A method, device, storage medium and program product for determining a vehicle driving profile

Abstract

The one or more embodiments of the specification provide a method, device, storage medium and program product for determining a driving contour of a vehicle, the method comprising: obtaining an expected driving trajectory of the vehicle and a plurality of position points on the expected driving trajectory; for each position point, determining a plurality of feature points on a vehicle contour corresponding to the position point based on an expected pose of the vehicle at the position point, to obtain a feature point set composed of the plurality of feature points corresponding to each position point; obtaining a side point set corresponding to the vehicle contour at a starting point of the trajectory and at an ending point of the trajectory respectively; and determining a boundary of a space occupied by the vehicle along the expected driving trajectory as a whole, as the driving contour of the vehicle, according to the feature point set and the side point set, thereby significantly reducing the computational overhead while ensuring the integrity of the contour, and meeting the dual requirements of real-time performance and safety for intelligent driving.

Description

Technical Field

[0001] This specification relates to the field of vehicle technology, and more particularly to a method, apparatus, storage medium, and program product for determining a vehicle driving profile. Background Technology

[0002] In recent years, intelligent driving technology has been moving from theoretical research to engineering mass production and commercialization. Mid-to-high-level functions such as urban navigation assistance, traffic jam assistance, and memory parking have become core competitive advantages in the industry. However, these functions need to address unstructured road scenarios such as towns, industrial parks, and parking lots, which are generally characterized by narrow road widths, lack of clear road markings, irregular boundary shapes, and highly dynamic environments. Vehicles in such scenarios often need to slow down or brake suddenly, and their anthropomorphic performance highly depends on the accurate calculation of detailed driving contours and the accuracy of collision detection. Therefore, how to accurately match the vehicle's actual potential occupied space, and thus provide high-precision boundary data for collision detection in drivable areas, has become a pressing technical problem that needs to be solved in this field.

[0003] In related technologies, there are generally two types of methods for determining the vehicle's driving profile. One type is the lateral translation method based on the planned driving path, which generates an approximate profile by laterally translating the planned driving path by a fixed distance. While logically simple, this method has coarse accuracy and cannot adapt to changes in the vehicle's pose, such as corner expansion during turns. Furthermore, the static offset is difficult to balance safety and spatial efficiency, potentially leading to collision risks or excessive conservatism. The other type is the dense sampling virtual rectangle overlay method, which involves densely sampling along the driving path and calculating the boundaries of virtual rectangles at each sampling point. The driving profile is approximated by overlaying all the rectangle boundaries. However, this method is prone to collision blind spots when sampling is sparse, and the large number of redundant edges involved in the calculation leads to wasted computational power, making it difficult to meet the real-time requirements of intelligent driving. Summary of the Invention

[0004] In view of the above, one or more embodiments of this specification provide the following technical solutions: According to a first aspect of one or more embodiments of this specification, a method for determining a vehicle driving profile is provided, comprising: Obtain the vehicle's expected driving trajectory and multiple location points on the expected driving trajectory, wherein the multiple location points include at least the trajectory start point and trajectory end point of the expected driving trajectory; For each location point, based on the expected pose of the vehicle at that location point, multiple feature points on the vehicle contour corresponding to that location point are determined to obtain a feature point set composed of multiple feature points corresponding to each location point, wherein the vehicle contour is used to characterize the boundary of the space occupied by the vehicle at the corresponding location point. Obtain the set of side points corresponding to the vehicle contours at the starting point and ending point of the trajectory, respectively; Based on the feature point set and the side point set, the boundary of the space occupied by the vehicle along the entire expected driving trajectory is determined, which serves as the driving profile of the vehicle.

[0005] According to a second aspect of one or more embodiments of this specification, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor implements the steps of the method described in the first aspect by executing the executable instructions.

[0006] According to a third aspect of this specification, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0007] According to a fourth aspect of this specification, a computer program product is provided, comprising a computer program / instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0008] As can be seen from the above embodiments, this specification first generates feature points on the corresponding vehicle contour at each location point based on the vehicle's expected pose. This allows the feature point set to accurately depict the actual occupancy shape of the vehicle at each location throughout the entire driving process, thereby significantly improving the accuracy of the driving contour and avoiding the rough contour accuracy defect of the traditional lateral translation method for planning paths. At the same time, this solution only needs to obtain the side point set at the trajectory start point and end point, and combine it with the feature point set of each location point to construct a complete boundary. There is no need to densely superimpose the virtual rectangles of all sampling points, which avoids the collision blind spot caused by sparse sampling and eliminates redundant calculations of a large number of invalid boundaries. While ensuring the integrity of the contour, it significantly reduces the computational overhead and meets the dual requirements of real-time performance and safety for intelligent driving. Attached Figure Description

