Parallel parking space parking-out path planning method and device and computer equipment

Abstract

This invention discloses a method, apparatus, and computer device for planning parking exit paths in parallel parking spaces. The method includes: determining a first turning center on the side of the target lane closest to the starting position of the vehicle's parking exit, based on the vehicle's minimum turning radius and the vehicle's starting position in the parallel parking space; determining the intersection point of the vehicle's front corner position with the first turning center on the long side line of the parallel parking space; obtaining the drivable area for the vehicle to exit from the parallel parking space to the target lane; if the intersection point is not within the drivable area, determining the reversing target position of the vehicle in the parallel parking space; and using the reversing target position as the starting position of the vehicle's parking exit in the parallel parking space, returning to the step of determining the first turning center. This improves the reliability of parking exit path planning in parallel parking spaces through dynamic adjustment strategies.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus, and computer equipment for planning parking exit routes for parallel parking spaces. Background Technology

[0002] With the rapid development of intelligent driving assistance technology, automatic parking assist systems have become one of the important intelligent features of mid-to-high-end mass-produced vehicles. Parallel parking exit function, as a key component of automatic parking assist systems, aims to automate the safe exit of vehicles from parallel parking spaces.

[0003] In related technologies, the implementation of parallel parking space exit functionality mainly relies on environmental models constructed by sensors such as ultrasonic radar and cameras, combined with rule-based or geometrically computational path planning algorithms. However, the path planning algorithms in these technologies often plan only a single path, resulting in a low success rate for vehicle exit under complex environmental conditions such as road obstacles. Therefore, the reliability of parallel parking space exit path planning algorithms under complex constraints urgently needs improvement. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in related technologies. To this end, this invention proposes a method, apparatus, and computer device for planning parking exit routes for parallel parking spaces. Through dynamic adjustment strategies during the route planning process, a more compact and efficient parking exit route is planned, effectively improving the reliability of parking exit route planning for parallel parking spaces.

[0005] To achieve the above objectives, a first aspect of the present invention provides a parking exit path planning method for parallel parking spaces. The method includes: determining a first turning center on the side of the parking exit starting position closer to the target lane, based on the vehicle's minimum turning radius and the vehicle's initial parking position in the parallel parking space; wherein the target lane is located on one side of one of the long side lines of the parallel parking space; the line connecting the first turning center and the initial parking position is perpendicular to the vehicle's centerline; determining the intersection point of the vehicle's front corner position with the first turning center as the center, moving forward to the corresponding intersection point on the straight line containing the long side line of the parallel parking space; wherein the front corner position is located on the side of the parallel parking space furthest from the target lane. The process involves: obtaining the drivable area from the parallel parking space to the target lane; if the intersection point is within the drivable area, performing path planning based on the parking start position, the minimum turning radius, the angle corresponding to the path from the vehicle's front corner to the intersection point, and the lane width of the target lane to obtain a first parking path; if the intersection point is not within the drivable area, determining the reversing target position of the vehicle in the parallel parking space; using the reversing target position as the parking start position of the vehicle in the parallel parking space, and returning to the step of determining the first turning center on the side of the parking start position closer to the target lane based on the vehicle's minimum turning radius and the vehicle's parking start position in the parallel parking space.

[0006] According to one embodiment of the present invention, determining the reversing target position of the vehicle in the parallel parking space includes: determining a second turning center symmetrical to the first turning center on the side away from the target lane from the parking starting position, based on the minimum turning radius and the parking starting position; determining the minimum movement angle of the left and right rear corner positions of the vehicle, respectively, to move backward to the boundary of the drivable area with the second turning center as the center; and performing reversing path planning based on the minimum movement angle, the parking starting position, and the second turning center to determine the reversing target position.

[0007] According to one embodiment of the present invention, the step of planning a reversing path based on the minimum movement angle, the parking start position, and the second turning center to determine the reversing target position includes: taking the parking start position as the starting point and the second turning center as the center, performing circular interpolation processing based on the minimum turning radius and the minimum movement angle in the reversing direction of the vehicle to obtain a reversing path; and determining the reversing target position based on the end point of the reversing path.

[0008] According to one embodiment of the present invention, the starting position of the vehicle's parking in the parallel parking space is the position of the rear axle center point of the vehicle; after obtaining the first parking path, the method further includes: if the vehicle contains obstacle information in the bounding box corresponding to the end point of the first parking path, determining the boundary projection point of the obstacle boundary point in the obstacle information on the straight line where the specified long side line of the parallel parking space is located; wherein, the end point of the first parking path is located on the center axis of the target lane; determining the target boundary point position based on the minimum distance between the projection point of the parking starting position on the straight line where the specified long side line of the parallel parking space is located and the boundary projection point; calculating the target parking position of the vehicle on the center axis of the target lane based on the target boundary point position, the distance between the rear axle center point of the vehicle and the foremost point of the vehicle, and a preset interval distance; and performing path planning based on the parking starting position, the target position, and the minimum turning radius to obtain a second parking path.

[0009] According to one embodiment of the present invention, the path planning based on the parking start position, the minimum turning radius, the angle corresponding to the path from the vehicle's front corner position to the intersection point, and the lane width of the target lane includes: taking the parking start position as the starting point and the first turning center as the center, performing circular arc interpolation processing in the vehicle's forward direction based on the minimum turning radius and the angle corresponding to the path from the vehicle's front corner position to the intersection point to obtain a first circular arc curve segment; replacing the circular arc segment of the first circular arc curve segment with a preset curve length at the end furthest from the parking start position with a first spiral curve segment to obtain a first trajectory segment; wherein, the curvature of the starting point of the first spiral curve segment is the curvature of the first circular arc trajectory, and the curvature of the ending point of the first spiral curve segment is 0; the preset curve length is determined according to the lane width of the target lane; taking the ending point of the first trajectory segment as the starting point, along... The endpoint direction of the first trajectory segment determines a first straight segment with a preset straight length, resulting in a second trajectory segment; wherein the preset straight length is determined based on the lane width of the target lane; taking the endpoint of the second trajectory segment as the starting point, a third trajectory segment is determined based on the endpoint direction of the second trajectory segment; wherein the third trajectory segment includes a second spiral curve segment connected to the second trajectory segment, a second arc curve segment connected to the second spiral curve segment, and a third spiral curve segment connected to the second arc curve segment; the second spiral curve segment and the third spiral curve segment have the preset curve length; the radius of the second arc curve segment is the minimum turning radius; taking the endpoint of the third trajectory segment as the starting point, a second straight segment with the preset straight length is determined along the endpoint direction of the third trajectory segment, resulting in a fourth trajectory segment; wherein the fourth trajectory segment is located on the centerline of the target lane.

[0010] According to one embodiment of the present invention, after obtaining the first parking exit path, the method further includes: if, during the process of the vehicle traveling along the first parking exit path, an obstacle is determined to exist within a preset range of the vehicle at the current trajectory point position, trajectory replanning is performed starting from the current trajectory point position.

