A vehicle body region display method, device and apparatus

Abstract

The application provides a vehicle body area display method, device and equipment, the method comprises the following steps: obtaining a target vehicle body edge point set, the target vehicle body edge point set comprises vehicle body edge points corresponding to a plurality of track points; displaying a 3D vehicle body coverage area in a target 3D image based on the target vehicle body edge point set; converting a first position of the vehicle body edge point in a world coordinate system into a second position of the vehicle body edge point in a first coordinate system, the 3D vehicle body coverage area comprises a plurality of 3D sub-areas, and the 3D sub-area comprises a region formed by the second positions of at least three vehicle body edge points; displaying a 2D vehicle body coverage area in a target 2D image based on the target vehicle body edge point set; and converting the first position into a third position of the vehicle body edge point in a second coordinate system, the 2D vehicle body coverage area comprises a plurality of 2D sub-areas, and the 2D sub-area comprises a region formed by the third positions of at least three vehicle body edge points. Through the technical scheme of the application, the safety of the vehicle driving track can be improved.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation, and in particular to a method, device and equipment for displaying vehicle body areas. Background Technology

[0002] With the continuous development of vehicle driver assistance technology and vehicle autonomous driving technology, many vehicles have automatic parking (automatic parking) function. Automatic parking refers to the vehicle automatically parking into a space without the need for manual control by the user. It can help users park automatically, avoid the need for manual parking, and improve the user experience.

[0003] To achieve automatic parking, the vehicle's trajectory (i.e., the path taken during parking) needs to be determined, and the vehicle is then automatically parked based on this trajectory. During automatic parking, it's necessary to detect whether the vehicle has collided with any obstacles. If a collision occurs, the vehicle's trajectory needs to be replanned. However, current technologies lack an effective method for detecting collisions, leading to inaccurate collision detection and low safety of the vehicle's trajectory. In other words, a collision may occur when the vehicle is automatically parked based on its trajectory. Summary of the Invention

[0004] This application provides a method for displaying vehicle body areas, the method comprising: Obtain the target vehicle body edge point set; wherein, the acquired vehicle driving trajectory includes multiple trajectory points, and the target vehicle body edge point set includes the vehicle body edge points corresponding to the multiple trajectory points; Based on the target vehicle body edge point set, a 3D vehicle body coverage area is displayed in the target 3D image; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area includes multiple 3D sub-regions, and each 3D sub-region includes a region composed of the second positions of at least three vehicle body edge points. And / or, display a 2D vehicle body coverage area in the target 2D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area includes multiple 2D sub-regions, and the 2D sub-regions include a region composed of the third positions of at least three vehicle body edge points.

[0005] This application provides a vehicle body area display device, the device comprising: The acquisition module is used to acquire a set of target vehicle body edge points; wherein, the acquired vehicle driving trajectory includes multiple trajectory points, and the set of target vehicle body edge points includes vehicle body edge points corresponding to the multiple trajectory points; The display module is used to display a 3D vehicle body coverage area in a target 3D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area includes multiple 3D sub-regions, each 3D sub-region including an area composed of at least three vehicle body edge points at their second positions; and / or, to display a 2D vehicle body coverage area in a target 2D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area includes multiple 2D sub-regions, each 2D sub-region including an area composed of at least three vehicle body edge points at their third positions.

[0006] This application provides an electronic device, including: a processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the vehicle body area display method of the above example of this application.

[0007] This application provides a computer program product, which includes a computer program that, when executed by a processor, implements the vehicle body area display method of the above example of this application.

[0008] This application provides a machine-readable storage medium storing machine-executable instructions that can be executed by a processor; wherein the processor is used to execute the machine-executable instructions to implement the vehicle body area display method of the above example of this application.

[0009] As can be seen from the above technical solutions, in this embodiment, after obtaining the target vehicle body edge point set, a 3D vehicle body coverage area can be displayed in the target 3D image based on the target vehicle body edge point set. The 3D vehicle body coverage area can be viewed through the target 3D image, and then it can be checked whether the 3D vehicle body coverage area collides with obstacles. This accurately detects whether a collision has occurred between the vehicle and obstacles, and replans the vehicle's trajectory for potential collisions, resulting in a higher safety level for the vehicle's trajectory. When the vehicle is automatically parked based on the vehicle's trajectory, no collision occurs. Furthermore, a 2D vehicle body coverage area can be displayed in the target 2D image based on the target vehicle body edge point set. The 2D vehicle body coverage area can be viewed through the target 2D image, and then it can be checked whether the 2D vehicle body coverage area collides with obstacles. This accurately detects whether a collision has occurred between the vehicle and obstacles, and replans the vehicle's trajectory for potential collisions. This also results in a higher safety level for the vehicle's trajectory, and when the vehicle is automatically parked based on the vehicle's trajectory, no collision occurs.

[0010] The 3D image of the target vehicle can be used to view the 3D area covered by the vehicle, while the 2D image of the target vehicle can be used to view the 2D area covered by the vehicle. This allows for the combined analysis of both images to determine if a collision has occurred between the vehicle's coverage area and an obstacle, making the image display more intuitive and the judgment easier. This mutual verification between the 3D and 2D images improves the accuracy of collision detection and ensures a safe driving path.

[0011] The vehicle body coverage area (3D vehicle body coverage area and 2D vehicle body coverage area) can be determined based on the target vehicle body edge point set. Only multiple trajectory points of the vehicle's driving trajectory are required to determine the target vehicle body edge point set based on these trajectory points, thereby accurately calculating the vehicle body coverage area swept by the vehicle during driving. It can detect and judge the path of blind spots at corners, thus avoiding traffic accidents. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating a vehicle body area display method according to one embodiment of this application; Figure 2 This is a flowchart illustrating the process of acquiring the target vehicle body edge point set in one embodiment of this application; Figure 3A This is a schematic diagram of the vehicle coordinate system, origin, and corner points in one embodiment of this application; Figure 3B This is a schematic diagram of the vehicle coordinate system, trajectory points, and corner points in one embodiment of this application; Figure 3C This is a schematic diagram of the intersection of the current trajectory point and the historical trajectory point in one embodiment of this application; Figure 3D This is a schematic diagram of the cubic spline fitting effect in one embodiment of this application; Figure 3E This is a schematic diagram of sampling N vehicle body edge points at equal intervals in one embodiment of this application; Figure 3F This is a schematic diagram illustrating the transformation between the vehicle body coordinate system and the world coordinate system in one embodiment of this application; Figure 4A This is a schematic flowchart of the target 3D image display process in one embodiment of this application; Figure 4B This is a schematic diagram of the viewpoint matrix in one embodiment of this application; Figure 4C This is a schematic diagram of perspective projection in one embodiment of this application; Figure 4D and Figure 4E This is a schematic diagram of a rendered target 3D image in one embodiment of this application; Figure 5A This is a flowchart illustrating the target 2D image display process in one embodiment of this application; Figure 5B This is a schematic diagram of a rendered target 2D image in one embodiment of this application; Figure 5C This is a conversion process for a close-up image in one embodiment of this application; Figure 5D This is a schematic diagram of a close-up image in one embodiment of this application; Figure 6 This is a schematic diagram of the structure of a vehicle body area display device according to one embodiment of this application; Figure 7 This is a hardware structure diagram of an electronic device according to one embodiment of this application. Detailed Implementation

[0013] This application proposes a vehicle body area display method, which can be applied to electronic devices. See [link to relevant documentation]. Figure 1 The diagram shown is a flowchart of the method for displaying the vehicle body area. This method may include: Step 101: Obtain the target vehicle body edge point set; wherein, the acquired vehicle driving trajectory includes multiple trajectory points, and the target vehicle body edge point set may include the vehicle body edge points corresponding to multiple trajectory points.

[0014] Step 102: Display a 3D vehicle body coverage area in the target 3D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area may include multiple 3D sub-regions, and the 3D sub-regions may include regions composed of the second positions of at least three vehicle body edge points. And / or, display a 2D vehicle body coverage area in the target 2D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area may include multiple 2D sub-regions, and the 2D sub-regions may include regions composed of the third positions of at least three vehicle body edge points.

[0015] For example, the first coordinate system may include a clipping space coordinate system, which converts the first position of each vehicle edge point in the world coordinate system into a second position of the vehicle edge point in the first coordinate system. This may include, but is not limited to: for any vehicle edge point, based on the acquired viewpoint matrix and perspective projection matrix, converting the first position of the vehicle edge point in the world coordinate system into a second position of the vehicle edge point in the clipping space coordinate system; wherein, the viewpoint matrix may be determined based on the camera's right direction vector, the camera's up direction vector, the camera's viewing direction vector, and the camera's translation component; wherein, the perspective projection matrix may be determined based on the distance between the near clipping plane and the camera, the distance between the right clipping plane and the camera, the distance between the left clipping plane and the camera, the distance between the top clipping plane and the camera, the distance between the bottom clipping plane and the camera, and the distance between the far clipping plane and the camera.

[0016] For example, the second coordinate system may include the screen coordinate system. The first position of each vehicle edge point in the world coordinate system in the target vehicle edge point set is converted into the third position of the vehicle edge point in the second coordinate system. This may include, but is not limited to: for any vehicle edge point, based on the acquired camera intrinsic and extrinsic parameters, converting the first position of the vehicle edge point in the world coordinate system into the distortion-free position of the vehicle edge point in the pixel coordinate system; converting the distortion-free position into the distortion position based on the acquired distortion coefficients; and converting the distortion position into the third position in the screen coordinate system based on the configured expected parameters.