[0009] Figure 1 This is a schematic flowchart illustrating a method for determining a vehicle driving profile according to an embodiment disclosed in this specification; Figure 2 This is a schematic diagram illustrating a location point sampling scenario for a driving trajectory, as shown in the embodiments disclosed in this specification. Figure 3 This is a schematic diagram illustrating a vehicle profile based on various points in a driving trajectory, as shown in the embodiments disclosed in this specification. Figure 4 This is a schematic diagram illustrating a feature point set according to an embodiment disclosed in this specification; Figure 5 This is a schematic diagram illustrating a side point set according to an embodiment disclosed in this specification; Figure 6This is a schematic diagram illustrating a boundary point set according to an embodiment disclosed in this specification; Figure 7 This is a schematic diagram illustrating the driving profile of a vehicle according to the embodiments disclosed in this specification; Figure 8 This is a schematic flowchart illustrating another method for determining the vehicle driving profile as shown in the embodiments disclosed in this specification; Figure 9 This is a schematic structural diagram of an electronic device shown in the embodiments of this specification; Figure 10 This is a schematic diagram of a vehicle travel profile determination device shown in an embodiment of this specification. Detailed Implementation

[0010] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this manual are all information and data authorized by the user or fully authorized by all parties. The collection, use and processing of related data shall comply with the relevant laws, regulations and standards of the relevant regions, and corresponding operation portals shall be provided for users to choose to authorize or refuse.

[0011] Figure 1 This is a schematic flowchart illustrating an exemplary embodiment of a method for determining a vehicle's driving profile as shown in this specification. Figure 1 As shown, the method may include the following steps: Step S102: Obtain the vehicle's expected driving trajectory and multiple location points on the expected driving trajectory, wherein the multiple location points include at least the trajectory start point and trajectory end point of the expected driving trajectory.

[0012] First, a contour determination system can acquire the vehicle's expected driving trajectory and determine multiple location points along that trajectory. These location points include at least the starting and ending points of the trajectory, ensuring coverage of the complete driving range from the starting point to the ending point. Simultaneously, this expected driving trajectory can originate from the path planning module of the intelligent driving system corresponding to the vehicle, or from the vehicle's own navigation system, user-defined driving paths, or other higher-level decision-making modules. It should be noted that the contour determination system described in this specification can be deployed on the vehicle's local computing unit, or, in scenarios with network communication capabilities, defined as being deployed on a cloud server connected to the vehicle, with the cloud completing trajectory acquisition and location point sampling before transmitting the results back to the vehicle. This specification does not impose any limitations on this approach.

[0013] After obtaining the expected driving trajectory, the above-mentioned contour determination system can determine multiple discrete location points on the trajectory so as to calculate the vehicle contour at each point in the subsequent calculation.

[0014] In one embodiment, the aforementioned multiple location points can be obtained by sampling the expected driving trajectory at equal intervals. Equal-interval sampling refers to selecting location points along the driving direction of the expected trajectory, i.e., according to the cumulative arc length of the trajectory from the starting point, at fixed intervals, such as every 0.2 meters, 0.5 meters, or 1 meter. The specific interval can be dynamically adjusted according to vehicle speed, turning radius, or accuracy requirements. Equal-interval sampling ensures that the sampling points are uniformly distributed on the trajectory, avoiding uneven sampling density due to curvature changes, thereby providing a stable and consistent spatial reference for subsequent contour calculations. Figure 2 For example, the multiple triangles in the figure located in the preset coordinate system can be regarded as position points on the curve represented by the expected driving trajectory, which are sampled at equal intervals along the path length.

[0015] It should be noted that the aforementioned location points can be used to characterize the spatial position of the vehicle on the trajectory. Specifically, they can correspond to a fixed reference point on the vehicle, such as the rear axle center, rear axle center, rear bumper midpoint, center of mass, or vehicle geometric center. This specification does not restrict the specific location of the location points, as long as the reference point has a definite geometric relationship with the vehicle's pose. For the convenience of subsequent discussion in this specification, the rear axle center of the vehicle is uniformly used as the representative location point in the following embodiments. That is, the rear axle center is placed at each sampling point, and the aforementioned vehicle contour is calculated based on the coordinates of the rear axle center and the vehicle's heading angle.

[0016] In summary, through the aforementioned equidistant sampling, the contour determination system can obtain a series of location points that at least include the starting and ending points of the trajectory, thus covering the complete distance from the starting point to the ending point of the vehicle. Subsequent steps will calculate the feature points and side point sets on the vehicle contour based on these location points, and finally synthesize the vehicle's driving contour.

[0017] Step S104: For each location point, based on the expected pose of the vehicle at that location point, determine multiple feature points on the vehicle contour corresponding to that location point to obtain a feature point set composed of multiple feature points corresponding to each location point, wherein the vehicle contour is used to characterize the boundary of the space occupied by the vehicle at the corresponding location point.

[0018] The system described above can determine multiple feature points on the vehicle profile corresponding to each location point identified above, based on the vehicle's expected pose at that location point. The expected pose can include coordinates and heading angle. The vehicle profile represents the spatial boundary occupied by the vehicle at that location point, and the feature points represent geometrically significant key positions on that vehicle profile. In summary, the system can aggregate the feature points corresponding to each location point to form a feature point set, which reflects the spatial occupancy of the vehicle at various times as it travels along the trajectory.

