for subsequent V2V communication and related ADAS scenarios, helping to reduce missed or false triggers, thereby enhancing road safety and driving experience.

[0139] The above description is merely a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A V2X vehicle classification method, characterized in that, Includes the following steps: Establish the vehicle's coordinate system based on the vehicle information and the received lane information; Establish a coordinate system for the remote vehicle based on the remote vehicle information and the received lane information; Calculate the position information of the remote vehicle in the remote vehicle coordinate system, and transform the position information to the local vehicle coordinate system; Establish a forward projection coordinate system based on the vehicle's heading angle; Project the position information of the remote vehicle in the local vehicle coordinate system onto the forward projection coordinate system to obtain the projection of the remote vehicle in the forward projection coordinate system. Based on the projection of the distant vehicle into the forward projection coordinate system, the relative positional relationship between the distant vehicle and the vehicle itself is determined, and the relative positional classification result is obtained. Based on the heading angles of the vehicle and the distant vehicle, as well as the angle between them and the direction of travel in the lane, the reverse driving coefficient is calculated, and the relative heading relationship between the two vehicles is determined based on the reverse driving coefficient to obtain the relative heading classification result. The relative position classification result and the relative heading classification result are combined by a bitwise OR operation to obtain the final vehicle classification result.

2. The method according to claim 1, characterized in that, The step of establishing a vehicle coordinate system based on the vehicle information and the acquired lane information specifically includes: Based on the vehicle information and the lane information received from the roadside unit (RSU), determine the road segment to which the vehicle's lane belongs, and obtain the centerline coordinate matrix of the road segment where the vehicle's lane is located. Calculate the coordinates of point O, the centerline point of the road segment closest to the vehicle's center of gravity. With point O as the origin, the tangent direction of the centerline at point O is the positive X-axis direction, and the normal direction of the tangent direction to the right is the positive Y-axis direction, thus establishing the vehicle coordinate system.

3. The method according to claim 1, characterized in that, The step of establishing a remote vehicle coordinate system based on the remote vehicle information and the received lane information specifically includes: Based on the remote vehicle information and the lane information received from the roadside unit (RSU), determine the road segment to which the remote vehicle's lane belongs, and obtain the centerline coordinate matrix of the road segment where the remote vehicle's lane is located. Calculate the coordinates of the road segment centerline point O' that is closest to the centroid of the distant vehicle; With point O' as the origin, the tangent direction of the centerline at point O' is the positive X-axis, and the normal direction of the tangent direction to the right is the positive Y-axis, the remote vehicle coordinate system is established; wherein, the tangent direction of the centerline at point O' is along the direction of curve O→O'.

4. The method according to claim 3, characterized in that, The position information of the remote vehicle in the remote vehicle coordinate system specifically refers to the coordinates of its centroid and vertices. Transforming the centroid and vertices of the remote vehicle to the local vehicle coordinate system specifically includes: The centroid and vertex coordinates of the distant vehicle are translated along the positive X-axis of the vehicle coordinate system by a distance S, where S is the actual curve distance between point O and point O' on the center line of the road segment. If the vector direction of the line connecting the nearest point O1 to point O in the curve O→O' is opposite to the direction of the tangent to the center line at point O, then the translated coordinates will be rotated 180° counterclockwise around point O. If point O and point O' are not on the same road segment, the translation distance is determined by the set of points on the center line of the road segment covered by the connection between the two points.

5. The method according to claim 2, characterized in that, The forward projection coordinate system is established based on the vehicle's heading angle, specifically by taking the midpoint of the vehicle's front boundary as the origin, the vehicle's heading angle as the positive X-axis direction, and the normal direction of the vehicle's heading angle to the right as the positive Y-axis direction.