[0017] For example, the target vehicle body edge point set may include a first vehicle body edge point set and a second vehicle body edge point set. The first vehicle body edge point set includes multiple left vehicle body edge points, and the second vehicle body edge point set includes multiple right vehicle body edge points. The 3D sub-region includes a region composed of the second positions of three vehicle body edge points, and the 2D sub-region includes a region composed of the third positions of three vehicle body edge points. The three vehicle body edge points include two left vehicle body edge points and one right vehicle body edge point, or one left vehicle body edge point and two right vehicle body edge points. When displaying the 3D sub-region, each pixel within the 3D sub-region is filled; when displaying the 2D sub-region, each pixel within the 2D sub-region is filled.

[0018] For example, the target vehicle edge point set may include a first vehicle edge point set and a second vehicle edge point set. Obtaining the target vehicle edge point set may include, but is not limited to: for any trajectory point of the vehicle's driving trajectory, determining the first candidate point corresponding to the vehicle being at that trajectory point, the first candidate point including two left corner points and two right corner points; wherein, the first candidate point also includes a left intersection point and a right intersection point, the left intersection point being the intersection of the line connecting the two left corner points and the line connecting the two left corner points of the historical trajectory point, and the right intersection point being the intersection of the line connecting the two right corner points and the line connecting the two right corner points of the historical trajectory point. Filtering the first candidate points corresponding to all trajectory points to obtain second candidate points; wherein, for any first candidate point, if the first candidate point is within the rectangular range of the vehicle body corresponding to any trajectory point, then the first candidate point is not considered as a second candidate point; otherwise, the first candidate point is considered as a second candidate point. Sort the multiple second candidate points located on the left side of the vehicle body to obtain a first initial edge point set; generate the first vehicle edge point set based on the first initial edge point set. The multiple second candidate points located on the right side of the vehicle body are sorted to obtain the second initial edge point set; the second vehicle body edge point set is generated based on the second initial edge point set.

[0019] For example, sorting multiple second candidate points located on the left side of the vehicle body to obtain a first initial edge point set may include, but is not limited to: after obtaining the sorted current second candidate points, for each unsorted second candidate point, determining the loss value corresponding to the second candidate point based on the reference distance and reference angle corresponding to the second candidate point; wherein, the reference distance may be the distance between the second candidate point and the current second candidate point, the reference angle may be the angle between the first displacement vector and the second displacement vector, the first displacement vector may be the displacement vector between the second candidate point and the current second candidate point, and the second displacement vector may be the displacement vector between the current second candidate point and the previous second candidate point. Based on the loss values ​​corresponding to each unsorted second candidate point, the second candidate point corresponding to the minimum loss value can be determined as the next second candidate point after the current second candidate point. Then, it can be determined whether the second candidate point is the last second candidate point on the left side of the vehicle body; if not, the second candidate point is taken as the sorted current second candidate point, and the operation of "determining the loss value corresponding to each unsorted second candidate point based on the reference distance and reference angle corresponding to the second candidate point" is returned. If so, then all sorted second candidate points form the first initial edge point set.

[0020] For example, determining the loss value corresponding to the second candidate point based on the reference distance and reference angle can include, but is not limited to: determining the maximum and minimum reference distances based on the reference distances corresponding to each unsorted second candidate point; normalizing the reference distances corresponding to the second candidate point based on the maximum and minimum reference distances to obtain the normalized distance corresponding to the second candidate point; determining the penalty value corresponding to the second candidate point based on the reference angle corresponding to the second candidate point, where the penalty value can be inversely proportional to the reference angle (i.e., negatively correlated); determining the maximum and minimum penalty values ​​based on the penalty values ​​corresponding to each unsorted second candidate point; normalizing the penalty value corresponding to the second candidate point based on the maximum and minimum penalty values ​​to obtain the normalized penalty value corresponding to the second candidate point; and then performing a weighted calculation on the normalized distance and the normalized penalty value corresponding to the second candidate point to obtain the loss value corresponding to the second candidate point.

[0021] The process of sorting multiple second candidate points located on the right side of the vehicle body to obtain the second initial edge point set is similar to the process of obtaining the first initial edge point set, except that the sorting is performed on multiple second candidate points located on the right side of the vehicle body, rather than on multiple second candidate points located on the left side of the vehicle body.

[0022] For example, the process of generating a first vehicle body edge point set based on a first initial edge point set may include, but is not limited to: for two adjacent second candidate points in the first initial edge point set, if the distance between the two second candidate points is greater than a distance threshold, inserting a new second candidate point between the two second candidate points to obtain an updated first initial edge point set; fitting a first curve based on the updated first initial edge point set; sampling N vehicle body edge points on the first curve to obtain a first target edge point set, which may include the positions of the N vehicle body edge points in the vehicle body coordinate system.

[0023] A first vehicle body edge point set is generated based on a first target edge point set; wherein, the position of each vehicle body edge point in the vehicle body coordinate system is converted into the first position of the vehicle body edge point in the world coordinate system, and the first vehicle body edge point set may include the first positions of N vehicle body edge points in the world coordinate system.

[0024] For example, the process of generating a second vehicle body edge point set based on a second initial edge point set may include, but is not limited to: for two adjacent second candidate points in the second initial edge point set, if the distance between these two second candidate points is greater than a distance threshold, inserting a new second candidate point between these two second candidate points to obtain an updated second initial edge point set; fitting the updated second initial edge point set to obtain a second curve; sampling N vehicle body edge points on the second curve to obtain a second target edge point set, which may include the positions of the N vehicle body edge points in the vehicle body coordinate system.

[0025] A second vehicle body edge point set is generated based on the second target edge point set; wherein, the position of each vehicle body edge point in the vehicle body coordinate system is converted into the first position of the vehicle body edge point in the world coordinate system, and the second vehicle body edge point set may include the first positions of N vehicle body edge points in the world coordinate system.

[0026] As can be seen from the above technical solutions, in this embodiment, after obtaining the target vehicle body edge point set, a 3D vehicle body coverage area can be displayed in the target 3D image based on the target vehicle body edge point set. The 3D vehicle body coverage area can be viewed through the target 3D image, and then it can be checked whether the 3D vehicle body coverage area collides with obstacles. This accurately detects whether a collision has occurred between the vehicle and obstacles, and replans the vehicle's trajectory for potential collisions, resulting in a higher safety level for the vehicle's trajectory. When the vehicle is automatically parked based on the vehicle's trajectory, no collision occurs. Furthermore, a 2D vehicle body coverage area can be displayed in the target 2D image based on the target vehicle body edge point set. The 2D vehicle body coverage area can be viewed through the target 2D image, and then it can be checked whether the 2D vehicle body coverage area collides with obstacles. This accurately detects whether a collision has occurred between the vehicle and obstacles, and replans the vehicle's trajectory for potential collisions. This also results in a higher safety level for the vehicle's trajectory, and when the vehicle is automatically parked based on the vehicle's trajectory, no collision occurs.

[0027] The 3D image of the target vehicle can be used to view the 3D area covered by the vehicle, while the 2D image of the target vehicle can be used to view the 2D area covered by the vehicle. This allows for the combined analysis of both images to determine if a collision has occurred between the vehicle's coverage area and an obstacle, making the image display more intuitive and the judgment easier. This mutual verification between the 3D and 2D images improves the accuracy of collision detection and ensures a safe driving path.

[0028] The vehicle body coverage area (3D vehicle body coverage area and 2D vehicle body coverage area) can be determined based on the target vehicle body edge point set. Only multiple trajectory points of the vehicle's driving trajectory are required to determine the target vehicle body edge point set based on these trajectory points, thereby accurately calculating the vehicle body coverage area swept by the vehicle during driving. It can detect and judge the path of blind spots at corners, thus avoiding traffic accidents.

[0029] The technical solutions described above in the embodiments of this application will be explained below in conjunction with specific application scenarios.

[0030] This application proposes a vehicle body area display method, which can be applied to electronic devices. These electronic devices can be in-vehicle devices or control devices for in-vehicle devices (such as personal computers, terminal devices, etc.). Taking an in-vehicle device as an example, the in-vehicle device can be an autonomous driving device (an in-vehicle device supporting an autonomous driving system) or an assisted driving device (an in-vehicle device supporting an assisted driving system). The path determined by the autonomous driving device or the assisted driving device can be called the vehicle's driving trajectory, that is, the driving trajectory of the vehicle deployed by the autonomous driving device or the assisted driving device. For example, the vehicle's driving trajectory can be the driving trajectory during parking, the driving trajectory when leaving a parking space, or other driving trajectories.

[0031] Autonomous driving or driver assistance systems have automatic parking functions, such as AVP (Automated Valet Parking) or PAVP (Public Automated Valet Parking). Through these functions, the autonomous driving or driver assistance systems can automatically park the vehicle without requiring manual control from the user.

[0032] The vehicle in this embodiment is an intelligent vehicle capable of implementing assisted driving or autonomous driving functions. For example, if the vehicle is equipped with an autonomous driving device, then the vehicle is an intelligent vehicle capable of autonomous driving, and the vehicle body area display method is implemented through the autonomous driving device. Alternatively, if the vehicle is equipped with an assisted driving device, then the vehicle is an intelligent vehicle capable of assisted driving, and the vehicle body area display method is implemented through the assisted driving device. The vehicle may also be equipped with cameras and ultrasonic radar. Of course, in addition to cameras and ultrasonic radar, the vehicle may also be equipped with millimeter-wave radar and / or lidar; there is no limitation on this. Obstacles around the vehicle can be detected through cameras and ultrasonic radar; there is no limitation on the obstacle detection method.