[0019] The specific shape of the vehicle outline can be achieved using different geometric models depending on actual needs, while the selection method of feature points is closely related to the shape type of the vehicle outline.

[0020] In one embodiment, the vehicle outline can be represented by a polygon with preset geometric parameters, such as a rectangle reflecting the vehicle's outer contour. In this polygonal model, the aforementioned feature points are at least some of the corner points of the polygon. For example, when using a rectangular outline, the feature points can be all or some of the four corner points, such as the front left, rear left, front right, and rear right corner points. These corner points are the critical areas where the vehicle is most likely to collide with external obstacles when turning. Using corner points as feature points can significantly reduce the computational load while ensuring accuracy.

[0021] In another embodiment, the vehicle profile can also be a non-polygonal shape, such as an ellipse, a rounded rectangle, or a closed shape enclosed by an irregular curve, to more accurately fit the actual vehicle outline. In this non-polygonal model, the aforementioned feature points are at least a portion of the boundary points of the protruding parts of the non-polygonal outer edge. A protruding part refers to an area on the vehicle profile that has a large radial distance relative to the vehicle center, such as the endpoint of the major axis of an ellipse or the arc vertex of a rounded rectangle. These areas are also the most prone to interference when the vehicle changes direction or passes through narrow passages. Using the boundary points of these protruding parts as feature points can balance profile accuracy and computational efficiency.

[0022] To facilitate the explanation of the core concepts of this specification, the following embodiments uniformly use a rectangular outline as an example of the vehicle outline, and use the four corner points of the rectangle, namely the aforementioned front left, rear left, front right, and rear right, as feature points. For details, please refer to [link / reference needed]. Figure 3 It consists of the expected driving trajectory, multiple position points, and a rectangular vehicle outline corresponding to each position point. The vehicle outline represented by the black rectangle can be used to simulate the posture of the vehicle when it travels to the corresponding position point, providing a pose reference for the subsequent calculation of feature point set and side point set.

[0023] Of course, regardless of whether a polygonal or non-polygonal contour is used, the feature points are determined based on the geometric shape of the vehicle contour at the corresponding location point, and each feature point is associated with the vehicle's pose at that location point. In other words, when the location point changes, for example, when the vehicle moves along the trajectory sampling point, its vehicle contour translates and rotates with the pose, and each of its original feature points also changes synchronously, thus forming a continuous feature point trajectory, which is the aforementioned feature point set. In this way, the feature point set can accurately depict the actual occupied boundary of the vehicle at each moment throughout the entire driving process, laying the foundation for subsequent fusion with the side point set.

[0024] Based on the above description, in the embodiments of this specification for obtaining a feature point set, the feature points can be exemplarily represented as the four corner points of a rectangle. See details... Figure 4 For each sampling point (position point) on the expected driving trajectory, the rear axle center of the vehicle is placed at that sampling point, and the coordinates of the four corner points of the rectangular contour are calculated based on the vehicle's pose at that point: Left Front (LF), Left Back (LB), Right Front (RF), and Right Back (RB). Here, "left" and "right" are relative to the vehicle's driving direction, while "front" and "back" correspond to the vehicle's forward and backward directions.

[0025] After traversing all sampling points, the set of coordinates of corner points of the same type at different locations constitutes the point set of that corner point. Specifically, the LF points of all sampling points constitute the left front corner point set, the LB points constitute the left rear corner point set, the RF points constitute the right front corner point set, and the RB points constitute the right rear corner point set. Figure 4 The curves swept by these four sets of corner points are schematically depicted using different line types and arrows: the two thick solid lines represent the LF and LB point sets respectively, and the two dashed lines represent the RF and RB point sets respectively.

[0026] Those skilled in the art will understand that these four sets of corner points can be considered as a concrete implementation of the aforementioned "feature point set" within a rectangular outline—each set corresponds to the same feature on the vehicle outline, i.e., the trajectory of the same corner point across the entire path. In this way, the originally abstract feature point set is visualized as four clear corner point trajectory curves, providing a precise geometric basis for subsequent fusion with the side point set and lateral optimization processing.

[0027] Step S106: Obtain the side point sets corresponding to the vehicle contours at the starting point and ending point of the trajectory, respectively.

[0028] The system described above can also obtain the side point sets corresponding to the vehicle contour at the starting point and ending point of the trajectory, respectively. The side point set refers to the set of points on the left and right sides of the vehicle contour. These points can fill in any boundary gaps that might arise from relying solely on feature points between the starting and ending points, ensuring the continuity of the contour in the vehicle's direction of travel. It should be noted that this step is not sequential with the aforementioned step of determining the feature point set; they can be executed sequentially or in parallel, and this specification does not impose any restrictions on this.