[0011] According to one embodiment of the present invention, the trajectory replanning based on the current trajectory point position includes: replanning the trajectory based on the hybrid A* algorithm, using the current trajectory point position as the starting point.

[0012] According to one embodiment of the present invention, the step of determining to perform trajectory replanning starting from the current trajectory point position when the vehicle is traveling along the first parking path and an obstacle is determined to exist within a preset range of the vehicle at the current trajectory point position includes: when the vehicle is traveling along the first parking path and the front corner position is located within the range corresponding to the target lane, if an obstacle is determined to exist within a preset range of the vehicle at the current trajectory point position, trajectory replanning is performed starting from the current trajectory point position.

[0013] To achieve the above objectives, a second aspect of the present invention provides a parking exit path planning device for parallel parking spaces. The device includes: a turning center determination module, used to determine a first turning center on the side of the parking exit starting position closer to the target lane, based on the vehicle's minimum turning radius and the vehicle's parking exit starting position in the parallel parking space; wherein the target lane is located on one side of one of the long side lines of the parallel parking space; the line connecting the first turning center and the parking exit starting position is perpendicular to the vehicle's centerline in the vehicle's direction; and an intersection point determination module, used to determine the intersection point corresponding to the long side line of the parallel parking space where the vehicle's front corner position point moves forward with the first turning center as the center; wherein the front corner position point is located on the side of the parallel parking space away from the target lane. The path planning module is used to obtain the drivable area from the parallel parking space to the target lane. If the intersection is located within the drivable area, path planning is performed based on the parking start position, the minimum turning radius, the angle corresponding to the path from the front corner of the vehicle to the intersection, and the lane width of the target lane to obtain a first parking path. The reversing target position determination module is used to determine the reversing target position of the vehicle in the parallel parking space if the intersection is not located within the drivable area, and to use the reversing target position as the parking start position of the vehicle in the parallel parking space, returning to the step of determining the first turning center on the side of the parking start position closer to the target lane based on the minimum turning radius of the vehicle and the parking start position of the vehicle in the parallel parking space.

[0014] To achieve the above objectives, a third aspect of the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the berthing path planning method described in any of the preceding embodiments.

[0015] According to multiple embodiments of the present invention, whether the intersection point is within the drivable area is used as the criterion for determining whether a vehicle can directly exit from a parallel parking space. When direct forward movement is not feasible, a proactive and orderly reversing adjustment is introduced, and the path is replanned based on the adjusted position and posture. This allows the system to adapt to narrower or more complex parking scenarios, effectively solving the problem of "repeatedly maneuvering around the parking space" without a clear strategy, transforming "maneuvering around the parking space" into a purposeful and calculable optimization process. Through this dynamic adjustment strategy, a more compact and efficient parking path is planned, effectively improving the reliability of parallel parking space exit path planning, and is particularly suitable for parallel parking space exit scenarios in complex urban parking environments.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] Figure 1a This is a flowchart illustrating a parking exit path planning method for parallel parking spaces provided according to an embodiment of this specification.

[0018] Figure 1b This is a schematic diagram of a parallel parking space and drivable area scenario provided according to an embodiment of this specification.

[0019] Figure 1c This is a schematic diagram illustrating the determination of a first turning center according to an embodiment of this specification.

[0020] Figure 2 This is a flowchart illustrating a method for determining the target position of a vehicle reversing in a parallel parking space, according to an embodiment of this specification.

[0021] Figure 3a This is a flowchart illustrating another berthing path planning method provided according to the embodiments of this specification.

[0022] Figure 3b This is a schematic diagram of the endpoint of a first berthing path provided according to an embodiment of this specification.

[0023] Figure 3c This is a schematic diagram of a target location provided according to an embodiment of this specification.

[0024] Figure 4 This is a schematic diagram of a path planning process provided according to an embodiment of this specification. Detailed Implementation

[0025] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0026] With the rapid development of intelligent driving assistance technology, Automated Parking Assist (APA) systems have become one of the important intelligent features of mid-to-high-end mass-produced vehicles. The parallel parking exit function, as a key component of the APA system, aims to automate the safe exit of a vehicle from a parallel parking space through the coordinated operation of three modules: environmental perception, path planning, and vehicle control.

[0027] In related technologies, the implementation of parallel parking space exit functionality mainly relies on an environmental model constructed by sensors such as ultrasonic radar and cameras, combined with path planning algorithms based on rules or geometric calculations. Typical technical solutions often employ preset path templates, generating parking trajectories through combinations of arcs or straight lines with fixed curvature. For example, some systems use a "three-stage" planning strategy: first, reversing at the maximum steering angle to adjust the vehicle's direction; then, using a fixed-radius arc to complete lateral movement; and finally, driving out of the parking space in a straight line. This type of method can achieve basic functionality in open spaces or scenarios with standard parking space spacing (typically requiring 1.5 times the vehicle length of the car to be left between the front and rear vehicles).

[0028] However, in practical applications of mass-produced vehicles, the path planning schemes in related technologies have revealed significant shortcomings in environmental adaptability. Particularly in urban parking scenarios, the following prominent problems exist: First, the planned paths are too simplistic, leading to multiple parking failures when obstacles are present on the road; second, when the distance between the vehicle waiting to park and the vehicles in the spaces in front and behind is too small, it will fall into a state of repeated oscillations around the parking space, or experience multiple parking failures. Therefore, the reliability of existing parallel parking space exit path planning algorithms under complex constraints urgently needs to be improved.

[0029] To improve the reliability of parallel parking space exit path planning and thus increase the parking success rate of the automatic parking assistance system, it is necessary to propose a parallel parking space exit path planning method, device, and computer equipment. The parallel parking space exit path planning method provided in this specification first uses the vehicle's minimum turning radius as a basic parameter. Combined with the vehicle's starting position in the parallel parking space, a first turning center is determined on the side of the starting position closest to the target lane (the lane the vehicle needs to enter), ensuring that the line connecting the first turning center and the starting position is perpendicular to the vehicle's centerline. Then, using the corner of the vehicle's front end furthest from the target lane as the tracking target, the trajectory of the vehicle moving forward around the first turning center is simulated, and the intersection of this trajectory with the straight line containing the long side line (parking space boundary) of the parallel parking space is determined.

[0030] Based on the vehicle's drivable area, determine whether the intersection point is located within that drivable area. If the intersection point is located within the drivable area, then based on the parking start position, minimum turning radius, the angle corresponding to the path from the vehicle's front corner to the intersection point, and the width of the target lane, plan a parking path to generate the first parking path.

[0031] If the intersection point is not within the drivable area, the system automatically calculates the reversing target position and uses the reversing target position as the new parking start position. Based on the vehicle's pose at the reversing target position, the above path planning steps are repeated until a feasible path is generated.

[0032] Therefore, the criterion for determining whether a vehicle can directly exit a parallel parking space is whether the intersection point is within the drivable area. When direct forward movement is not feasible, a proactive and orderly reversing adjustment is introduced, and the path is replanned based on the adjusted position and posture. This allows the system to adapt to narrower or more complex parking scenarios, effectively solving the problem of "repeatedly maneuvering" without a clear strategy, and transforming "maneuvering" into a purposeful and calculable optimization process. Through this dynamic adjustment strategy, a more compact and efficient parking path is planned, effectively improving the reliability of parallel parking space exit path planning, especially suitable for parallel parking space exit scenarios in complex urban parking environments.