[0033] This application proposes a method for displaying a vehicle body region, involving processes such as acquiring a target vehicle body edge point set, displaying a target 3D image, and displaying a target 2D image. In the process of acquiring the target vehicle body edge point set, a target vehicle body edge point set can be acquired. In the process of displaying the target 3D image, a 3D vehicle body coverage area can be displayed in the target 3D image based on the target vehicle body edge point set. In the process of displaying the target 2D image, a 2D vehicle body coverage area can be displayed in the target 2D image based on the target vehicle body edge point set. The method may involve displaying only the 3D vehicle body coverage area in the target 3D image, only the 2D vehicle body coverage area in the target 2D image, or simultaneously displaying both. The processes of acquiring the target vehicle body edge point set, displaying the target 3D image, and displaying the target 2D image are described below.

[0034] First, regarding the process of acquiring the target vehicle body edge point set. See [link / reference]. Figure 2 The diagram shown illustrates the process of acquiring the target vehicle body edge point set. This process may include: Step 201: Obtain the first candidate point in the initial state of the vehicle. The first candidate point includes two left corner points and two right corner points. The two left corner points and two right corner points are the vehicle corner point data.

[0035] For example, when a vehicle needs to move into a parking space from its initial position, the state of the vehicle in that initial position is called the vehicle's initial state. A vehicle coordinate system is established in the vehicle's initial state. This coordinate system has its origin at any position on the vehicle; for example, a coordinate system with its origin at the rear axle center.

[0036] See Figure 3A The diagram shows the vehicle coordinate system, origin, and corner points. Trajectory point M is the origin of the vehicle coordinate system, with the X-axis pointing horizontally to the right and the Y-axis pointing vertically upwards. Within this vehicle coordinate system, vehicle corner point data (i.e., vehicle boundary corner points) can be obtained. This data can include two left corner points and two right corner points, which can be used as first candidate points.

[0037] The coordinates of the rear axle center are (0, 0). It is the left corner point (i.e., the left front corner point). It is the left corner point (i.e., the left rear corner point). It is the right corner point (i.e., the right front corner point). The right corner point (i.e., the right rear corner point) is shown in Formula (1). The left and right corner points mentioned above can be found in Formula (1).

[0038] Formula (1) In formula (1), For vehicle length, For vehicle width, The rear overhang is the distance from the rearmost point of the vehicle to the center of the rear axle. The vehicle length, vehicle width, and rear overhang are all known values.

[0039] Step 202: For any trajectory point of the vehicle's driving trajectory, determine the first candidate point corresponding to the vehicle being at that trajectory point. The first candidate point may include two left corner points and two right corner points.

[0040] For example, when a vehicle needs to move from its initial position into a parking space, its trajectory during movement can be acquired, known as the vehicle's driving trajectory. This trajectory can include multiple trajectory points. For any given trajectory point, it's necessary to determine the vehicle's four corner points in the vehicle's coordinate system (i.e., the coordinate system with the rear axle center as the origin when the vehicle is in its initial position). This corner point data can include two left corner points and two right corner points, which can be used as first-line candidate points. For each trajectory point, the vehicle's corner point data can be obtained; that is, the vehicle's corner point data can be obtained from the same trajectory point.

[0041] See Figure 3B The diagram shows the vehicle coordinate system, trajectory points, and corner points. Given any trajectory point (x...) of the vehicle's trajectory... i , y i , θ i Using the vehicle's coordinate system and rotation and translation matrices, the first candidate points (two left corner points and two right corner points) corresponding to each trajectory point are obtained. i , y i θ represents the position of the trajectory point in the vehicle coordinate system. i This represents the angle of the trajectory point in the vehicle coordinate system. For example, the trajectory point (x) can be determined using the following formulas (2)-(5). i , y i , θ i The first candidate point corresponding to ) Formula (2) Formula (3) Formula (4) Formula (5) In the above formula, This indicates the position of the left corner point (i.e., the front left corner point) in the vehicle coordinate system. This indicates the position of the right corner point (i.e., the right front corner point) in the vehicle coordinate system. This indicates the position of the right corner point (i.e., the right rear corner point) in the vehicle coordinate system. This indicates the position of the left corner point (i.e., the left rear corner point) in the vehicle coordinate system. Indicates the left front corner point Position in the vehicle body coordinate system Indicates the right front corner point Position in the vehicle body coordinate system Indicates the right rear corner point Position in the vehicle body coordinate system Indicates the left rear corner point Position in the vehicle body coordinate system.

[0042] In summary, for any trajectory point (x) of the vehicle's driving trajectory... i , y i , θ i (Based on the left front corner point in the vehicle body coordinate system) Right front corner Right rear corner and left rear corner point The trajectory point (x) can be determined. i ,y i , θ i The first candidate points (two left corner points and two right corner points) are corresponding to this.

[0043] Step 203: For any trajectory point of the vehicle's driving trajectory, determine the first candidate point corresponding to the vehicle's position at that trajectory point. The first candidate point may include a left intersection point and a right intersection point. The left intersection point is optional, meaning that the first candidate point corresponding to this trajectory point may or may not include a left intersection point. The right intersection point is also optional, meaning that the first candidate point corresponding to this trajectory point may or may not include a right intersection point.

[0044] For example, regarding the left intersection point, it is the intersection of the line connecting the two left corner points (such as the left front corner point and the left rear corner point) and the line connecting the two left corner points of the historical trajectory point. The historical trajectory point can be the first trajectory point preceding this trajectory point, or it can be multiple trajectory points preceding this trajectory point. Similarly, regarding the right intersection point, it is the intersection of the line connecting the two right corner points (such as the right front corner point and the right rear corner point) and the line connecting the two right corner points of the historical trajectory point.

[0045] For example, see Figure 3C The image shows a schematic diagram of the intersection points between the current trajectory point and historical trajectory points. (x) i_lf y i_lf ) and (x i_lr y i_lr (x) are the two left corner points corresponding to the current trajectory point, and the line connecting these two left corner points is denoted as line a1.i-1_lf y i-1_lf ) and (x i-1_lr y i-1_lr Let a1 and a2 be the two left corner points corresponding to the historical trajectory points. The line connecting these two left corner points is denoted as line a2. Clearly, the intersection of line a1 and line a2 is the left intersection point, meaning the left intersection point can be (x...). i_left y i_left ).

[0046] (x) i_rf y i_rf ) and (x i_rr y i_rr (x) are the two right corner points corresponding to the current trajectory point, and the line connecting these two right corner points is denoted as line a3. i-1_rf y i-1_rf ) and (x i-1_rr y i-1_rr Let a and b be the two right corner points corresponding to the historical trajectory points, and let a4 be the line connecting these two right corner points. Clearly, the intersection of line a3 and line a4 is the right intersection point, meaning the right intersection point can be (x...). i_right y i_right ).

[0047] For example, in order to determine the left and right intersection points, a line segment intersection detection method can be used to calculate the left and right intersection points of adjacent rectangles. This will be explained below.

[0048] First, the positional relationship between the two line segments is determined using the following formula: If den is 0, it means that the two line segments are parallel or collinear, and the intersection point does not need to be calculated; if den is not 0, the position of the intersection point on the line segment can be calculated.

[0049] Then, if den is not 0, the position adjustment amount corresponding to the intersection point can be determined using the following formula: , .

[0050] like and If both points are within the interval [0, 1], it means the intersection point lies on both line segments. The coordinates of the intersection point are calculated as follows: , .like Not in the interval [0, 1], and / or, If the intersection point is not within the interval [0, 1], it means that the intersection point is on the extension of the two line segments. The coordinates of the intersection point are not calculated. In other words, the coordinates of the intersection point do not need to be calculated in this case.

[0051] This indicates the left front corner point corresponding to the current trajectory point. This indicates the left rear corner point corresponding to the current trajectory point. This represents the left front corner point corresponding to the historical trajectory point. This indicates the left rear corner point corresponding to the historical trajectory point. This indicates the intersection point on the left. Or, This indicates the right front corner point corresponding to the current trajectory point. This indicates the right rear corner point corresponding to the current trajectory point. This indicates the right front corner point corresponding to the historical trajectory point. This indicates the right rear corner point corresponding to the historical trajectory point. This indicates the intersection point on the right.

[0052] Step 204: Filter the first candidate points corresponding to all trajectory points to obtain the second candidate points; wherein, for any first candidate point, if the first candidate point is within the rectangular range of the vehicle body corresponding to any trajectory point, the first candidate point is not used as the second candidate point; otherwise, the first candidate point is used as the second candidate point.

[0053] For example, for any trajectory point in a vehicle's driving trajectory, this trajectory point corresponds to multiple first candidate points, such as two left corner points, two right corner points, a left intersection point, and a right intersection point. Based on this, these first candidate points can be filtered, and the remaining first candidate points are used as second candidate points.

[0054] For example, after obtaining multiple first candidate points, each first candidate point is traversed sequentially. For the currently traversed first candidate point (such as the left corner point, right corner point, left intersection point, or right intersection point), the following method is used to determine whether the first candidate point is within the vehicle body rectangle: translate the first candidate point to the vehicle body coordinate system (a coordinate system with the rear axle center as the origin), and rotate the first candidate point in the opposite direction so that the rotated first candidate point is aligned with the original rectangle (i.e., the vehicle body rectangle area in the initial state of the vehicle, as determined by...). , for, and Align the regions formed by the rotation and determine whether the first candidate point after rotation is within the original rectangle. If yes, it means that the first candidate point is within the vehicle body rectangle and is not considered as the second candidate point. If no, it means that the first candidate point is not within the vehicle body rectangle and is considered as the second candidate point.