[0029] In one embodiment, the aforementioned side point set may specifically include four sets: a first left side point set and a first right side point set of the vehicle contour at the trajectory starting point, and a second left side point set and a second right side point set of the vehicle contour at the trajectory ending point. Here, the terms "left side" and "right side" are relative to the vehicle's direction of travel, referring respectively to the line segments on the left and right boundaries of the vehicle contour. The first left side point set and the first right side point set represent the points on the left and right sides of the vehicle contour at the starting point pose, respectively; the second left side point set and the second right side point set represent the points on the left and right sides of the vehicle contour at the ending point pose, respectively.

[0030] by Figure 5 For example, the figure schematically depicts four sets of side point sets using different line types and arrows: the left side point set (Left Current, LC) and the right side point set (Right Current, RC) at the starting point of the trajectory, and the left side point set (Left End, LE) and the right side point set (Right End, RE) at the ending point of the trajectory. These side point sets are represented by different line types, namely two bold solid lines LE and LC, and two dashed lines RE and RC, to distinguish them from the subsequent corner point sets. By obtaining these four sets of side point sets and combining them with the aforementioned feature point sets, complete boundary data can be provided for the subsequent union fusion operation.

[0031] It is important to emphasize that the side point set is substantially different from the aforementioned feature point set, such as the corner point set: the feature point set describes the motion trajectory of specific discrete parts of the vehicle contour, such as corner points, on the entire trajectory, while the side point set only involves the complete side line segments at two specific locations: the starting point and the ending point. It should be noted that the reason only the side lines of the starting and ending points are selected in this specification is that in the middle section of the driving trajectory, the left and right boundaries of the vehicle contour are naturally covered by the corresponding parts in the feature point set, such as the line connecting the front and rear corner points on the left side, eliminating the need to repeatedly calculate the side lines of all sampled points; while at the starting and ending points, due to the lack of connecting feature points in front or behind, the side line segments can be used as additional separate supplements, thus forming a closed driving contour without gaps.

[0032] Step S108: Based on the feature point set and the side point set, determine the boundary of the space occupied by the vehicle along the entire expected driving trajectory, so as to serve as the driving profile of the vehicle.

[0033] Finally, based on the aforementioned feature point set and the side point sets at the trajectory start and end points, the system determines the boundary of the space occupied by the vehicle along the entire expected driving trajectory, and uses this boundary as the vehicle's driving profile. By fusing the feature point set and the side point set, the system characterizes the key occupied areas of the vehicle at each location and provides continuous boundaries at the beginning and end of the expected driving trajectory. Thus, the system constructs a complete, continuous, and gap-free driving profile, providing high-precision boundary information for subsequent drivable area collision detection and control decisions such as deceleration and braking.

[0034] At this point, the system can group and merge points according to geometric correspondence, rather than simply merging all feature points and all side points into a single point set. Specifically, the feature point set can be divided into multiple feature point subsets based on the different locations of the feature points on the vehicle contour. For example, in the rectangular contour embodiment described above, it is divided into the left front corner point subset LF, the left rear corner point subset LB, the right front corner point subset RF, and the right rear corner point subset RB. Simultaneously, the side point sets are also divided into four groups according to the position of the "starting point or ending point" and the lateral direction (left or right): the starting point left side point set LC, the starting point right side point set RC, the ending point left side point set LE, and the ending point right side point set RE.

[0035] The so-called union operation refers to merging a subset of feature points with corresponding relationships with a set of side points to form a new set of boundary points. The correspondence here is based on the pre-determined location of the feature points on the vehicle contour and the position and lateral orientation of the side point sets: feature points on the front left (e.g., LF) correspond to the left side of the starting point (LC), because both are located on the left side of the vehicle and close to the front; feature points on the rear left (e.g., LB) correspond to the left side of the ending point (LE), both are located on the left side and close to the rear; similarly, feature points on the front right (e.g., RF) correspond to the right side of the starting point (RC), and feature points on the rear right (e.g., RB) correspond to the right side of the ending point (RE). By unioning these corresponding point sets, four sets of boundary points are obtained: {LF, LC}, {LB, LE}, {RF, RC}, and {RB, RE}.

[0036] For details, please refer to Figure 6The figure illustrates these four sets of boundary points using different line types and arrows: the bold solid lines represent {LF, LC} and {LB, LE}, and the dashed lines represent {RF, RC} and {RB, RE}. It can be seen that the union operation organically merges the corner point trajectories, i.e., the aforementioned feature point sets, with the side line segments of the start and end points. The corner point trajectories cover the movement areas of the four corners most prone to collision during vehicle movement, while the side segments of the start and end points fill in the boundary gaps caused by the lack of adjacent corner points at the beginning and end. Through this fusion, the system obtains a complete and uninterrupted boundary foundation, providing an accurate data source for subsequent processing such as position alignment and lateral optimization.

[0037] It should be noted that this step only completes the union and fusion of point sets, resulting in multiple sets of boundary points. How to ultimately determine the driving profile based on these multiple sets of boundary points will be discussed in the following section. Figure 7 The following detailed descriptions are provided in subsequent embodiments.

[0038] First, this specification introduces a preset coordinate system to facilitate subsequent processing of the aforementioned sets of boundary points. Within this coordinate system, two mutually perpendicular directions can be defined: a first direction and a second direction.