[0033] This specification presents a method for planning parking exit routes for parallel parking spaces, referencing... Figure 1a As shown, the berthing route planning method may include the following steps.

[0034] S110. Based on the vehicle's minimum turning radius and the vehicle's starting position for exiting the parallel parking space, determine the first turning center on the side of the starting position closer to the target lane; wherein the target lane is located on one side of one of the long side lines of the parallel parking space; the line connecting the first turning center and the starting position for exiting the parallel parking space is perpendicular to the vehicle's centerline in the direction of the vehicle body.

[0035] S120. Determine the position of the vehicle's front corner and move forward from the center of the first turn to the intersection point on the straight line containing the long side of the parallel parking space; wherein, the position of the front corner is located on the side of the vehicle away from the target lane in the parallel parking space.

[0036] S130. Obtain the drivable area when the vehicle parks from the parallel parking space to the target lane. If the intersection is located within the drivable area, perform path planning based on the parking start position, minimum turning radius, the angle corresponding to the path from the front corner of the vehicle to the intersection, and the lane width of the target lane to obtain the first parking path.

[0037] S140. If the intersection is not located within the drivable area, determine the target position for reversing the vehicle in the parallel parking space.

[0038] S150, Taking the reversing target position as the starting position for the vehicle to exit in the parallel parking space, return to the step of determining the first turning center on the side of the starting position closer to the target lane based on the vehicle's minimum turning radius and the starting position for the vehicle to exit in the parallel parking space.

[0039] Parallel parking spaces refer to parking spaces where the direction of the vehicle is parallel to the direction of the road.

[0040] The vehicle's centerline refers to the straight line connecting the center points of the front and rear axles.

[0041] The front corner position point is the corresponding position point of the front corner point on the path planning plane.

[0042] The long side line of a parallel parking space refers to the longer boundary line on both sides of the parallel parking space that is parallel to the direction of the road.

[0043] It should be noted that the first turning center, parallel parking space, target lane, parking start position, vehicle centerline, front corner position, and drivable area are all located within the same path planning plane.

[0044] Specifically, the above-mentioned parking path planning method is applied to a control system with automatic parking function (e.g., an APA system, hereinafter referred to as the system). The system first determines the first turning center based on the vehicle's minimum turning radius and the vehicle's starting position in the parallel parking space. This turning center is located on the side of the starting position closer to the target lane (i.e., the side of the road the vehicle will enter), and the line connecting the first turning center and the starting position is perpendicular to the vehicle's centerline in the direction of the vehicle body at the starting position, to ensure compliance with the vehicle's actual steering kinematic constraints.

[0045] Then, the system selects the corner of the vehicle's front end in the parallel parking space, away from the target lane. The corresponding corner position is used as the tracking point. The system simulates the movement of this corner position forward around the first turning center, and calculates the intersection point of its trajectory with the long side line of the parallel parking space (i.e., the intersection point when the corner position moves forward around the first turning center to the long side line of the parallel parking space). This intersection point is recorded as the intersection point. It's important to understand that the long side line of the parallel parking space here refers to the side closest to the target lane.

[0046] The system acquires the drivable area for a vehicle to exit from its parallel parking space to the target lane. This is typically determined by sensors (such as ultrasonic radar, cameras, etc.) detecting surrounding obstacles (such as oncoming vehicles, curbs, other stationary objects, etc.) in real time, and then judging whether the intersection point is within the drivable area. If the intersection point is within the drivable area, it indicates that it is safe and feasible for the vehicle to drive forward directly from its current parking start position and pose. Then, based on parameters such as the parking start position, the vehicle's minimum turning radius, the central angle of the arc swept by the vehicle's front corner position point to the intersection point (i.e., the angle corresponding to the path of the vehicle's front corner position point to the intersection point), and the width of the target lane, the system performs path planning to generate the first parking path.

[0047] If the intersection point is outside the drivable area (usually even if it's on the boundary of the drivable area, it's still considered outside the drivable area), it means that driving directly forward would result in a collision with an obstacle, so a reversing adjustment strategy is introduced. That is, the system determines a target reversing position, uses this target position as the new parking starting position, and then returns to the steps described above—determining the first turning center based on the vehicle's minimum turning radius and the parking starting position—to plan the parking path until a feasible path is generated. Through this iterative approach, the system can explore ways to optimize the parking path by first reversing and then moving forward—a "parking maneuver."

[0048] For example, the parking start position can be determined based on the center point of the vehicle's rear axle or the geometric center point of the vehicle body. Taking the center point of the vehicle's rear axle as an example, refer to... Figure 1b As shown, the target lane is above the parallel parking space (i.e., to the left of the vehicle to be parked) and adjacent to the parallel parking space. Based on the vehicle's minimum turning radius and its starting position in the parallel parking space, when determining the first turning center on the side closest to the target lane from the starting position, the center point of the rear axle of the vehicle to be parked can be used as the coordinate origin O(X). O Y O The positive x-axis is defined as the direction from the center point of the rear axle of the vehicle towards the front (parallel to the center line of the vehicle's direction), and the positive y-axis is defined as the direction from the center point of the rear axle of the vehicle towards the target lane (perpendicular to the x-axis). Figure 1bIn the initial parking state shown, the center point of the rear axle of the vehicle to be parked is the origin O(X). O Y O (At this time X) O =Y O =0), the horizontal direction is the x-axis, the vertical direction is the y-axis, and the counterclockwise rotation direction is the positive direction.

[0049] When the system obtains the drivable area of ​​a vehicle from a parallel parking space to the target lane, it can use sensors to obtain information about the vehicle to be parked, surrounding vehicles and obstacles, and then establish the global coordinate system mentioned above and create a grid map.

[0050] Specifically, let m be the distance m extended from point O in the positive and negative x-axis directions, and n be the distance n extended in the positive and negative y-axis directions, with length and width as... The system creates a grid map based on the surrounding vehicles and obstacles. Figure 1b The drivable area is shown, and each grid cell is assigned a status value. A single grid cell typically contains the following information: (1) grid number; (2) whether it is an obstacle grid cell; (3) the x-direction distance and y-direction distance to the nearest obstacle grid cell; (4) the Euclidean distance to the nearest obstacle grid cell. It should be noted that the above extension distances m and n can be set according to actual application requirements, etc., and this manual does not impose specific limitations. Meanwhile, Figure 1b The boundary of the drivable area shown is only an example boundary. In actual scenarios, the boundary of the drivable area may be based on lane lines, parking space lines, etc.

[0051] exist Figure 1b In the parallel parking space and drivable area scenario shown, the minimum turning radius R of the vehicle to be parked is determined using the Ackermann steering mechanism principle. min Then we have:

[0052] in, This refers to the vehicle's wheelbase. The wheelbase of the vehicle. This is the maximum steering angle of the steering wheel.