[0055] For example, the first candidate point after rotation can be determined using the following formula (6): Formula (6) In formula (6), This represents the coordinates of the first candidate point in the vehicle body coordinate system. This represents the coordinates of the rear axle center (or any other location besides the rear axle center, such as the center point of the vehicle body rectangle, without restriction) in the vehicle body coordinate system within the i-th vehicle body rectangle. The angle between the line connecting the center point of the i-th rectangular area of ​​the vehicle body and the origin of the vehicle body coordinate system and the horizontal axis of the vehicle body coordinate system.

[0056] For any trajectory point of the vehicle's driving trajectory, taking the i-th trajectory point as an example, we can determine the four corner points corresponding to the i-th trajectory point. The rectangular area formed by these four corner points is the i-th vehicle body rectangle. We can then obtain the coordinates of the center point of the i-th vehicle body rectangle in the vehicle body coordinate system.

[0057] This represents the first candidate point after rotation. If the first candidate point after rotation lies within the original rectangle, then it is considered the first candidate point. If a point is located within the i-th vehicle body rectangle, it is not considered a second candidate point. If the rotated first candidate point is not within the original rectangle, it means that the first candidate point is not within the i-th vehicle body rectangle. The process continues to determine whether the first candidate point is within another vehicle body rectangle, and so on, until the first candidate point is within a certain vehicle body rectangle, or the first candidate point is not within any of the vehicle body rectangles. In this case, the first candidate point can be considered a second candidate point.

[0058] In summary, after processing all first candidate points as described above, multiple second candidate points can be obtained. These second candidate points may include left corner points, right corner points, left intersection points, and right intersection points.

[0059] Step 205: Sort the multiple second candidate points located on the left side of the vehicle body to obtain the first initial edge point set; sort the multiple second candidate points located on the right side of the vehicle body to obtain the second initial edge point set.

[0060] For example, second candidate points located on the left side of the vehicle can be selected from all second candidate points, such as all left corner points and left intersection points. Based on this, the second candidate points located on the left side of the vehicle are sorted to obtain the first initial edge point set. Similarly, second candidate points located on the right side of the vehicle can be selected from all second candidate points, such as all right corner points and right intersection points. Based on this, the second candidate points located on the right side of the vehicle are sorted to obtain the second initial edge point set. The process of obtaining the second initial edge point set is similar to that of the first initial edge point set; the following explanation will use the process of obtaining the first initial edge point set as an example.

[0061] Based on the second candidate point located on the left side of the vehicle body, the first initial edge point set is obtained using the following steps: Step S11: After obtaining the sorted current second candidate points, when filtering the next second candidate point after the current second candidate point, for each unsorted second candidate point, determine the reference distance corresponding to the second candidate point. The reference distance can be the distance between the second candidate point and the current second candidate point.

[0062] For example, first determine the first second candidate point within the first initial edge point set. For example, when the vehicle is moving forward, the left front corner of the vehicle in its initial state is used as the second candidate point. When the vehicle is moving backward, the left rear corner point in the initial state of the vehicle is the second candidate point. Then, determine the second candidate point within the first initial edge point set. For example, with the second candidate point The nearest second candidate point is used as the second candidate point. Therefore, when determining the third and subsequent second candidate points within the first initial edge point set, the reference distance and reference angle can be used as a basis for determination, which will be explained below.

[0063] Record the currently sorted second candidate point as the second candidate point. The second candidate point The previous second candidate point is denoted as the second candidate point. Each unsorted second candidate point is recorded as a second candidate point. Obviously, it is necessary to select from all second candidate points. Select one of the second candidate points as the second candidate point. The next second candidate point, as i increases, can sort all the second candidate points in turn.

[0064] For each unsorted second candidate point The second candidate point The corresponding reference distance is denoted as ,but . Indicates the second candidate point coordinates Indicates the second candidate point coordinates express and The distance between them.

[0065] Step S12: Based on the reference distances corresponding to each unsorted second candidate point, determine the maximum reference distance and the minimum reference distance; normalize the reference distances corresponding to the second candidate point based on the maximum reference distance and the minimum reference distance to obtain the normalized distance corresponding to the second candidate point.

[0066] For example, the normalized distance corresponding to the second candidate point can be determined using the following formula: .in, Indicates the second candidate point The corresponding reference distance, This represents the minimum reference distance among the reference distances corresponding to all unsorted second candidate points. This represents the maximum reference distance among the reference distances corresponding to all unsorted second candidate points. Indicates the second candidate point The corresponding normalized distance.

[0067] Step S13: For each unsorted second candidate point, determine the reference angle corresponding to the second candidate point. The reference angle can be the angle between the first displacement vector and the second displacement vector. The first displacement vector can be the displacement vector between the second candidate point and the current second candidate point. The second displacement vector can be the displacement vector between the current second candidate point and the previous second candidate point.

[0068] For example, for each unsorted second candidate point We can define two displacement vectors, which are called the first displacement vector. Second displacement vector . It is the second candidate point With the second candidate point The displacement vector between them , It is the second candidate point With the second candidate point The displacement vector between them Based on this, the angle between the first displacement vector and the second displacement vector can be calculated using the dot product formula. , denoted as the second candidate point The corresponding reference angle.

[0069] For example, calculating the angle between the first displacement vector and the second displacement vector. The derivation of the formula is as follows: , , , As can be seen from the above formula, it can be based on the first displacement vector. Second displacement vector Determine the included angle That is, to obtain the reference angle.

[0070] Step S14: For each unsorted second candidate point, determine the penalty value corresponding to the second candidate point based on the reference angle corresponding to the second candidate point. The penalty value can be inversely proportional to the reference angle (i.e., negatively correlated).

[0071] For example, the second candidate point can be determined using the following formula. The corresponding penalty value (i.e., a penalty that increases with the steering angle): . Indicates the second candidate point The corresponding reference angle, Indicates the second candidate point The corresponding penalty value. Clearly, in A value of 0 indicates a U-shaped turn, with the maximum penalty value. for When the time is right, it indicates going straight, and the penalty value is the minimum (0).

[0072] Step S15: Based on the penalty values ​​corresponding to each unsorted second candidate point, determine the maximum penalty value and the minimum penalty value; based on the maximum penalty value and the minimum penalty value, normalize the penalty value corresponding to the second candidate point to obtain the normalized penalty value corresponding to the second candidate point.

[0073] For example, the normalized penalty value corresponding to the second candidate point can be determined using the following formula: .in, Indicates the second candidate point The corresponding normalized penalty value, This represents the minimum penalty value among the penalty values ​​corresponding to each unsorted second candidate point. This represents the maximum penalty value among the penalty values ​​corresponding to the unsorted second candidate points. Indicates the second candidate point The corresponding normalized penalty value.

[0074] Step S16: Perform a weighted calculation on the normalized distance and the normalized penalty value corresponding to the second candidate point to obtain the loss value corresponding to the second candidate point.

[0075] For example, the second candidate point can be determined using the following formula. Corresponding loss value: . Indicates the second candidate point The corresponding loss value, Indicates the second candidate point The corresponding normalized distance, Indicates the second candidate point The corresponding normalized penalty value. and This represents a weighting coefficient, which can be a value between 0 and 1, or greater than 1. It controls the importance of distance and steering and can be configured according to actual needs. For example, the weighting coefficient... The weighting coefficient is 0.9. It is 0.1.

[0076] Step S17: Based on the loss values ​​corresponding to each unsorted second candidate point, determine the second candidate point corresponding to the minimum loss value as the next second candidate point after the current second candidate point.

[0077] Then, determine the second candidate point. Is it the last second candidate point on the left side of the vehicle? If not, then select the second candidate point corresponding to the minimum loss value. Updated to the currently sorted second candidate point Repeat steps S11-S17, and so on, until the last second candidate point on the left side of the vehicle is reached. In this way, all second candidate points have been sorted, and the first initial edge point set is obtained.

[0078] Step 206: Perform interpolation on the first initial edge point set to obtain the updated first initial edge point set. Perform interpolation on the second initial edge point set to obtain the updated second initial edge point set. For ease of description, the following process will use the processing of the first initial edge point set as an example.

[0079] For example, the first initial edge point set includes a plurality of sorted second candidate points. For any two adjacent second candidate points in the first initial edge point set, if the distance between these two second candidate points is greater than a distance threshold, a new second candidate point is inserted between them; if the distance between these two second candidate points is not greater than the distance threshold, no new second candidate point is inserted between them. After performing the above operation on any two second candidate points, an updated first initial edge point set is obtained.

[0080] For example, the distance d between two adjacent second candidate points can be determined using the following formula: , This represents the coordinates of a second candidate point. This represents the coordinates of another second candidate point. If the distance d is greater than a pre-configured distance threshold d... max If so, then linear interpolation is performed between these two second candidate points to supplement the second candidate points.

[0081] For example, the new second candidate point inserted between these two second candidate points is: Then the new second candidate point can be determined using the following formula. : , If a new second candidate point is inserted, then j is 0. If two new second candidate points are inserted, then j for the two new second candidate points is 0 and 1 respectively, and so on. Obviously, the index j of the inserted new second candidate point starts from 0, and at least one second candidate point can be inserted.

[0082] Step 207: Fit the updated first initial edge point set to obtain the first curve; Step 206 is optional. If Step 206 is not executed, the first curve is obtained by fitting the first initial edge point set (i.e., the first initial edge points without interpolation). Fit the updated second initial edge point set to obtain the second curve. The following process takes the processing of the first initial edge point set as an example.

[0083] For example, fitting means approximating the overall trend of a set of data points using a mathematical model or curve. A first curve can be obtained by fitting all second candidate points within the first initial edge point set, and there are no restrictions on the fitting method. For instance, a cubic spline fitting method can be used to fit all second candidate points within the first initial edge point set to obtain the first curve; the cubic spline fitting method is merely an example.