[0039] In one embodiment, the aforementioned preset coordinate system is the SL coordinate system, also known as the natural coordinate system or the Frenet coordinate system. In the SL coordinate system, the first direction is the direction of travel along the expected travel trajectory, which can be represented by the vertical coordinate S, representing the arc length measured from the starting point of the trajectory. The second direction is perpendicular to the direction of travel, which can be represented by the horizontal coordinate L, representing the lateral distance from the center line of the trajectory. In other words, the first direction is used to characterize the position of the vehicle along the trajectory, and the second direction is used to characterize the lateral offset of the vehicle's profile relative to the trajectory. The SL coordinate system will be used by default in the subsequent embodiments of this specification for ease of explanation. Of course, similar functions can be achieved even with other coordinate systems. For example, when the preset coordinate system is a Cartesian coordinate system, the aforementioned first direction is the tangent direction of the trajectory, and the aforementioned second direction is the normal direction. This specification does not impose any restrictions on this.

[0040] Based on the aforementioned preset coordinate system, the system can first divide multiple sets of boundary points into a left boundary point set and a right boundary point set according to the lateral direction corresponding to each set of boundary points. Still using... Figure 6Taking a rectangular outline as an example, the left boundary point set includes two groups: {LF, LC} and {LB, LE}, while the right boundary point set includes two groups: {RF, RC} and {RB, RE}. The two left point sets together characterize the potential occupancy boundary on the left side of the vehicle, and the two right point sets together characterize the potential occupancy boundary on the right side of the vehicle. Since the processing logic for the left and right sides is exactly the same, the following embodiment will use the left boundary point set as an example for detailed explanation, and the right side can be deduced similarly.

[0041] Specifically, the boundary point sets in the aforementioned left boundary point set group can be aligned along the first direction. Position alignment means that points in different boundary point sets have the same sampling position in the first direction, i.e., the coordinate of the driving direction S, so as to compare the second direction offset of each point set at the same S coordinate.

[0042] In one embodiment, the alignment process can be achieved using equidistant interpolation. The specific steps are as follows: First, determine the minimum and maximum coordinate values ​​of each set of boundary points on the left in the first direction (e.g., for the {LF, LC} point set, calculate its minimum S-coordinate value S_LFLC_min and maximum S_LFLC_max; for the {LB, LE} point set, calculate S_LBLE_min and S_LBLE_max). Then, determine the coordinate interval for the entire left side, extending from the minimum left coordinate value SL_min to the maximum left coordinate value SL_max. Next, perform equidistant dense interpolation within this interval at fixed intervals, such as 0.1 meters or 0.2 meters, so that both sets of boundary points on the left have the same S-coordinate sequence after interpolation. This operation yields two sets of interpolated point sets: the front left interpolated point set (corresponding to {LF, LC}, denoted as LFC) and the rear left interpolated point set (corresponding to {LB, LE}, denoted as LBE). These two sets of points are perfectly aligned in the first direction—that is, for the same S-coordinate value, there exists a corresponding point in both LFC and LBE.

[0043] Similarly, the same interpolation operation is performed on the right boundary point set to obtain the right front interpolation point set (corresponding to {RF,RC}, denoted as RFC) and the right rear interpolation point set (corresponding to {RB, RE}, denoted as RBE), which are also perfectly aligned in the first direction.

[0044] After completing the alignment, the system can compare the offsets in the second direction point by point for lateral optimization. For the left side, at each aligned S-coordinate position, the L-coordinate of that point in the LFC is compared with the L-coordinate of the same point in the LBE. The point with the larger L-coordinate value—that is, the point further away from the centerline of the expected driving trajectory and the further outward—is selected. All selected points constitute the left contour point set LPS. From the results, this operation ensures that the left contour is always determined by the set of boundary points closer to the outside of the two sets, thus accurately covering the leftmost limit area that the vehicle may actually occupy. For the right side, at each S-coordinate position, the L-coordinates of the corresponding points in the RFC and RBE are compared. However, note that the offset on the right side is usually negative or positive, and a unified sign convention is needed. In this specification, "larger offset" or "optimization" refers to a larger numerical value, that is, further to the right and further away from the trajectory centerline. For ease of understanding, the absolute value of the L-coordinate can be used uniformly, or the scalar value can be directly compared. In short, as long as the point that is further outward and farther from the trajectory centerline is selected, it is acceptable. By comparison, the points with the wider outer edge are selected to form the right contour point set RPS.

[0045] Note that, provided the accuracy requirements are met, this specification may also employ other alternative methods to achieve the alignment and selection process described above. For example, the nearest neighbor matching method can be used to directly find the points with the closest vertical coordinates among the original sampled points of each boundary point set for lateral comparison; or curve fitting, such as B-splines or cubic splines, can be performed on each boundary point set first, and then resampled in a unified parameter space for comparison; or piecewise constant approximation can be used in scenarios where the contour accuracy requirements are not high, with the maximum lateral offset within the interval representing the overall offset of that interval.