6. The method according to claim 1, characterized in that, The step of projecting the position information of the remote vehicle in the local vehicle coordinate system onto the forward projection coordinate system to obtain the projection of the remote vehicle in the forward projection coordinate system specifically includes: Represent the vehicle and the distant vehicle as rectangles, and calculate the projection point of the rectangle vertex of the distant vehicle onto the rectangle of the vehicle in the forward projection coordinate system. If the rectangular vertex of the distant vehicle and the two rectangular boundaries of the vehicle itself both have projections, then only the rectangular boundary of the vehicle itself closest to the rectangular vertex of the distant vehicle is taken as the projection. If there is no corresponding projection, calculate the projection points of the rectangular vertices of the distant vehicle on the X and Y axes; If the line connecting the vertex of the remote vehicle's rectangle to its projection point along the X-axis or Y-axis first intersects the rectangular boundary of the remote vehicle itself, then the projection point is considered invalid.

7. The method according to claim 6, characterized in that, The method of determining the relative positional relationship between the distant vehicle and the main vehicle based on the projection of the distant vehicle in the forward projection coordinate system, and obtaining the relative position classification result, includes at least one of the following: Based on the projection points and projection line segments of the vehicle in the forward projection coordinate system, relative positional relationships are classified according to preset rules to obtain the relative positional classification results; specifically, the relative positional relationship classification according to preset rules is determined based on the existence of projection points on specific line segments and the comparison of the lengths of the projection line segments in the X and Y axis directions with preset thresholds; or Calculate the overlap rate between the area of ​​the distant vehicle's projection within the safe area of ​​the vehicle and the actual projected area of ​​the distant vehicle; when the overlap rate exceeds a preset threshold, a collision risk is determined, and the relative position classification result is obtained based on the different overlap rates.

8. The method according to claim 1, characterized in that, When the distant vehicle and the vehicle in question are not on the same road segment, the determination of their relative positional relationship to obtain a relative positional classification result specifically includes: Place the intersection of the distant vehicle and the road segment where this vehicle is located in the coordinate system of this vehicle; For curved road sections, an approximation of straight road is used, while preserving the included angle at the intersection of road sections, and using the curved distance to replace the straight distance; Based on the centroids of the distant vehicle and the local vehicle, and the projections of the two vehicles onto the local vehicle's road segment, the relative positions of the distant vehicle and the local vehicle are determined and classified.

9. The method according to claim 1, characterized in that, The process involves calculating a reversal coefficient based on the heading angles of the vehicle and the distant vehicle, as well as the angle between these angles and the direction of travel in the lane. The reversal coefficient is then used to determine the relative heading relationship between the two vehicles, resulting in a relative heading classification. Specifically, this includes: The reverse driving coefficient of this vehicle is calculated based on the angle between the heading angle of this vehicle and the driving direction of the lane in which this vehicle is located; The reverse driving coefficient of the distant vehicle is calculated based on the angle between the heading angle of the distant vehicle and the prescribed driving direction of the lane where the vehicle is located. Determine whether the reverse driving coefficient of this vehicle is equal to that of the distant vehicle. If they are equal, the two vehicles are traveling in the same direction; otherwise, they are traveling in opposite directions.

10. The method according to claim 1, characterized in that, The process involves calculating a reversal coefficient based on the heading angles of the vehicle and the distant vehicle, as well as the angle between these angles and the direction of travel in the lane. The reversal coefficient is then used to determine the relative heading relationship between the two vehicles, resulting in a relative heading classification. Specifically, this includes: Calculate the reverse driving coefficient of this vehicle based on the angle between the heading angle of this vehicle and the direction of travel of the lane it is in; Based on the heading angle of the distant vehicle and its relative relationship with the direction of travel in the lane where the main vehicle is located, and combined with the road segment information where the distant vehicle is located, the reverse driving coefficient of the distant vehicle is calculated. Multiply the reverse driving coefficient of this vehicle by the reverse driving coefficient of the distant vehicle to obtain the reverse driving coefficient of the two vehicles; The relative heading relationship between the two vehicles is determined based on the value of the reverse driving coefficient. If the value is 0, they are judged to be traveling in the same direction; if the value is 1, they are judged to be traveling in opposite directions.

11. A V2X vehicle classification device, characterized in that, include: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform the V2X vehicle classification method as described in any one of claims 1 to 10.

12. A computer program product, characterized in that, Includes computer instructions that instruct a computer device to perform an operation corresponding to the method as described in any one of claims 1 to 10.

Images