[0084] Cubic spline fitting enriches the point set data and ensures consistent data volume between the first and second initial edge point sets. Cubic splines are applied to both sets, and the corresponding y-values ​​are calculated using an interpolation object to form smooth curves. The smooth curve corresponding to the first initial edge point set is the first curve, and the smooth curve corresponding to the second initial edge point set is the second curve. For example, cubic splines construct piecewise cubic polynomials to ensure continuity of function values, first and second derivatives at connection points between adjacent segments. See also... Figure 3D The figure shows a schematic diagram of the cubic spline fitting effect.

[0085] Step 208: Sample N vehicle edge points on the first curve to obtain a first target edge point set. The first target edge point set may include the positions of the N vehicle edge points in the vehicle coordinate system.

[0086] N vehicle edge points are sampled on the second curve to obtain a second target edge point set, which may include the positions of the N vehicle edge points in the vehicle coordinate system.

[0087] For example, after obtaining the first curve, N points can be discretized and sampled on the first curve, and these N points are called vehicle body edge points. For instance, N vehicle body edge points can be sampled at equal intervals on the first curve, and these N vehicle body edge points are their positions (coordinates) in the vehicle body coordinate system. Based on this, the first target edge point set can include the positions of N vehicle body edge points in the vehicle body coordinate system.

[0088] Similarly, after obtaining the second curve, N points can be discretized and sampled on this second curve. These N points are called vehicle body edge points. For example, N vehicle body edge points can be sampled at equal intervals on the second curve. These N vehicle body edge points are their positions (coordinates) in the vehicle body coordinate system. Based on this, the second target edge point set can include the positions of N vehicle body edge points in the vehicle body coordinate system.

[0089] For example, see Figure 3E The diagram illustrates the sampling of N vehicle body edge points at equal intervals. N vehicle body edge points are sampled at equal intervals on the first curve; for example, eight vehicle body edge points are denoted as Left[n-1], Left[n], Left[n+1], Left[n+2], Left[n+3], Left[n+4], Left[n+5], and Left[n+6]. N (eight) vehicle body edge points are sampled at equal intervals on the second curve, denoted as Right[n], Right[n+1], Right[n+2], Right[n+3], Right[n+4], Right[n+5], Right[n+6], and Right[n+7].

[0090] Based on this, triangular segmentation can be performed, with the segmented areas corresponding to the vehicle body coverage area. For example, the first and second vehicle body edge points on the first curve and the first vehicle body edge point on the second curve form a triangle; the second vehicle body edge point on the first curve and the first and second vehicle body edge points on the second curve form a triangle; the second and third vehicle body edge points on the first curve and the second vehicle body edge point on the second curve form a triangle; the third vehicle body edge point on the first curve and the second and third vehicle body edge points on the second curve form a triangle, and so on. Figure 3E The example below illustrates how to cut a triangle by connecting the vehicle body edge points with equal subscripts and then connecting them to the vehicle body edge point with subscript -1.

[0091] Step 209: Generate a first vehicle body edge point set based on the first target edge point set; wherein, the position of each vehicle body edge point in the vehicle body coordinate system is converted into the first position of the vehicle body edge point in the world coordinate system, and the first vehicle body edge point set may include the first positions of N vehicle body edge points in the world coordinate system.

[0092] In addition, a second vehicle body edge point set is generated based on the second target edge point set; wherein, the position of each vehicle body edge point in the vehicle body coordinate system is converted into the first position of the vehicle body edge point in the world coordinate system, and the second vehicle body edge point set may include the first positions of N vehicle body edge points in the world coordinate system.

[0093] For example, the first target edge point set (second target edge point set) includes the positions of the vehicle body edge points in the vehicle body coordinate system (the vehicle body coordinate system is a coordinate system with the rear axle center as the origin). These positions need to be transformed to the world coordinate system. See [link / reference]. Figure 3F The diagram shown is a conversion diagram between the vehicle body coordinate system and the world coordinate system. The world coordinate system is a coordinate system with the center of the vehicle body as the origin.

[0094] Based on this, for any vehicle body edge point within the first target edge point set (second target edge point set), the first position of the vehicle body edge point in the world coordinate system can be determined using the following formula (7): Formula (7) In formula (7), This indicates the position of a point on the edge of the vehicle body in the vehicle body coordinate system. This indicates the first position of a point on the edge of the vehicle body in the world coordinate system. For vehicle length, It is a rear suspension.

[0095] In summary, for any vehicle body edge point within the first target edge point set (second target edge point set), the first position of that vehicle body edge point in the world coordinate system can be obtained, and the first positions of all vehicle body edge points in the world coordinate system constitute the first vehicle body edge point set (second vehicle body edge point set).

[0096] This completes the acquisition of the target vehicle body edge point set. The target vehicle body edge point set can include a first vehicle body edge point set and a second vehicle body edge point set. The first vehicle body edge point set includes multiple left-side vehicle body edge points, and the second vehicle body edge point set includes multiple right-side vehicle body edge points. For example, if the vehicle's driving trajectory includes multiple trajectory points, the first vehicle body edge point set can include the left-side vehicle body edge points corresponding to the multiple trajectory points, and the second vehicle body edge point set can include the right-side vehicle body edge points corresponding to the multiple trajectory points.

[0097] Second, regarding the process of displaying the target 3D image. See also... Figure 4A The diagram shown illustrates the process of displaying a target 3D image, which may include the following steps: Step 401: Obtain the configured viewpoint matrix. The viewpoint matrix can be determined based on the camera's right direction vector, the camera's top direction vector, the camera's viewing direction vector, and the camera's translation component.

[0098] For example, see Figure 4B The diagram shows a viewpoint matrix. The vehicle's trajectory is observed and imaged by a camera. The observation matrix can be a viewpoint matrix, which can be found in formula (8). = Formula (8) In formula (8), Represents the viewpoint matrix. This represents the rightward direction vector of the camera, and can be a known value. This represents the upward direction vector of the camera, which can be a known value. This represents the camera's viewing direction vector, which can be a known value. The translation component of the camera can be a known value. In summary, the viewpoint matrix can be pre-configured. .

[0099] Step 402: Obtain the configured perspective projection matrix. The perspective projection matrix is ​​determined based on the distances between the near clipping plane and the camera, the right clipping plane and the camera, the left clipping plane and the camera, the top clipping plane and the camera, the bottom clipping plane and the camera, and the far clipping plane and the camera.

[0100] For example, see Figure 4C The diagram shows a perspective projection, which transforms the object from the viewing space to the clipping space and achieves the perspective projection deformation effect. The perspective projection matrix can be found in formula (9): Formula (9) In formula (9), Represents the perspective projection matrix. This indicates the distance between the near-end clipping plane and the camera. Figure 4C The Near clip plane. This indicates the distance between the right-side clipping plane and the camera, corresponding to... Figure 4C The Right in the middle. This indicates the distance between the left-side cutout plane and the camera, corresponding to... Figure 4C Left in the middle. This indicates the distance between the top clipping plane and the camera, corresponding to... Figure 4C Top in the middle. This indicates the distance between the bottom cutout surface and the camera, corresponding to... Figure 4C Bottom in the middle. Indicates the distance between the far-end clipping plane and the camera, corresponding to Figure 4C The far clip plane in the image. In summary, the perspective projection matrix can be pre-configured.

[0101] Step 403: For any vehicle edge point in the target vehicle edge point set, based on the viewpoint matrix and perspective projection matrix, convert the first position of the vehicle edge point in the world coordinate system to the second position of the vehicle edge point in the first coordinate system. For example, the first coordinate system can be the clipping space coordinate system or other coordinate systems. Taking the clipping space coordinate system as an example, the first position of the vehicle edge point in the world coordinate system is converted to the second position of the vehicle edge point in the clipping space coordinate system.

[0102] For example, for any edge point of the vehicle body, the transformation from the world coordinate system to the clipping space coordinate system is completed using formula (10), and the second position of the edge point of the vehicle body in the clipping space coordinate system is obtained: Formula (10) In formula (10), This indicates the first position of the edge point of the vehicle body in the world coordinate system. This indicates the second position of the vehicle body edge point in the clipping space coordinate system. This represents the homogeneous term of the vehicle body edge point in the clipping space coordinate system. Obviously, based on formula (10), the first position in the world coordinate system can be transformed into the second position in the clipping space coordinate system based on the viewpoint matrix and perspective projection matrix.

[0103] Step 404: Display the 3D vehicle body coverage area in the target 3D image. The 3D vehicle body coverage area includes multiple 3D sub-regions, and each 3D sub-region includes an area composed of the second positions of at least three vehicle body edge points.

[0104] For example, a 3D sub-region may include a region consisting of the second positions of three vehicle body edge points. The three vehicle body edge points may include two left vehicle body edge points and one right vehicle body edge point, or the three vehicle body edge points may include one left vehicle body edge point and two right vehicle body edge points.

[0105] For example, with Figure 3E The model is divided into triangles. The first and second body edge points in the first set of body edge points and the first body edge point in the second set of body edge points are used as three body edge points. Based on the second position of these three body edge points in the clipping space coordinate system, a 3D sub-region is formed and displayed in the target 3D image. When displaying this 3D sub-region, each pixel inside the 3D sub-region is filled, that is, each pixel inside the triangle is filled (e.g., with color, which can be the same color or different colors), thus displaying the 3D sub-region as a rendered effect. Rendering refers to the process of converting a model, scene, or data into an image or visualization result.

[0106] The second body edge point in the first body edge point set and the first and second body edge points in the second body edge point set are used as three body edge points. Based on the second position of these three body edge points in the clipping space coordinate system, a 3D sub-region is formed and displayed in the target 3D image.