[0046] Finally, the driving profile can be generated based on the left contour point set LPS and the right contour point set RPS. (See reference...) Figure 7 Connect the beginning and end points of LPS and RPS respectively to form a closed or non-closed boundary composed of multiple smooth line segments, usually two curves on the left and right. However, the beginning and end can also be set to be connected by the outline of the front bumper and rear bumper of the vehicle according to actual needs, but this is not the focus of this manual. Figure 7 The final generated driving profile is schematically depicted with bold solid lines. This profile fully covers all possible spatial boundaries that the vehicle may occupy during its journey from the starting point to the end point, including details such as the outward expansion of the corner point when turning and the extension of the front and rear of the vehicle.

[0047] Thus, through the aforementioned steps of grouping and unioning, position alignment, lateral optimization, and contour generation, this specification has achieved a high-precision, gap-free, and computationally efficient method for determining driving contours, providing reliable boundary basis for intelligent driving vehicles to make decisions such as deceleration and emergency braking on unstructured roads.

[0048] After determining the vehicle's driving profile through the above steps, the system can further utilize this driving profile to make driving decisions, thereby ensuring the vehicle's driving safety on unstructured roads.

[0049] Specifically, intelligent driving systems typically maintain a drivable space, which is the boundary of an unobstructed area around the vehicle where safe passage is possible. This boundary can be detected in real time by a perception module using sensors such as cameras and lidar. This application compares the generated driving profile with the boundary of this drivable space to determine if they overlap. If the driving profile overlaps with the boundary of the drivable space, it means that when the vehicle travels along the expected trajectory, its body or its potential occupied space will interfere with the boundary of the drivable space, i.e., there is a risk of collision or the vehicle is about to leave the drivable area. In response to the detection of this overlap, the system can generate corresponding deceleration or emergency braking control commands to reduce the vehicle speed or apply braking, thereby avoiding a collision or ensuring that the vehicle always operates within the drivable space.

[0050] By employing the above methods, this application directly applies high-precision driving profiles to real-time conflict detection and control decisions, significantly improving the safety and human-likeness of intelligent driving vehicles in unstructured narrow road scenarios.

[0051] The following is based on Figure 8 Taking this as an example, and combining it with the above-mentioned expected driving trajectory, we will introduce the specific process of another method for determining the vehicle's driving profile, such as... Figure 8 As shown, the method may include the following steps: Step S802: Obtain the expected driving trajectory.

[0052] In one embodiment, the contour determination system acquires the vehicle's expected driving trajectory. This expected driving trajectory may originate from the path planning module of an intelligent driving system, a navigation system, or user settings.

[0053] Step S804: Obtain location points by sampling at equal intervals.

[0054] In one embodiment, the system samples at equal intervals along the direction of travel of the expected trajectory and adds the trajectory endpoint to the set of sampling points, resulting in multiple location points. These multiple location points include at least the trajectory start point and the trajectory end point.

[0055] Step S806: Place the vehicle reference point at each sampling point.

[0056] In one embodiment, the system virtually places a preset reference point of the vehicle (e.g., the rear axle center) at each sampling point and combines it with the vehicle's expected pose (coordinates, heading angle) at that point to provide a pose reference for subsequent contour calculation.

[0057] Step S808: Generate the corner point set at each sampling point.

[0058] In one embodiment, for each sampling point, the system calculates the four corner points of the vehicle contour, which is a rectangle, to obtain the left front corner point set LF, the left rear corner point set LB, the right front corner point set RF, and the right rear corner point set RB. These corner point sets constitute the specific representation of the feature point set.

[0059] Step S810: Generate the set of side points at the starting and ending points.

[0060] In one embodiment, the system acquires the left side point set LC and the right side point set RC of the vehicle profile at the starting point of the trajectory, and the left side point set LE and the right side point set RE of the vehicle profile at the ending point of the trajectory.

[0061] Step S812: Grouping and union operation.

[0062] In one embodiment, the system performs a union operation on the corner point subset and the side point set that have corresponding relationships, resulting in four sets of boundary point sets: {LF, LC}, {LB, LE}, {RF, RC}, and {RB, RE}. The corresponding relationships are predetermined based on the location of the feature point (front left, back left, front right, back right) and the position (starting point or ending point) and lateral orientation (left or right) of the side point set.

[0063] Step S814: Calculate the vertical coordinate intervals on the left and right sides.

[0064] In one embodiment, the system calculates the minimum value SLmin and the maximum value SLmax of the two sets of boundary points on the left ({LF, LC} and {LB, LE}) in the first direction, i.e. the driving direction S coordinate, and the minimum value SRmin and the maximum value SRmax of the two sets of boundary points on the right ({RF, RC} and {RB, RE}).

[0065] Step S816: Achieve position alignment by using dense interpolation with equal spacing.