[0053] refer to Figure 1c As shown, based on the minimum turning radius R min The first turning center O1 (X) can be determined in the positive y-axis direction. O1 Y O1 (Initial state X) O1 =X O =0, Y O1 =R minAt this point, the right front corner position of the vehicle on the side furthest from the target lane is designated as point P. Simulate the movement of point P forward around the first turning center O1, and calculate the intersection of the trajectory of point P with the straight line containing the long side of the parallel parking space. This calculates the radius R of point P. P Let C1 be the intersection point of the arc whose radius is rotated counterclockwise around the center O1 and the straight line containing the upper edge of the parking space. The turning radius R of point P is... P for:

[0054] Among them, continue to refer to Figure 1c As shown, The distance from the front wheel to the front of the car. This refers to the distance from the rear wheel to the rear of the vehicle. The description of scenarios where the target lane is below the parallel parking space (or, the target lane is to the right of the vehicle) is similar to the above description and will not be repeated here.

[0055] In some feasible methods, if the system determines that there are no obstacles behind the vehicle, it can simulate the vehicle moving backward (i.e., in the direction of the parking space) a certain distance in a straight line until the intersection of the trajectory of the front corner of the vehicle and the long side line of the parallel parking space is located within the drivable area, so as to determine the target position for reversing.

[0056] In other feasible methods, the vehicle can be simulated to turn backward and move a certain distance until the intersection of the trajectory of the front corner of the vehicle and the long side line of the parallel parking space is within the drivable area, or until a certain rear corner of the vehicle reaches the edge line of the parallel parking space or the boundary of the drivable area, in order to determine the target position for reversing.

[0057] It should be noted that the target lane can be adjacent to a parallel parking space, or it can be separated from a parallel parking space by a certain distance, or there can be a lane between it and the parallel parking space. The specific requirements are determined based on the actual application scenario.

[0058] Path planning is performed based on the starting position of parking, the minimum turning radius, the angle corresponding to the path from the vehicle's front corner position to the intersection point, and the width of the target lane. It can be implemented using any of the following methods: geometric curve planning algorithm, sampling-based path planning algorithm, or search-based path planning algorithm. No specific limitation is made in this specification.

[0059] In the above implementation, whether the intersection point is within the drivable area is used as the criterion for determining whether a vehicle can directly exit from a parallel parking space. When direct forward movement is not feasible, a proactive and orderly reversing adjustment is introduced, and the path is replanned based on the adjusted position and posture, enabling the system to adapt to narrower or more complex parking scenarios. Compared to related technologies that use preset path templates to generate parking trajectories through combinations of arcs or straight lines with fixed curvature, the parallel parking space exit path planning method provided in this specification effectively solves the problem of "repeatedly rubbing the parking space" without a clear strategy, transforming "rubbing the parking space" into a purposeful and calculable optimization process, improving the system's environmental adaptability and robustness. Through this dynamic adjustment strategy, a more compact and efficient parking path is planned, reducing unnecessary movement distance and time, effectively improving the reliability of parallel parking space exit path planning, and achieving a safer, more efficient, and intelligent automatic parking function, especially suitable for parallel parking space exit scenarios in complex urban parking environments.

[0060] In some implementations, reference Figure 2 As shown, determining the target position for reversing a vehicle in a parallel parking space can include the following steps.

[0061] S210. Based on the minimum turning radius and the parking start position, determine a second turning center that is symmetrical to the first turning center on the side of the parking start position away from the target lane.

[0062] S220. Determine the minimum angle of motion for the left and right rear corner positions of the vehicle to move backward to the boundary of the drivable area with the second turning center as the center.

[0063] S230. Based on the minimum movement angle, the parking start position, and the second turning center, plan the reversing path and determine the reversing target position.

[0064] The left rear corner position point is the position point corresponding to the left corner of the vehicle's rear on the path planning plane, also known as the left rear corner position point; the right rear corner position point is the position point corresponding to the right corner of the vehicle's rear on the path planning plane, also known as the right rear corner position point.

[0065] Specifically, if the intersection is not located within the drivable area, the system uses the vehicle's minimum turning radius and parking start position as a reference to determine a second turning center that is symmetrical to the first turning center on the side of the parking start position away from the target lane. That is, the second turning center that is symmetrical to the first turning center is determined with the vehicle's centerline as the axis of symmetry.

[0066] The system tracks the left and right rear corner positions of the vehicle, simulating their backward trajectories around the second turning center. It calculates the angles of the paths taken by each rear corner position when they reach the boundary of the drivable area, and selects the smallest angle as the minimum motion angle. Based on this minimum motion angle, the parking start position, and the second turning center, a reversing path is planned. That is, using the minimum motion angle as a constraint, combined with the parking start position and the second turning center, a reversing trajectory is planned starting from the current parking start position, and the endpoint of the trajectory is the target reversing position. It can be understood that "the two rear corner positions reaching the boundary of the drivable area" means that each rear corner position reaches the nearest boundary of the drivable area.

[0067] For example, in Figure 1b Based on the initial parking state shown and its corresponding initial coordinate system, and according to the minimum turning radius R... min Determine the first turning center O1 (X) in the positive y-axis direction. O1 Y O1 (Initial state X) O1 =X O =0, Y O1 =R min If the intersection point is not located within the drivable area, the system determines a second turning center O0 (X) symmetrical to the first turning center in the negative y-direction, with the x-axis as the axis of symmetry. O1 -Y O1 Then O0(X) O1 -Y O1 Taking R as the turning center, calculate the directions counterclockwise. P1 (Turn radius of point P1 at the left rear corner), R P2 The intersection point F of the arc with radius (the turning radius of the right rear corner point P2) and the boundary of the drivable area. P1 F P2 Calculate the central angles corresponding to the two arcs respectively, and take the smaller central angle as the minimum motion angle.

[0068] It should be noted that the turning radius R of the left rear corner position point P1 is... P1 The turning radius R of the right rear corner position point P2 P2 The calculation method is the same as the turning radius R of the aforementioned front corner position point P. P The calculation method is the same, and the details will not be repeated here.

[0069] In some implementations, the reversing path planning based on the minimum movement angle, the parking start position, and the second turning center to determine the reversing target position may include: taking the parking start position as the starting point and the second turning center as the center, performing circular interpolation processing based on the minimum turning radius and minimum movement angle in the vehicle's reversing direction to obtain the reversing path; and determining the reversing target position based on the end point of the reversing path.

[0070] Specifically, the system takes the parking start position as the starting point, the second turning center as the center, and the minimum turning radius as the radius. It then performs circular interpolation within the minimum motion angle range in the vehicle's reverse direction to generate a series of trajectory points within that range. Connecting these points yields the reversing path, and the endpoint of this path is determined as the target reversing position.

[0071] For example, when performing circular interpolation based on the minimum turning radius and minimum motion angle in the reverse direction of the vehicle, the step size can be 0.05, 0.03, or 0.1 arc lengths.

[0072] It should be noted that the step size for circular interpolation can be determined by setting a preset angle or preset arc length according to actual application requirements, and this manual does not specify a specific limit.