[0107] Similarly, multiple triangles can be cut, each corresponding to a 3D sub-region, thus displaying multiple 3D sub-regions in the target 3D image. The combination of these multiple 3D sub-regions can be called the 3D vehicle body coverage area, i.e., the 3D vehicle body coverage area displayed in the target 3D image. For example, see... Figure 4D and Figure 4E The image shown is a schematic diagram of the rendered target 3D image, with the blue area representing the 3D vehicle body coverage area.

[0108] Third, regarding the display process of the target 2D image. See also Figure 5A The diagram shown illustrates the process of displaying a target 2D image, which may include the following steps: Step 501: For any vehicle edge point in the target vehicle edge point set, based on the camera intrinsic and extrinsic parameters, convert the first position of the vehicle edge point in the world coordinate system into the distortion-free position of the vehicle edge point in the pixel coordinate system (i.e., the pixel position in the image without distortion).

[0109] For example, for any edge point of the vehicle body, the transformation from the world coordinate system to the pixel coordinate system is completed using formula (11), and the distortion-free position of the edge point of the vehicle body in the pixel coordinate system is obtained: Formula (11) In formula (11), This indicates the first position of a point on the edge of the vehicle body in the world coordinate system. This indicates the distortion-free position of a point on the edge of the vehicle body in the pixel coordinate system. For camera internal parameters, , and These are the principal point coordinates of the image, and the positions of the camera's optical center in the x and y directions. and It is the focal length of the camera in the x and y directions, measured in pixels. This represents the rotation parameters (i.e., position angle parameters) in the camera's extrinsic parameters. ). This represents the translation parameters (i.e., position parameters) in the camera's extrinsic parameters. ). This represents the camera's depth parameters, specifically the pixel depth value.

[0110] Step 502: Based on the obtained distortion coefficients, convert the undistorted position into a distorted position (i.e., the pixel position in the distorted image), which can be a distorted position in a fisheye image.

[0111] For example, the distortion-free location can be converted into a distortion location using the following formula (12): Formula (12) In the above formula, This indicates the location without distortion. Indicates the location of the distortion. This represents the distortion coefficient, which can be determined based on the distance distortion coefficient and the angle distortion coefficient. and This refers to the position of the camera's optical center in the x and y directions. The following explains this process.

[0112] For example, a coefficient table can be generated in advance, which includes distance distortion coefficients. and angular distortion coefficient The mapping relationship between them. For example, distance distortion coefficient. With angle distortion coefficient The following expression applies between them: , =1, The distortion parameters are intrinsic camera parameters and can be known values. Based on this, the angular distortion coefficients can be iterated over. Multiple values ​​for the angle distortion coefficient For each value, the distance distortion coefficient is obtained through the above expression. The value of can be obtained in this way, thus yielding the distance distortion coefficient. and angular distortion coefficient The coefficient table between them.

[0113] In step 502, based on the distortion-free position The distance distortion coefficient can be determined in the following way. : , , .

[0114] Obviously, due to the absence of distortion position Camera internal parameters ( , ), camera internal parameters ( , Since all the values ​​are known, the distance distortion coefficient can be obtained. After obtaining the distance distortion coefficient Then, by consulting the coefficient table, the distance distortion coefficient can be obtained. Corresponding angular distortion coefficient Based on this, the distortion coefficient can be determined using the following formula. : .

[0115] Step 503: Based on the configured desired parameters, convert the distortion location into the third position of the vehicle body edge point in the second coordinate system. For example, the second coordinate system can be the screen coordinate system (i.e., the coordinate system of the screen displaying the target 2D image), or it can be another coordinate system. Taking the screen coordinate system as an example, the distortion location is converted into the third position of the vehicle body edge point in the screen coordinate system. In summary, the first position of the vehicle body edge point in the world coordinate system can be converted into the third position of the vehicle body edge point in the screen coordinate system.

[0116] For example, the distortion position can be converted to the third position in the screen coordinate system using the following formula (13): Formula (13) In formula (13), This indicates the third position of the vehicle body edge point in the screen coordinate system. This indicates the location of the distortion. This represents the rotation angle parameter, which is the desired parameter. It indicates the angle that the user expects the target 2D image to be relative to the horizontal axis of the screen coordinate system. This rotation angle parameter can be configured according to actual needs. and This represents the translation parameter, which is the desired parameter, indicating the distance of the user's desired 2D image relative to the origin of the screen coordinate system. Indicates the distance along the horizontal axis. (This represents the distance along the vertical axis), and this translation parameter can be configured according to actual needs. In summary, the distortion position can be determined based on the desired parameters. Convert to the third position of the vehicle body edge point in the screen coordinate system .

[0117] In summary, for any vehicle edge point, based on steps 501-503, the first position of the vehicle edge point in the world coordinate system can be converted into the third position of the vehicle edge point in the screen coordinate system.

[0118] Step 504: Display the 2D vehicle body coverage area in the target 2D image. The 2D vehicle body coverage area includes multiple 2D sub-regions, and each 2D sub-region includes an area composed of the third position of at least three vehicle body edge points.

[0119] For example, a 2D sub-region may include a region consisting of the third position of three vehicle body edge points. The three vehicle body edge points may include two left vehicle body edge points and one right vehicle body edge point, or the three vehicle body edge points may include one left vehicle body edge point and two right vehicle body edge points.

[0120] For example, with Figure 3E The image is divided into triangles. The first and second body edge points from the first set of body edge points and the first body edge point from the second set of body edge points are used as three body edge points. Based on the third position of these three body edge points in the screen coordinate system, a 2D sub-region is formed and displayed in the target 2D image. When displaying this 2D sub-region, each pixel inside the 2D sub-region is filled, that is, each pixel inside the triangle is filled (e.g., with color, which can be the same color or different colors), thus displaying the 2D sub-region as a rendered effect.

[0121] The second body edge point in the first body edge point set and the first and second body edge points in the second body edge point set are used as three body edge points. Based on the third position of these three body edge points in the screen coordinate system, a 2D sub-region is formed and the 2D sub-region is displayed in the target 2D image.

[0122] Similarly, multiple triangles can be cut, each corresponding to a 2D sub-region, thus displaying multiple 2D sub-regions in the target 2D image. The combination of these multiple 2D sub-regions can be called the 2D vehicle body coverage area, i.e., the 2D vehicle body coverage area displayed in the target 2D image. For example, see... Figure 5B The image shown is a schematic diagram of the rendered target 2D image, with the blue area representing the 2D vehicle body coverage area.

[0123] For example, a target 2D image can be referred to as a close-up image (2D close-up image). See also... Figure 5C The diagram illustrates the conversion process for a close-up image. First, the initial position in world coordinates is converted to a distortion-free position in pixel coordinates. Then, the distortion-free position in pixel coordinates is converted to a distorted position in pixel coordinates (i.e., the distorted fisheye position). Finally, the distorted position in pixel coordinates is converted to a close-up image in screen coordinates. See also... Figure 5D The image shown is a schematic diagram of a close-up view. The first image is of the undistorted area, the second image is of the distorted area, and the third image is a close-up view.

[0124] In one possible implementation, all data involved in this embodiment is obtained and used only with the knowledge and authorization of the relevant users.

[0125] As seen from the above technical solutions, the vehicle body coverage area (3D and 2D vehicle body coverage areas) can be determined based on the target vehicle body edge point set. Only multiple trajectory points of the vehicle's driving trajectory are needed to determine the target vehicle body edge point set, thereby accurately calculating the vehicle body coverage area swept by the vehicle during driving. This enables path detection and judgment of blind spots at corners, preventing traffic accidents. By viewing the 3D vehicle body coverage area through the target 3D image and the 2D vehicle body coverage area through the target 2D image, the combined 3D and 2D images can be used to check whether a collision has occurred between the vehicle body coverage area and an obstacle, making the image display more intuitive and the judgment easier. The mutual confirmation between the target 3D and 2D images improves the accuracy of collision detection and ensures the safety of the driving path.

[0126] Based on the same concept as the above method, this application proposes a vehicle body area display device, see [link to relevant documentation]. Figure 6 The diagram shown is a structural schematic of the device, which may include: Acquisition module 61 is used to acquire a target vehicle body edge point set; wherein the acquired vehicle driving trajectory includes multiple trajectory points, and the target vehicle body edge point set includes vehicle body edge points corresponding to the multiple trajectory points; display module 62 is used to display a 3D vehicle body coverage area in a target 3D image based on the target vehicle body edge point set; wherein the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area includes multiple 3D sub-regions, each 3D sub-region including an area composed of at least three vehicle body edge points at their second positions; and / or, to display a 2D vehicle body coverage area in a target 2D image based on the target vehicle body edge point set; wherein the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area includes multiple 2D sub-regions, each 2D sub-region including an area composed of at least three vehicle body edge points at their third positions.

[0127] For example, the first coordinate system includes a clipping space coordinate system. When the display module 62 converts the first position of each vehicle edge point in the world coordinate system in the target vehicle edge point set into the second position of the vehicle edge point in the first coordinate system, it specifically performs the following: for any vehicle edge point, based on the viewpoint matrix and the perspective projection matrix, converts the first position of the vehicle edge point in the world coordinate system into the second position of the vehicle edge point in the clipping space coordinate system; wherein, the viewpoint matrix is ​​determined based on the camera's right direction vector, the camera's up direction vector, the camera's viewing direction vector, and the camera's translation component; wherein, the perspective projection matrix is ​​determined based on the distance between the near clipping plane and the camera, the distance between the right clipping plane and the camera, the distance between the left clipping plane and the camera, the distance between the top clipping plane and the camera, the distance between the bottom clipping plane and the camera, and the distance between the far clipping plane and the camera.