[0066] In one embodiment, the system performs equally spaced dense interpolation on the two sets of boundary points on the left side within the interval [SLmin, SLmax] to obtain the left front interpolation point set LFC and the left rear interpolation point set LBE; and performs equally spaced dense interpolation on the two sets of boundary points on the right side within the interval [SRmin, SRmax] to obtain the right front interpolation point set RFC and the right rear interpolation point set RBE. After interpolation, each set of points has the same sampling position in the first direction.

[0067] Step S818: Generate the left contour point set by horizontal selection.

[0068] In one embodiment, the system compares the offsets of LFC and LBE in the second direction, i.e., the L coordinate perpendicular to the driving direction, point by point, and selects the point with the larger offset to form the left contour point set LPS.

[0069] Step S820: Generate the right contour point set by horizontal selection.

[0070] In one embodiment, the system compares the offsets of RFC and RBE point by point in the second direction, and selects the points with larger offsets (i.e., more outward expansion) to form the right contour point set RPS.

[0071] Step S822: Construct the driving profile.

[0072] In one embodiment, the system connects the first and last endpoints based on the left contour point set LPS and the right contour point set RPS to generate the driving contour of the space boundary occupied by the vehicle along the expected driving trajectory.

[0073] Figure 9 This is a schematic structural diagram of an electronic device according to an exemplary embodiment. Please refer to... Figure 9 At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile storage, and may also include other necessary hardware. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it, forming a determination device based on the vehicle's driving profile at the logical level. Of course, this specification does not exclude other implementation methods besides software implementation, such as logic devices or a combination of hardware and software, etc. In other words, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0074] Figure 10 This specification illustrates a block diagram of a device for determining the driving profile of a vehicle, as shown in the embodiments. Please refer to... Figure 10 The device includes: The trajectory acquisition unit 1002 is used to acquire the expected driving trajectory of the vehicle and multiple location points on the expected driving trajectory, wherein the multiple location points include at least the trajectory start point and trajectory end point of the expected driving trajectory. The feature point set acquisition unit 1004 is used to determine multiple feature points on the vehicle contour corresponding to each location point based on the expected pose of the vehicle at that location point, so as to obtain a feature point set composed of multiple feature points corresponding to each location point, wherein the vehicle contour is used to characterize the boundary of the space occupied by the vehicle at the corresponding location point. Side point set acquisition unit 1006 is used to acquire the side point sets corresponding to the vehicle contours at the starting point of the trajectory and the ending point of the trajectory, respectively. The trajectory contour determination unit 1008 is used to determine the boundary of the space occupied by the vehicle along the entire expected driving trajectory based on the feature point set and the side point set, so as to serve as the driving contour of the vehicle.

[0075] Optionally, the trajectory acquisition unit 1002 is specifically used for: The expected driving trajectory is sampled at equal intervals to obtain the multiple location points.

[0076] Optionally, when the vehicle outline is a polygon, the plurality of feature points are at least a portion of the corner points of the polygon; When the vehicle outline is non-polygonal, the plurality of feature points are at least a portion of the boundary points of the protruding outer edge of the non-polygonal part.

[0077] Optionally, the side point set acquisition unit 1006 is specifically used for: Obtain the first left side point set and the first right side point set of the vehicle contour at the starting point of the trajectory, and the second left side point set and the second right side point set of the vehicle contour at the ending point of the trajectory. The first left side point set and the second left side point set are used to represent the points on the left side of the vehicle contour at the corresponding position point, and the first right side point set and the second right side point set are used to represent the points on the right side of the vehicle contour at the corresponding position point.

[0078] Optionally, the trajectory contour determination unit 1008 is specifically used for: Each subset of feature points in the feature point set is combined with the corresponding side point sets in the first left side point set, the first right side point set, the second left side point set, and the second right side point set to obtain multiple sets of boundary points. The correspondence is predetermined based on the vehicle contour part corresponding to each feature point subset and the position and lateral direction corresponding to the side point set. The driving profile is determined based on the multiple sets of boundary points.

[0079] Optionally, the trajectory contour determination unit 1008 is specifically used for: Based on the lateral direction corresponding to each set of boundary points, the multiple sets of boundary points are divided into a left boundary point set group and a right boundary point set group. In a preset coordinate system, the boundary point sets in the left boundary point set are aligned along the first direction, and the offsets in the second direction are compared point by point after alignment. The points with larger offsets are selected to form the left contour point set. In addition, the boundary point sets in the right boundary point set are aligned along the first direction, and the offsets in the second direction are compared point by point after alignment. The points with larger offsets are selected to form the right contour point set. The driving profile is generated based on the left contour point set and the right contour point set.

[0080] Optionally, the preset coordinate system is the SL coordinate system; the first direction is the driving direction along the expected driving trajectory, and the second direction is perpendicular to the driving direction.

[0081] Optionally, the trajectory contour determination unit 1008 is specifically used for: Determine the coordinate interval of each boundary point set in the first direction, and perform equal-interval interpolation within the coordinate interval so that different boundary point sets have the same coordinate values ​​in the first direction.

[0082] Optionally, the device further includes: The instruction generation unit is used to generate corresponding deceleration or emergency braking control instructions when the driving profile overlaps with the boundary of the vehicle's drivable space.