[0073] In some implementations, the starting position of the vehicle when exiting the parallel parking space is the position of the center point of the vehicle's rear axle; see reference. Figure 3a As shown, after obtaining the first berthing path, the berthing path planning method may also include the following steps.

[0074] S310. If the bounding box corresponding to the end point of the first parking path contains obstacle information, determine the boundary projection point of the obstacle boundary point in the obstacle information on the straight line where the specified long side line of the parallel parking space is located; wherein, the end point of the first parking path is located on the center axis of the target lane.

[0075] S320. Determine the location of the target boundary point based on the minimum distance between the projection point of the starting position of parking on the straight line of the specified long side of the parallel parking space and the boundary projection point.

[0076] S330: Based on the target boundary point location, the distance between the rear axle center point of the vehicle and the frontmost point of the vehicle, and the preset interval distance, calculate and determine the target parking position of the vehicle on the centerline of the target lane.

[0077] S340. Based on the starting position, target position, and minimum turning radius, a path is planned to obtain the second berthing path.

[0078] The bounding box is a rectangular box used to represent the space occupied by the vehicle.

[0079] The centerline of the target lane is the middle line of the width of the target lane.

[0080] The designated long side line of a parallel parking space can be any long side line of the parallel parking space.

[0081] The target boundary point location contains the coordinate information that has the greatest impact on the vehicle's parking endpoint, and is used as a reference for obstacle avoidance.

[0082] The preset interval distance is a preset safety margin between the vehicle and surrounding objects.

[0083] Specifically, after obtaining the first parking exit path, the system can determine the endpoint of the first parking exit path and whether the bounding box of the vehicle at that endpoint contains obstacle information detected by sensors. If obstacle information is included, the system extracts the boundary points of the obstacles (i.e., obstacle boundary points) from the obstacle information and projects the extracted obstacle boundary points vertically onto the straight line of the designated long side of the parallel parking space (usually the long boundary line of the parking space closer to the target lane) to obtain the corresponding boundary projection points.

[0084] The system also projects a perpendicular image from the starting position of the parking maneuver onto the straight line containing the designated long side of the parallel parking space, obtaining the corresponding projection point. The system calculates the distance between each boundary projection point and the projection point corresponding to the starting position of the parking maneuver, determines the minimum distance, and uses the boundary projection point corresponding to the minimum distance to determine the position of the target boundary point.

[0085] Using the target boundary point as a reference, the offset distance is calculated along a line parallel to the lane edge towards the parking space where the vehicle is parked. This offset distance is equal to "the distance from the front of the vehicle to the center of the rear axle + a preset interval distance". The offset position is then determined, and this point is projected onto the center line of the target lane to obtain the final position that the center point of the vehicle's rear axle should reach, i.e., the target parking position. The system performs path planning based on the parking start position, target position, and minimum turning radius to obtain the second parking path.

[0086] In some cases, the boundary point of the obstacle closest to the vehicle can be extracted from the obstacle information and used as the obstacle boundary point.

[0087] For example, refer to Figure 3b As shown, the target lane is adjacent to the parallel parking space where the vehicle is located, and the width of the parallel parking space is D. L The target lane width is W R / 2, The centerline of the target lane is y=D L / 2+W R / 4. After obtaining the first berthing exit path through system planning, the endpoint of the first berthing exit path is determined to be E1 (X). E1 YE1 ), where X E1 =X O +D x Y E1 =D L / 2+W R / 4.

[0088] If, at the end point E1 of the first parking exit path, the system determines that the vehicle's bounding box contains information about an obstacle vehicle (represented by grid information), then refer to... Figure 3c As shown, the system determines the target boundary point position F(X) based on the boundary projection point of the obstacle boundary point on the straight line containing the long side line between the parallel parking space and the target lane, and the projection point of the parking start position point O on the straight line containing the long side line. F Y F ), where the target boundary point F is the point projected onto the centerline of the target lane.

[0089] The system uses F as a reference and calculates the offset distance along the centerline of the target lane towards the parking space where the vehicle is parked. The offset distance is the distance D from the foremost point of the vehicle to the center of the rear axle. c Distance D from the preset interval s The system determines the offset position point E2 (X) based on the target boundary point position F and the offset distance. E2 Y E2 ), that is, the new target location. Where X... E2 =X F -D s -D c Y E2 =Y F =D L / 2+W R / 4.

[0090] The system performs path planning based on the starting position O, the target position E2, and the minimum turning radius to obtain the second exit path.

[0091] Understandably, since the obstacle boundary points are determined based on the obstacle information contained in the bounding box corresponding to the end point of the vehicle's first parking path, therefore, continued reference... Figure 3c As shown, the x-coordinate (i.e., the x-coordinate X) of the target boundary point F is... F () is the coordinate range in the y-axis direction. The minimum x-coordinate among all boundary points within the vehicle's track. Here, w is the vehicle's wheelbase.

[0092] It should be noted that the target boundary point can be determined on the straight line of the specified long side of the parallel parking space, or on the original obstacle boundary point, or on the centerline of the target lane. The specific location can be determined according to the actual application scenario, etc., and this manual does not make specific limitations.

[0093] The preset interval distance can be determined according to the actual application scenario or needs, such as 30cm or 50cm, etc., and this manual does not make a specific limitation.

[0094] In the above embodiments, by introducing a perception and response mechanism for the end environment of the first parking path, when there is a possibility of collision with an obstacle at the end of the first parking path, the parking target end point is dynamically adjusted, and a second parking path with higher accuracy is generated. This avoids the situation where the vehicle needs to be repeatedly adjusted or the parking fails because the end of the first parking path cannot be parked, thereby improving the vehicle parking efficiency and effectively improving the reliability of parking path planning.

[0095] In some implementations, reference Figure 4 As shown, path planning based on the parking start position, minimum turning radius, the angle corresponding to the path from the vehicle's front corner position to the intersection point, and the lane width of the target lane can include: S410. Starting from the parking start position and with the first turning center as the center, perform circular interpolation processing based on the minimum turning radius and the angle corresponding to the path from the front corner position point to the intersection point in the vehicle's forward direction to obtain the first circular arc curve segment.

[0096] S420. Replace the arc segment of the first arc curve segment with the arc segment of the preset curve length at the end far from the starting position of parking with the first spiral curve segment to obtain the first trajectory segment; wherein, the starting curvature of the first spiral curve segment is the curvature of the first arc trajectory, and the ending curvature of the first spiral curve segment is 0; the preset curve length is determined according to the lane width of the target lane.

[0097] S430. Taking the end point of the first trajectory as the starting point, determine a first straight line segment with a preset straight line length along the direction of the end point of the first trajectory segment to obtain the second trajectory segment; wherein, the preset straight line length is determined according to the lane width of the target lane.

[0098] S440. Taking the end point of the second trajectory as the starting point, determine the third trajectory according to the direction of the end point of the second trajectory; wherein, the third trajectory includes a second spiral curve segment connected to the second trajectory, a second circular arc curve segment connected to the second spiral curve segment, and a third spiral curve segment connected to the second circular arc curve segment; the second spiral curve segment and the third spiral curve segment have preset curve lengths; the radius of the second circular arc curve segment is the minimum turning radius.