[0128] For example, the second coordinate system includes the screen coordinate system. When the display module 62 converts the first position of each vehicle edge point in the world coordinate system in the target vehicle edge point set into the third position of the vehicle edge point in the second coordinate system, it specifically performs the following steps: for any vehicle edge point, based on camera intrinsic and extrinsic parameters, converts the first position of the vehicle edge point in the world coordinate system into the distortion-free position of the vehicle edge point in the pixel coordinate system; converts the distortion-free position into a distorted position based on the acquired distortion coefficients; and converts the distorted position into the third position in the screen coordinate system based on the desired parameters.

[0129] For example, the target vehicle body edge point set includes a first vehicle body edge point set and a second vehicle body edge point set. The first vehicle body edge point set includes multiple left-side vehicle body edge points, and the second vehicle body edge point set includes multiple right-side vehicle body edge points. The 3D sub-region includes a region composed of the second positions of three vehicle body edge points, and the 2D sub-region includes a region composed of the third positions of three vehicle body edge points. The three vehicle body edge points include two left-side vehicle body edge points and one right-side vehicle body edge point, or one left-side vehicle body edge point and two right-side vehicle body edge points. When displaying the 3D sub-region, each pixel inside the 3D sub-region is filled. When displaying the 2D sub-region, each pixel inside the 2D sub-region is filled.

[0130] For example, the target vehicle edge point set includes a first vehicle edge point set and a second vehicle edge point set. When the acquisition module 61 acquires the target vehicle edge point set, it specifically performs the following: for any trajectory point of the vehicle's driving trajectory, it determines a first candidate point corresponding to the vehicle being at that trajectory point. The first candidate point includes two left corner points and two right corner points; wherein, the first candidate point further includes a left intersection point and a right intersection point, wherein the left intersection point is the intersection of the line connecting the two left corner points and the line connecting the two left corner points of the historical trajectory point, and the right intersection point is the intersection of the line connecting the two right corner points and the line connecting the two right corner points of the historical trajectory point. The intersection points between lines; filtering the first candidate points corresponding to all trajectory points to obtain second candidate points; wherein, for any first candidate point, if the first candidate point is within the rectangular range of the vehicle body corresponding to any trajectory point, then the first candidate point is not used as the second candidate point; otherwise, the first candidate point is used as the second candidate point; sorting the multiple second candidate points located on the left side of the vehicle body to obtain a first initial edge point set; generating the first vehicle body edge point set based on the first initial edge point set; sorting the multiple second candidate points located on the right side of the vehicle body to obtain a second initial edge point set; generating the second vehicle body edge point set based on the second initial edge point set.

[0131] For example, when the acquisition module 61 sorts multiple second candidate points located on the left side of the vehicle body to obtain a first initial edge point set, it is specifically used for: after obtaining the sorted current second candidate points, for each unsorted second candidate point, determining the loss value corresponding to the second candidate point based on the reference distance and reference angle corresponding to the second candidate point; wherein, the reference distance is the distance between the second candidate point and the current second candidate point, the reference angle is the angle between the first displacement vector and the second displacement vector, the first displacement vector is the displacement vector between the second candidate point and the current second candidate point, and the second displacement vector is the displacement vector between the current second candidate point and the previous second candidate point; based on the loss values ​​corresponding to each unsorted second candidate point, determining the second candidate point corresponding to the minimum loss value as the next second candidate point of the current second candidate point.

[0132] For example, when the acquisition module 61 determines the loss value corresponding to the second candidate point based on the reference distance and reference angle corresponding to the second candidate point, it specifically performs the following steps: determining the maximum and minimum reference distances based on the reference distances corresponding to each unsorted second candidate point; normalizing the reference distances corresponding to the second candidate point based on the maximum and minimum reference distances to obtain the normalized distance corresponding to the second candidate point; determining the penalty value corresponding to the second candidate point based on the reference angle corresponding to the second candidate point, wherein the penalty value is inversely proportional to the reference angle; determining the maximum and minimum penalty values ​​based on the penalty values ​​corresponding to each unsorted second candidate point; normalizing the penalty value corresponding to the second candidate point based on the maximum and minimum penalty values ​​to obtain the normalized penalty value corresponding to the second candidate point; and performing a weighted calculation on the normalized distance and the normalized penalty value corresponding to the second candidate point to obtain the loss value corresponding to the second candidate point.

[0133] For example, when the acquisition module 61 generates the first vehicle body edge point set based on the first initial edge point set, it is specifically used to: for two adjacent second candidate points in the first initial edge point set, if the distance between the two second candidate points is greater than a distance threshold, insert a new second candidate point between the two second candidate points to obtain an updated first initial edge point set; fit the updated first initial edge point set to obtain a first curve; sample N vehicle body edge points on the first curve to obtain a first target edge point set, the first target edge point set including the positions of the N vehicle body edge points in the vehicle body coordinate system; generate the first vehicle body edge point set based on the first target edge point set; wherein, the positions of each vehicle body edge point in the vehicle body coordinate system are converted into the first positions of the vehicle body edge points in the world coordinate system, the first vehicle body edge point set including the first positions of the N vehicle body edge points in the world coordinate system.

[0134] Based on the same concept as the above method, this application proposes an electronic device, see [link to previous application]. Figure 7 As shown, the electronic device includes a processor 71 and a machine-readable storage medium 72, the machine-readable storage medium 72 storing machine-executable instructions that can be executed by the processor 71; the processor 71 is used to execute the machine-executable instructions to implement the vehicle body area display method disclosed in the above example of this application.

[0135] Based on the same concept as the above method, this application also provides a machine-readable storage medium storing a plurality of computer instructions, which, when executed by a processor, can implement the vehicle body area display method disclosed in the above examples of this application.

[0136] The aforementioned machine-readable storage medium can be any electronic, magnetic, optical, or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, machine-readable storage media can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard disk drives), solid-state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or combinations thereof.

[0137] Based on the same concept as the methods described above, this application also provides a computer program product, which may include a computer program. When executed by a processor, the computer program implements the vehicle body area display method disclosed in the examples above.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for displaying vehicle body areas, characterized in that, The method includes: Obtain the target vehicle body edge point set; wherein, the acquired vehicle driving trajectory includes multiple trajectory points, and the target vehicle body edge point set includes the vehicle body edge points corresponding to the multiple trajectory points; Based on the target vehicle body edge point set, a 3D vehicle body coverage area is displayed in the target 3D image; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area includes multiple 3D sub-regions, and each 3D sub-region includes a region composed of the second positions of at least three vehicle body edge points. And / or, display a 2D vehicle body coverage area in the target 2D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into the third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area includes multiple 2D sub-regions, and the 2D sub-regions include a region composed of the third positions of at least three vehicle body edge points.

2. The method according to claim 1, characterized in that, The first coordinate system includes a clipping space coordinate system. The step of converting the first position of each vehicle edge point in the world coordinate system within the target vehicle edge point set into a second position in the first coordinate system includes: for any vehicle edge point, based on the acquired viewpoint matrix and perspective projection matrix, converting the first position of the vehicle edge point in the world coordinate system into a second position in the clipping space coordinate system; wherein, the viewpoint matrix is ​​determined based on the camera's right direction vector, the camera's up direction vector, the camera's viewing direction vector, and the camera's translation component; wherein, the perspective projection matrix is ​​determined based on the distances between the near clipping plane and the camera, the right clipping plane and the camera, the left clipping plane and the camera, the top clipping plane and the camera, the bottom clipping plane and the camera, and the far clipping plane and the camera; The second coordinate system includes the screen coordinate system. The step of converting the first position of each vehicle edge point in the world coordinate system in the target vehicle edge point set into the third position of the vehicle edge point in the second coordinate system includes: for any vehicle edge point, based on the acquired camera intrinsic and extrinsic parameters, converting the first position of the vehicle edge point in the world coordinate system into the distortion-free position of the vehicle edge point in the pixel coordinate system; converting the distortion-free position into a distorted position based on the acquired distortion coefficients; and converting the distorted position into the third position in the screen coordinate system based on the configured expected parameters.

3. The method according to claim 1, characterized in that, The target vehicle body edge point set includes a first vehicle body edge point set and a second vehicle body edge point set. The first vehicle body edge point set includes multiple left vehicle body edge points, and the second vehicle body edge point set includes multiple right vehicle body edge points. The 3D sub-region includes a region composed of the second positions of three vehicle body edge points, and the 2D sub-region includes a region composed of the third positions of three vehicle body edge points. The three vehicle body edge points include two left vehicle body edge points and one right vehicle body edge point, or one left vehicle body edge point and two right vehicle body edge points; When displaying the 3D sub-region, each pixel inside the 3D sub-region is filled; when displaying the 2D sub-region, each pixel inside the 2D sub-region is filled.

4. The method according to claim 1, characterized in that, The target vehicle edge point set includes a first vehicle edge point set and a second vehicle edge point set. Obtaining the target vehicle edge point set includes: For any trajectory point of the vehicle's driving trajectory, a first candidate point corresponding to the vehicle being at that trajectory point is determined. The first candidate point includes two left corner points and two right corner points. The first candidate point also includes a left intersection point and a right intersection point. The left intersection point is the intersection of the line connecting the two left corner points and the line connecting the two left corner points of the historical trajectory point. The right intersection point is the intersection of the line connecting the two right corner points and the line connecting the two right corner points of the historical trajectory point. The first candidate points corresponding to all trajectory points are filtered to obtain the second candidate points; wherein, for any first candidate point, if the first candidate point is within the rectangular range of the vehicle body corresponding to any trajectory point, the first candidate point is not used as the second candidate point; otherwise, the first candidate point is used as the second candidate point. The multiple second candidate points located on the left side of the vehicle body are sorted to obtain a first initial edge point set; the first vehicle body edge point set is generated based on the first initial edge point set. The multiple second candidate points located on the right side of the vehicle body are sorted to obtain a second initial edge point set; the second vehicle body edge point set is generated based on the second initial edge point set.