[0083] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this specification according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0084] Based on the same concept as the methods described above, this specification also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor performs the steps of the method as described in any of the above embodiments by executing the executable instructions.

[0085] Based on the same concept as the methods described above, this specification also provides a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0086] Based on the same concept as the methods described above, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0087] The embodiments of the subject matter and functional operation described in this specification can be implemented in the following ways: digital electronic circuits, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing apparatus or for controlling the operation of a data processing apparatus. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or combinations thereof.

[0088] The processing and logic flow described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output. The processing and logic flow can also be executed by dedicated logic circuitry—such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the device can also be implemented as dedicated logic circuitry.

[0089] Computers suitable for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a GPS receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0090] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0091] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily intended to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation thereof.

[0092] Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0093] Therefore, specific embodiments of the subject matter have been described. Furthermore, the processes depicted in the figures are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0094] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A method for determining the driving profile of a vehicle, characterized in that, include: Obtain the vehicle's expected driving trajectory and multiple location points on the expected driving trajectory, wherein the multiple location points include at least the trajectory start point and trajectory end point of the expected driving trajectory; For each location point, based on the expected pose of the vehicle at that location point, multiple feature points on the vehicle contour corresponding to that location point are determined to obtain a feature point set composed of multiple feature points corresponding to each location point, wherein the vehicle contour is used to characterize the boundary of the space occupied by the vehicle at the corresponding location point. Obtain the side point sets corresponding to the vehicle contours at the starting point and ending point of the trajectory, respectively. The side point sets are different from the feature point sets. The side point sets are used to supplement and describe the side line segments that are missing from the feature point sets at the starting point and ending point of the trajectory. Based on the feature point set and the side point set, the boundary of the space occupied by the vehicle along the entire expected driving trajectory is determined, which serves as the driving profile of the vehicle.

2. The method according to claim 1, characterized in that, Obtaining multiple location points on the expected driving trajectory includes: The expected driving trajectory is sampled at equal intervals to obtain the multiple location points.

3. The method according to claim 1, characterized in that, When the vehicle outline is a polygon, the plurality of feature points are at least a portion of the corner points of the polygon; When the vehicle outline is non-polygonal, the plurality of feature points are at least a portion of the boundary points of the protruding outer edge of the non-polygonal part.

4. The method according to claim 1, characterized in that, The step of obtaining the side point sets corresponding to the vehicle contours at the starting point and ending point of the trajectory includes: Obtain the first left side point set and the first right side point set of the vehicle contour at the starting point of the trajectory, and the second left side point set and the second right side point set of the vehicle contour at the ending point of the trajectory. The first left side point set and the second left side point set are used to represent the points on the left side of the vehicle contour at the corresponding position point, and the first right side point set and the second right side point set are used to represent the points on the right side of the vehicle contour at the corresponding position point.

5. The method according to claim 4, characterized in that, The step of determining the boundary of the space occupied by the vehicle along the entire expected driving trajectory based on the feature point set and the side point set, as the driving profile of the vehicle, includes: Each subset of feature points in the feature point set is combined with the corresponding side point sets in the first left side point set, the first right side point set, the second left side point set, and the second right side point set to obtain multiple sets of boundary points. The correspondence is predetermined based on the vehicle contour part corresponding to each feature point subset and the position and lateral direction corresponding to the side point set. The driving profile is determined based on the multiple sets of boundary points.

6. The method according to claim 5, characterized in that, Determining the driving profile based on the multiple sets of boundary points includes: Based on the lateral direction corresponding to each set of boundary points, the multiple sets of boundary points are divided into a left boundary point set group and a right boundary point set group. In a preset coordinate system, the boundary point sets in the left boundary point set are aligned along the first direction, and the offsets in the second direction are compared point by point after alignment. The points with larger offsets are selected to form the left contour point set. In addition, the boundary point sets in the right boundary point set are aligned along the first direction, and the offsets in the second direction are compared point by point after alignment. The points with larger offsets are selected to form the right contour point set. The driving profile is generated based on the left contour point set and the right contour point set.

7. The method according to claim 6, characterized in that, The preset coordinate system is the SL coordinate system; the first direction is the driving direction along the expected driving trajectory, and the second direction is perpendicular to the driving direction.

8. The method according to claim 6, characterized in that, The position alignment along the first direction includes: Determine the coordinate interval of each boundary point set in the first direction, and perform equal-interval interpolation within the coordinate interval so that different boundary point sets have the same coordinate values ​​in the first direction.

9. The method according to claim 1, characterized in that, The method further includes: When the driving profile overlaps with the boundary of the vehicle's drivable space, a corresponding deceleration or emergency braking control command is generated.

10. An electronic device, characterized in that, include: processor; A memory for storing processor-executable instructions; wherein the processor implements the steps of the method as described in any one of claims 1-9 by executing the executable instructions.

11. A computer-readable storage medium, characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 9.

12. A computer program product, characterized in that, Includes a computer program / instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 9.

Images