[0099] S450. Taking the end point of the third trajectory as the starting point, determine a second straight line segment with a preset straight line length along the direction of the end point of the third trajectory to obtain the fourth trajectory; wherein, the fourth trajectory is located on the centerline of the target lane.

[0100] Among them, the first, second, third, and fourth trajectory segments are located within the same path planning plane as the centerline of the target lane.

[0101] Specifically, when planning a path based on the starting position of the vehicle's exit, the minimum turning radius, the angle corresponding to the path from the vehicle's front corner to the intersection point, and the width of the target lane, the system uses a geometric curve path planning algorithm. First, taking the starting position of the vehicle's exit in the parallel parking space as the starting point and the first turning center as the center, the system performs arc interpolation in the forward direction, using the minimum turning radius as the radius, according to the central angle corresponding to the path from the vehicle's front corner to the intersection point. The interpolated arc is the first arc curve segment.

[0102] Secondly, the arc portion of the first circular curve segment (the end furthest from the starting position) with a predetermined curve length is replaced with a spiral curve segment, namely the first spiral curve segment. The starting curvature of the first spiral curve segment is the same as that of the first circular curve segment, and the ending curvature is 0, thus solving the problem of abrupt curvature change at the end of the circular path and achieving a smooth transition from turning to straight-line travel. The first trajectory segment is constructed based on the unreplaced preceding circular arc segment and the replaced first spiral curve segment.

[0103] Secondly, taking the end point of the first trajectory as the starting point, extend a straight line segment of a predetermined length along the direction of the end point of the first trajectory or the tangent direction of the end point (i.e., the direction of vehicle movement here), which is the first straight line segment, to form the second trajectory segment.

[0104] Then, using the end point of the second trajectory segment as the starting point and the direction of that end point as the initial direction, the third trajectory segment is constructed. The third trajectory segment consists of a second spiral curve segment, a second circular arc curve segment, and a third spiral curve segment. The starting point of the second spiral curve segment is the end point of the second trajectory segment, and the length of the second spiral curve segment is a preset curve length. The end point of the second spiral curve segment is the starting point of the second circular arc curve segment. The radius of the second circular arc curve segment is the minimum turning radius of the vehicle; therefore, the curvature at the starting point of the second spiral curve segment is 0, and the curvature at the ending point is the curvature of the second circular arc curve segment.

[0105] The end point of the second circular curve segment is the starting point of the third spiral curve segment. The length of the third spiral curve segment is the preset curve length, and the end point of the third spiral curve segment is tangent to the centerline of the target lane.

[0106] Finally, starting from the end of the third trajectory segment or the third spiral curve segment, extend a straight segment of a predetermined length along the direction of the end of the third trajectory segment or the third spiral curve segment, i.e., the second straight segment, to form the fourth trajectory segment. Since the end of the third trajectory segment or the third spiral curve segment is tangent to the centerline of the target lane, the fourth trajectory segment is located on the centerline of the target lane.

[0107] The preset curve length of each spiral curve segment and the preset straight length of each straight line segment can be determined according to the lane width of the target lane. For example, it can be achieved by any of the following methods: a ratio-based mapping algorithm or a lookup table method.

[0108] In some implementations, after obtaining the first parking exit path, the parking exit path planning method may further include: if, during the process of the vehicle traveling along the first parking exit path, there is an obstacle within a preset range of the vehicle determined at the current trajectory point position, trajectory replanning is performed starting from the current trajectory point position.

[0109] Specifically, as the vehicle travels along the first parking path, each time it reaches a trajectory point on the first parking path, sensors continuously monitor the area around the vehicle for obstacles within a preset range to obtain the obstacle detection result corresponding to that trajectory point. When the vehicle moves to the current trajectory point on the first parking path, the control system with automatic parking function can obtain the obstacle detection result corresponding to that current trajectory point. If the obstacle detection result determines that there is an obstacle within the vehicle's preset range at that current trajectory point, the system will replan the trajectory starting from the current trajectory point.

[0110] For example, trajectory replanning can be achieved using any of the following methods: A* algorithm, Model Predictive Control (MPC) algorithm, graph search algorithm, etc.

[0111] Furthermore, if the bounding box corresponding to the end point of the first parking path contains obstacle information, the system redetermines the target parking position of the vehicle on the centerline of the target lane, and performs path planning based on the parking start position, target position, and minimum turning radius to obtain a second parking path. After obtaining the second parking path, the parking path planning method may further include: if, during the vehicle's travel along the second parking path, an obstacle is determined to exist within a preset range of the vehicle at a certain current trajectory point position in the second parking path, trajectory replanning is performed starting from that trajectory point position.

[0112] It should be noted that the preset range can be determined according to the vehicle's safety requirements, such as a range of 1 meter or 3 meters around the vehicle, etc., and this manual does not specify a specific range.

[0113] In some implementations, trajectory replanning starting from the current trajectory point position may include: replanning the trajectory based on the hybrid A-Star algorithm starting from the current trajectory point position.

[0114] Specifically, taking the current trajectory point position as the starting point (also known as the planning start node), with a step size L... H Perform point exploration, expanding n nodes (n in different directions) under the current planned node at a time. Calculate the state of the expanded nodes using the grid diagram. If the expanded node is not occupied by obstacles, it is a valid node. If all n expanded nodes are occupied by obstacles (falling into the obstacle grid), then perform point exploration again in other directions.

[0115] For a given valid node, calculate the cost C1 under the A* algorithm and the cost C2 under the Reeds-Shepp curve (RS) planning at that node. Denote the final cost at that valid node as C = max{C1, C2}. At the current planning node, identify the valid node among its extended nodes that has the minimum cost C, and use this valid node as the starting node for the next planning step. Repeat the above steps until the planning endpoint is reached. The planning focus can be dynamically determined using a hybrid A* algorithm.

[0116] In some implementations, if an obstacle is found within a preset range of the vehicle at the current trajectory point location while the vehicle is traveling along the first parking path, and trajectory replanning is determined to start from the current trajectory point location, this may include: if an obstacle is found within a preset range of the vehicle at the current trajectory point location while the vehicle is traveling along the first parking path and the corner of the vehicle is located within the range corresponding to the target lane, and trajectory replanning is determined to start from the current trajectory point location.

[0117] The phrase "the corner of the vehicle is located within the range corresponding to the target lane" refers to the corner of the vehicle crossing the parallel parking space line. In this manual, the parking path after the corner of the vehicle crosses the parallel parking space line is called the "post-exit path," and the path replanning after the corner of the vehicle crosses the parallel parking space line is called the "post-exit path planning."

[0118] Specifically, after the vehicle travels along the first parking path and crosses the parallel parking space boundary line at the corner of the front of the vehicle on the side away from the target lane, and enters the area corresponding to the target lane, the control system with automatic parking function obtains the obstacle detection result corresponding to the current trajectory point position. If the obstacle detection result determines that there is an obstacle within the vehicle's preset range at the current trajectory point position, the system will replan the trajectory starting from the current trajectory point position.