5. The method according to claim 4, characterized in that, The process of sorting multiple second candidate points located on the left side of the vehicle body to obtain a first initial edge point set includes: After obtaining the sorted current second candidate points, for each unsorted second candidate point, the loss value corresponding to the second candidate point is determined based on the reference distance and reference angle corresponding to the second candidate point; wherein, the reference distance is the distance between the second candidate point and the current second candidate point, and the reference angle is the angle between the first displacement vector and the second displacement vector, the first displacement vector is the displacement vector between the second candidate point and the current second candidate point, and the second displacement vector is the displacement vector between the current second candidate point and the previous second candidate point; Based on the loss values ​​corresponding to each unsorted second candidate point, the second candidate point corresponding to the minimum loss value is determined as the next second candidate point after the current second candidate point.

6. The method according to claim 5, characterized in that, The step of determining the loss value corresponding to the second candidate point based on the reference distance and reference angle corresponding to the second candidate point includes: Based on the reference distances corresponding to each unsorted second candidate point, determine the maximum reference distance and the minimum reference distance; normalize the reference distances corresponding to the second candidate point based on the maximum reference distance and the minimum reference distance to obtain the normalized distance corresponding to the second candidate point. The penalty value corresponding to the second candidate point is determined based on the reference angle corresponding to the second candidate point, and the penalty value is inversely proportional to the reference angle; the maximum penalty value and the minimum penalty value are determined based on the penalty values ​​corresponding to each unsorted second candidate point; the penalty value corresponding to the second candidate point is normalized based on the maximum penalty value and the minimum penalty value to obtain the normalized penalty value corresponding to the second candidate point. The loss value corresponding to the second candidate point is obtained by weighting the normalized distance and the normalized penalty value corresponding to the second candidate point.

7. The method according to claim 4, characterized in that, The step of generating the first vehicle body edge point set based on the first initial edge point set includes: For any two adjacent second candidate points within the first initial edge point set, if the distance between the two second candidate points is greater than a distance threshold, a new second candidate point is inserted between the two second candidate points to obtain an updated first initial edge point set; a first curve is obtained by fitting based on the updated first initial edge point set; N vehicle edge points are sampled on the first curve to obtain a first target edge point set, the first target edge point set including the positions of the N vehicle edge points in the vehicle coordinate system; The first vehicle body edge point set is generated based on the first target edge point set; wherein, the position of each vehicle body edge point in the vehicle body coordinate system is converted into the first position of the vehicle body edge point in the world coordinate system, and the first vehicle body edge point set includes the first positions of the N vehicle body edge points in the world coordinate system.

8. A vehicle body area display device, characterized in that, The device includes: The acquisition module is used to acquire a set of target vehicle body edge points; wherein, the acquired vehicle driving trajectory includes multiple trajectory points, and the set of target vehicle body edge points includes vehicle body edge points corresponding to the multiple trajectory points; The display module is used to display a 3D vehicle body coverage area in a target 3D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a second position of the vehicle body edge point in the first coordinate system, and the 3D vehicle body coverage area includes multiple 3D sub-regions, each 3D sub-region including an area composed of at least three vehicle body edge points at their second positions; and / or, to display a 2D vehicle body coverage area in a target 2D image based on the target vehicle body edge point set; wherein, the first position of each vehicle body edge point in the target vehicle body edge point set in the world coordinate system is converted into a third position of the vehicle body edge point in the second coordinate system, and the 2D vehicle body coverage area includes multiple 2D sub-regions, each 2D sub-region including an area composed of at least three vehicle body edge points at their third positions.

9. The apparatus according to claim 8, characterized in that, The first coordinate system includes a clipping space coordinate system. When the display module converts the first position of each vehicle edge point in the world coordinate system to a second position in the first coordinate system, it specifically performs the following: for any vehicle edge point, based on the acquired viewpoint matrix and perspective projection matrix, converts the first position of the vehicle edge point in the world coordinate system to a second position in the clipping space coordinate system. The viewpoint matrix is ​​determined based on the camera's right direction vector, the camera's up direction vector, the camera's viewing direction vector, and the camera's translation component. The perspective projection matrix is ​​determined based on the distances between the near clipping plane and the camera, the right clipping plane and the camera, the left clipping plane and the camera, the top clipping plane and the camera, the bottom clipping plane and the camera, and the far clipping plane and the camera. The second coordinate system includes the screen coordinate system. When the display module converts the first position of each vehicle edge point in the world coordinate system to the third position of the vehicle edge point in the second coordinate system, it specifically performs the following steps: for any vehicle edge point, based on the acquired camera intrinsic and extrinsic parameters, converts the first position of the vehicle edge point in the world coordinate system to a distortion-free position in the pixel coordinate system; converts the distortion-free position to a distorted position based on the acquired distortion coefficients; and converts the distorted position to a third position in the screen coordinate system based on the configured desired parameters. Alternatively, the target vehicle body edge point set includes a first vehicle body edge point set and a second vehicle body edge point set. The first vehicle body edge point set includes multiple left-side vehicle body edge points, and the second vehicle body edge point set includes multiple right-side vehicle body edge points. The 3D sub-region includes a region composed of the second positions of three vehicle body edge points, and the 2D sub-region includes a region composed of the third positions of three vehicle body edge points. The three vehicle body edge points include two left-side vehicle body edge points and one right-side vehicle body edge point, or one left-side vehicle body edge point and two right-side vehicle body edge points. When displaying the 3D sub-region, each pixel inside the 3D sub-region is filled; when displaying the 2D sub-region, each pixel inside the 2D sub-region is filled. Alternatively, the target vehicle edge point set includes a first vehicle edge point set and a second vehicle edge point set. When the acquisition module acquires the target vehicle edge point set, it specifically performs the following: for any trajectory point of the vehicle's driving trajectory, it determines a first candidate point corresponding to the vehicle being at that trajectory point. The first candidate point includes two left corner points and two right corner points. The first candidate point further includes a left intersection point and a right intersection point. The left intersection point is the intersection of the line connecting the two left corner points and the line connecting the two left corner points of the historical trajectory point. The right intersection point is the intersection of the line connecting the two right corner points and the line connecting the two right corner points of the historical trajectory point. The intersection points between the trajectory points are identified; the first candidate points corresponding to all trajectory points are filtered to obtain second candidate points; wherein, for any first candidate point, if the first candidate point is within the rectangular range of the vehicle body corresponding to any trajectory point, then the first candidate point is not considered as the second candidate point; otherwise, the first candidate point is considered as the second candidate point; the multiple second candidate points located on the left side of the vehicle body are sorted to obtain a first initial edge point set; the first vehicle body edge point set is generated based on the first initial edge point set; the multiple second candidate points located on the right side of the vehicle body are sorted to obtain a second initial edge point set; the second vehicle body edge point set is generated based on the second initial edge point set; Alternatively, when the acquisition module sorts multiple second candidate points located on the left side of the vehicle to obtain a first initial edge point set, it specifically performs the following steps: After obtaining the sorted current second candidate points, for each unsorted second candidate point, based on the reference distance and reference angle corresponding to that second candidate point, it determines the loss value corresponding to that second candidate point; wherein, the reference distance is the distance between the second candidate point and the current second candidate point, and the reference angle is the angle between a first displacement vector and a second displacement vector, the first displacement vector is the displacement vector between the second candidate point and the current second candidate point, and the second displacement vector is the displacement vector between the current second candidate point and the previous second candidate point; based on the loss values ​​corresponding to each unsorted second candidate point, the second candidate point corresponding to the minimum loss value is determined as the next second candidate point after the current second candidate point; Alternatively, when the acquisition module determines the loss value corresponding to the second candidate point based on the reference distance and reference angle corresponding to the second candidate point, it specifically performs the following steps: Based on the reference distances corresponding to each unsorted second candidate point, determine the maximum and minimum reference distances; normalize the reference distances corresponding to the second candidate point based on the maximum and minimum reference distances to obtain the normalized distance corresponding to the second candidate point; determine the penalty value corresponding to the second candidate point based on the reference angle corresponding to the second candidate point, where the penalty value is inversely proportional to the reference angle; based on the penalty values ​​corresponding to each unsorted second candidate point, determine the maximum and minimum penalty values; normalize the penalty value corresponding to the second candidate point based on the maximum and minimum penalty values ​​to obtain the normalized penalty value corresponding to the second candidate point; and perform a weighted calculation on the normalized distance and the normalized penalty value corresponding to the second candidate point to obtain the loss value corresponding to the second candidate point. Alternatively, when the acquisition module generates the first vehicle body edge point set based on the first initial edge point set, it is specifically used for: for two adjacent second candidate points in the first initial edge point set, if the distance between the two second candidate points is greater than a distance threshold, inserting a new second candidate point between the two second candidate points to obtain an updated first initial edge point set; fitting a first curve based on the updated first initial edge point set; sampling N vehicle body edge points on the first curve to obtain a first target edge point set, the first target edge point set including the positions of the N vehicle body edge points in the vehicle body coordinate system; generating the first vehicle body edge point set based on the first target edge point set; wherein, the positions of each vehicle body edge point in the vehicle body coordinate system are converted into the first positions of the vehicle body edge points in the world coordinate system, the first vehicle body edge point set including the first positions of the N vehicle body edge points in the world coordinate system.

10. An electronic device, characterized in that, include: A processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions that can be executed by the processor; The processor is configured to execute machine-executable instructions to implement the method of any one of claims 1-7.

Images