[0119] Furthermore, if the bounding box corresponding to the end point of the first parking path contains obstacle information, the system redetermines the target parking position of the vehicle on the centerline of the target lane, and performs path planning based on the parking start position, target position, and minimum turning radius to obtain a second parking path. After obtaining the second parking path, the parking path planning method may further include: if, during the process of the vehicle traveling along the second parking path, the corner of the vehicle is within the range corresponding to the target lane, and if an obstacle is determined to exist within a preset range of the vehicle at a certain current trajectory point position in the second parking path, trajectory replanning is performed starting from that trajectory point position.

[0120] For example, this specification provides a parking exit path planning method for parallel parking spaces. First, parking logic is determined based on the drivable area, and a parking exit path is planned. Then, during the vehicle's exit process according to the parking path, if an obstacle or obstructing vehicle is detected after the vehicle's front corner crosses the parallel parking space boundary line, a hybrid A* algorithm is used for post-exit path planning. This method has a higher success rate in path planning under complex scenarios such as those with obstacles. Furthermore, for different scenarios, the proposed path planning logic employs different path planning methods, which can reduce the path planning time required for automatic vehicle parking to a certain extent and improve planning efficiency. The main process of this method is as follows.

[0121] First, by default, path planning is performed using a combination of circular curves and spirals: (1) with Figure 3b Taking the parallel parking space and driveable area scenario shown as an example, continue to refer to... Figure 1c As shown, the minimum turning radius R of the vehicle is... min As the turning radius, the vehicle is currently located at the origin in the parallel parking space, with the turning center at O1(X). O1 Y O1 ), calculate in counterclockwise direction with R P The point C1 where the arc of radius R intersects the edge of the parking space. min It is an important parameter for RS curve, geometric curve, and hybrid A-satellite path planning, affecting the coordinate position of trajectory points.

[0122] (2) If C1 is within the drivable area, then calculate the segment with R. P Using the angle of an arc with radius , and with an arc length step of 0.05, the first arc trajectory point is calculated using the arc interpolation method. The length S at the end of the first arc trajectory is... e1 The preset curve length is replaced with a spiral segment to buffer the trajectory curvature, complete the vehicle's in-cabin path planning, and obtain the first trajectory segment.

[0123] (3) Taking the end point of the first segment of the trajectory as the starting point, add a line of length S after the first segment of the trajectory. e2A straight line (i.e., a preset straight line length) is used as the second segment of the trajectory, and this straight line is tangent to the end point of the first segment of the trajectory.

[0124] (4) Using the end point of the second trajectory as the starting point, construct the third trajectory. The third trajectory consists of a spiral curve segment, a circular arc curve segment, and a spiral curve segment connected in sequence. The first spiral curve segment is tangent to the second trajectory segment, and the second spiral curve segment is tangent to the line y = D. L / 2+ W R / 4. The lengths of both spiral curve segments are S. e1 The radius of the arc segment is the minimum turning radius R. min .

[0125] (5) The fourth segment of the trajectory is a straight line, starting from the end point of the third segment of the trajectory, and ending at E1.

[0126] (6) If C1 is not within the drivable area, then O0(X) O1 -Y O1 ) is the turning center, and the direction is counterclockwise. Calculate R. P1 (Turn radius of point P1 at the left rear corner), R P2 The intersection point F of the arc with radius (the turning radius of the right rear corner point P2) and the boundary of the drivable area. P1 F P2 The central angles corresponding to the two arcs are calculated separately. The smaller central angle is taken, and the arc interpolation method is used to calculate the arc trajectory points of the reversing segment with an arc length step of 0.05. Among them, P1 and P2 are the left rear corner point and right rear corner point of the vehicle, respectively. The subsequent trajectory segments are planned by completing the parking path of the vehicle using the above steps (1) to (5).

[0127] Secondly, if the vehicle's bounding box contains obstacle information (i.e., obstacle grids) at the endpoint E1, trajectory replanning is performed: (7) If the vehicle bounding box contains obstacle information (i.e., contains obstacle grids) at the endpoint E1, continue to refer to Figure 3c As shown, the target location E2 is redefined according to the aforementioned method, and then the parking exit path planning for the vehicle is completed using the above steps with target location E2 as the path endpoint.

[0128] Then, during the process of the vehicle parking out along the planned path, after exiting the garage (meaning after the vehicle's corner point P crosses the garage edge line), a hybrid A* algorithm is used to dynamically adjust the path: (8) After leaving the warehouse, take the end point of the first segment of the trajectory planned in step (2) above as the planning starting node.

[0129] (9) With step size L HPerform point exploration, expanding n nodes at a time under the current planned node. Calculate the state of the expanded nodes using the grid diagram. If an expanded node is not occupied by an obstacle grid, it is a valid node. If all expanded nodes are occupied by obstacle grids, then perform point exploration again in other directions.

[0130] (10) Calculate the cost C1 under the A-star algorithm planning and the cost C2 under the RS curve planning at a certain effective node. Record the final cost value at the node as C=max{C1, C2}. Take the effective node with the smallest C value under the current planning node as the starting node of the next planning.

[0131] (11) Repeat steps (9) to (10) until the planned endpoint is reached.

[0132] This specification also provides a parking exit path planning device for parallel parking spaces, which may include: a turning center determination module, an intersection point determination module, a path planning module, and a reversing target position determination module.

[0133] The turning center determination module is used to determine the first turning center on the side of the starting position of parking in the parallel parking space, which is closer to the target lane, based on the vehicle's minimum turning radius and the starting position of parking in the parallel parking space. The target lane is located on one of the long side lines of the parallel parking space. The line connecting the first turning center and the starting position of parking is perpendicular to the vehicle's centerline in the direction of the vehicle body.

[0134] The intersection point determination module is used to determine the intersection point on the straight line where the vehicle's front corner position point moves forward from the first turning center to the long side line of the parallel parking space; wherein, the front corner position point is located on the side of the vehicle away from the target lane in the parallel parking space.

[0135] The path planning module is used to obtain the drivable area from the parallel parking space to the target lane. If the intersection is located within the drivable area, the first parking path is obtained based on the parking start position, minimum turning radius, the angle corresponding to the path from the front corner of the vehicle to the intersection, and the lane width of the target lane.

[0136] The reversing target position determination module is used to determine the reversing target position of the vehicle in the parallel parking space if the intersection is not located in the drivable area, and to use the reversing target position as the starting position for the vehicle to exit the parallel parking space. It then returns to the step of determining the first turning center on the side of the starting position closer to the target lane based on the vehicle's minimum turning radius and the starting position for the vehicle to exit the parallel parking space.

[0137] Specific limitations regarding the berthing path planning device can be found in the limitations of the berthing path planning method described above, and will not be repeated here. Each module in the aforementioned berthing path planning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0138] This specification also provides a computer device, which may include a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the berthing path planning method in any of the foregoing embodiments.

[0139] This specification also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the outbound path planning method in any of the foregoing embodiments.

[0140] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0141] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0142] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0143] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0144] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0145] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for planning parking exit routes for parallel parking spaces, characterized in that, The method includes: Based on the vehicle's minimum turning radius and the vehicle's starting position for exiting the parallel parking space, a first turning center is determined on the side of the starting position closer to the target lane; wherein the target lane is located on one of the long side lines of the parallel parking space; the line connecting the first turning center and the starting position for exiting the parallel parking space is perpendicular to the vehicle's centerline in the vehicle's direction. The vehicle's front corner position is determined and moved forward from the first turning center to the intersection point on the straight line containing the long side of the parallel parking space; wherein, the front corner position is located on the side of the vehicle in the parallel parking space away from the target lane; Obtain the drivable area from the parallel parking space to the target lane. If the intersection is located within the drivable area, perform path planning based on the parking start position, the minimum turning radius, the angle corresponding to the path from the front corner of the vehicle to the intersection, and the lane width of the target lane to obtain the first parking path. If the intersection point is not located within the drivable area, determine the reversing target position of the vehicle in the parallel parking space; Using the reversing target position as the starting position for the vehicle to exit the parallel parking space, return to the step of determining the first turning center on the side of the starting position closer to the target lane based on the vehicle's minimum turning radius and the starting position for exiting the parallel parking space.

2. The method according to claim 1, wherein determining the reversing target position of the vehicle in the parallel parking space comprises: Based on the minimum turning radius and the parking start position, a second turning center symmetrical to the first turning center is determined on the side of the parking start position away from the target lane; Determine the minimum angle of motion for the left and right rear corner positions of the vehicle to move backward to the boundary of the drivable area with the second turning center as the center. Based on the minimum motion angle, the parking start position, and the second turning center, a reversing path is planned to determine the reversing target position.

3. The method according to claim 2, characterized in that, The step of planning the reversing path based on the minimum motion angle, the parking start position, and the second turning center to determine the reversing target position includes: Starting from the parking exit position and with the second turning center as the center, circular interpolation is performed in the reverse direction of the vehicle based on the minimum turning radius and the minimum movement angle to obtain the reversing path; The reversing target position is determined based on the end point of the reversing path.

4. The method according to claim 1, characterized in that, The starting position of the vehicle when it leaves the parallel parking space is the position of the center point of the rear axle of the vehicle. After obtaining the first departure path, the method further includes: If the vehicle contains obstacle information in the bounding box corresponding to the end of the first parking path, determine the boundary projection point of the obstacle boundary point in the obstacle information on the straight line where the specified long side line of the parallel parking space is located; wherein, the end of the first parking path is located on the center axis of the target lane. The target boundary point position is determined based on the minimum distance between the projection point of the parking start position on the straight line of the specified long side of the parallel parking space and the boundary projection point. The target parking position of the vehicle is determined on the center axis of the target lane based on the location of the target boundary point, the distance between the center point of the rear axle of the vehicle and the foremost point of the vehicle, and the preset interval distance. Based on the starting position of berthing, the target position, and the minimum turning radius, a second berthing path is obtained through path planning.

5. The method according to claim 1, characterized in that, The path planning based on the parking start position, the minimum turning radius, the angle corresponding to the path from the vehicle's front corner position to the intersection point, and the lane width of the target lane includes: Starting from the parking start position and taking the first turning center as the center, circular interpolation is performed on the direction of the vehicle's movement based on the minimum turning radius and the angle corresponding to the path from the front corner position point to the intersection point to obtain the first circular arc curve segment. The first arc curve segment with a preset curve length at the end furthest from the parking start position is replaced with a first spiral curve segment to obtain the first trajectory segment; wherein, the curvature of the starting point of the first spiral curve segment is the curvature of the first arc trajectory, and the curvature of the ending point of the first spiral curve segment is 0; the preset curve length is determined according to the lane width of the target lane. Taking the end point of the first trajectory segment as the starting point, a first straight line segment with a preset straight line length is determined along the direction of the end point of the first trajectory segment to obtain the second trajectory segment; wherein, the preset straight line length is determined according to the lane width of the target lane; Taking the end point of the second trajectory segment as the starting point, a third trajectory segment is determined based on the direction of the end point of the second trajectory segment; wherein, the third trajectory segment includes a second spiral curve segment connected to the second trajectory segment, a second circular arc curve segment connected to the second spiral curve segment, and a third spiral curve segment connected to the second circular arc curve segment; the second spiral curve segment and the third spiral curve segment have the preset curve length; the radius of the second circular arc curve segment is the minimum turning radius; Taking the end point of the third trajectory as the starting point, a second straight line segment with the preset straight line length is determined along the direction of the end point of the third trajectory to obtain a fourth trajectory; wherein, the fourth trajectory is located on the centerline of the target lane.

6. The method according to claim 1, characterized in that, After obtaining the first departure path, the method further includes: If an obstacle is found within a preset range of the vehicle at the current trajectory point location while the vehicle is traveling along the first parking path, the trajectory is replanned starting from the current trajectory point location.

7. The method according to claim 6, characterized in that, The trajectory replanning based on the current trajectory point position includes: Starting from the current trajectory point position, trajectory replanning is performed based on the hybrid A-star algorithm.

8. The method according to claim 6, characterized in that, If, during the process of the vehicle traveling along the first parking path, an obstacle is determined to exist within a preset range of the vehicle at the current trajectory point location, trajectory replanning is determined to start from the current trajectory point location, including: If the vehicle is traveling along the first parking path and the front corner of the vehicle is within the range corresponding to the target lane, and if an obstacle is found within a preset range of the vehicle at the current trajectory point, trajectory replanning is performed starting from the current trajectory point.

9. A parking exit route planning device for parallel parking spaces, characterized in that, The device includes: A turning center determination module is used to determine a first turning center on the side of the starting position of parking in a parallel parking space, close to the target lane, based on the vehicle's minimum turning radius and the vehicle's starting position in the parallel parking space; wherein the target lane is located on one of the long side lines of the parallel parking space; the line connecting the first turning center and the starting position of parking is perpendicular to the vehicle's body centerline. The intersection point determination module is used to determine the intersection point on the straight line where the front corner position of the vehicle moves forward with the first turning center as the center to the long side line of the parallel parking space; wherein, the front corner position point is located on the side of the vehicle in the parallel parking space away from the target lane; The path planning module is used to obtain the drivable area from the parallel parking space to the target lane. If the intersection is located within the drivable area, the module performs path planning based on the parking start position, the minimum turning radius, the angle corresponding to the path from the front corner of the vehicle to the intersection, and the lane width of the target lane to obtain the first parking path. The reversing target position determination module is used to determine the reversing target position of the vehicle in the parallel parking space if the intersection is not located within the drivable area, and to use the reversing target position as the parking start position of the vehicle in the parallel parking space, returning to the step of determining the first turning center on the side of the parking start position close to the target lane based on the vehicle's minimum turning radius and the parking start position of the vehicle in the parallel parking space.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images