Parking space inspection methods, media and vehicles

By identifying free areas in the target map that meet the parking space conditions, the problem of automatic parking without parking lines or reference obstacle corners is solved. Automatic parking recognition and path planning without parking lines or obstacle corners are realized, improving the capabilities of the automatic parking system.

CN118269937BActive Publication Date: 2026-11-10BYD CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202311227955.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-21
Publication Date
2026-11-10
Estimated Expiration
2043-09-21

AI Technical Summary

Technical Problem

In existing technologies, automatic parking systems require parking space lines or reference obstacle corners to identify parking spaces, making it difficult to achieve automatic parking in the absence of parking space lines or reference obstacle corners, thus increasing the difficulty of operation for users.

Method used

By identifying a free area without obstacles in the target map, and then identifying areas within that area that meet parking space conditions, parking spaces are identified using these conditions. This includes determining candidate parking areas and target parking spaces. Safe path planning is then performed by combining vehicle positioning and obstacle information to achieve automatic parking.

Benefits of technology

Even in situations where there are no parking space lines or reference obstacle corners, the vehicle can identify and automatically park, increasing its automatic parking capability and reducing the difficulty of operation for users.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118269937B_ABST
    Figure CN118269937B_ABST
Patent Text Reader

Abstract

This disclosure relates to a parking space detection method, medium, and vehicle, belonging to the field of vehicle technology. The method includes: determining a free area of ​​the vehicle in a target map, wherein the free area is an area in the target map that does not include obstacles; and determining an area in the free area that meets parking space conditions for parking. Thus, this disclosure determines an area in the free area that can be used for parking based on parking space conditions, enabling parking even without identifying parking space lines or reference obstacle corners, thereby increasing the vehicle's automatic parking capability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of vehicle technology, specifically to a parking space detection method, medium, and vehicle. Background Technology

[0002] In related technologies, automatic parking integrates ultrasonic radar and surround view perception to obtain information on parking spaces, parking space groups, and obstacles around the vehicle, thereby controlling automatic parking. However, surround view perception requires parking space lines to identify parking spaces and parking space groups, while ultrasonic radar requires reference corner points of obstacles around the parking space to identify the parking space. Summary of the Invention

[0003] To achieve the above objectives, this disclosure provides a parking space detection method, medium, and vehicle.

[0004] The first aspect of this disclosure provides a parking space detection method, comprising:

[0005] Determine the free area of ​​the vehicle in the target map, where the free area is a region in the target map that does not include obstacles;

[0006] Within the free area, areas that meet the parking space requirements are designated for parking.

[0007] Optionally, within the free area, determining an area that meets the parking space conditions for parking includes:

[0008] Within the free area, candidate parking areas that meet the parking space conditions are identified;

[0009] Within the candidate parking area, a target parking space is selected for parking.

[0010] Optionally, in the free area, determining candidate parking areas that meet the parking space conditions includes:

[0011] Within the free area, the boundary of the area that satisfies the parking space conditions is determined;

[0012] The area corresponding to the boundary of the area is determined as the candidate parking area.

[0013] Optionally, in the free area, determining the area boundary that satisfies the parking space conditions includes:

[0014] Within the free grid region, define region boundaries that satisfy at least the following conditions:

[0015] The first boundary, the second boundary, the third boundary, and the fourth boundary are all parallel to each other. The first boundary, the second boundary, and the third boundary are adjacent and all located on one side of the vehicle. The fourth boundary is located on the other side of the vehicle, opposite to the first boundary.

[0016] The lengths of the first boundary, the second boundary, the third boundary, and the fourth boundary satisfy a first threshold condition, and the distance between the target boundary among the first boundary, the second boundary, and the third boundary and the fourth boundary satisfies a second threshold condition.

[0017] Optionally, the first threshold condition includes:

[0018] The sum of the lengths of the first boundary, the second boundary, and the third boundary is greater than or equal to a first threshold, and the length of the fourth boundary is greater than or equal to the first threshold;

[0019] The second threshold condition includes:

[0020] The distance between the first target boundary and the fourth boundary is greater than or equal to a second threshold, the length of the first target boundary is greater than or equal to a third threshold, and the distance between the second target boundary and the fourth boundary is greater than or equal to a fourth threshold. The first target boundary is the boundary with the largest distance from the fourth boundary among the first boundary, the second boundary, and the third boundary. The second target boundary is any boundary among the first boundary, the second boundary, and the third boundary except for the first target boundary. The second threshold is greater than the fourth threshold, and the third threshold is less than the first threshold.

[0021] Optionally, in the free area, determining the area boundary that satisfies the parking space conditions includes:

[0022] Within the free region, define a region boundary that satisfies at least the following conditions:

[0023] The fifth, sixth, and seventh boundaries are all parallel to each other. The fifth and sixth boundaries are adjacent and both are located on one side of the vehicle. The seventh boundary is located on the other side of the vehicle, opposite to the first boundary.

[0024] The lengths between the fifth boundary, the sixth boundary, and the seventh boundary satisfy the third threshold condition, and the distances between the target boundary in the fifth and sixth boundaries and the seventh boundary satisfy the fourth threshold condition.

[0025] Optionally, the third threshold condition includes:

[0026] The sum of the lengths of the fifth boundary and the sixth boundary is greater than or equal to the fifth threshold, and the length of the seventh boundary is greater than or equal to the fifth threshold;

[0027] The fourth threshold condition includes:

[0028] The distance between the third target boundary and the seventh boundary is greater than or equal to the sixth threshold, the length of the third target boundary is greater than the seventh threshold, and the distance between the fourth target boundary and the seventh boundary is greater than or equal to the eighth threshold. The third target boundary is the boundary with the largest distance from the seventh boundary among the fifth and sixth boundaries, the fourth target boundary is the other boundary among the fifth and sixth boundaries besides the third target boundary, the seventh threshold is less than the fifth threshold, and the eighth threshold is greater than the sixth threshold.

[0029] Optionally, in the candidate parking area, determining the target parking space for parking includes:

[0030] Obtain vehicle location information and obstacle information;

[0031] Based on the positioning information and the obstacle information, a target parking space that meets the safety conditions is determined in the candidate parking area. The safety conditions are used to ensure that the vehicle does not collide with the obstacle corresponding to the obstacle information during the parking process.

[0032] Optionally, determining the target parking space that meets the safety conditions in the candidate parking area based on the positioning information and the obstacle information includes:

[0033] Based on the location information, the target parking area, and the obstacle information, a parking path is planned to obtain the target planned path. The target parking area is the candidate parking area whose distance from the vehicle is less than a distance threshold.

[0034] If the target planned path meets the safety conditions, the parking space in the target parking area shall be used as the target parking space.

[0035] Optionally, determining the target parking space that meets the safety conditions in the candidate parking area based on the positioning information and the obstacle information includes:

[0036] Based on the location information and the obstacle information, parking path planning is repeatedly performed on the candidate parking areas according to the distance between the candidate parking areas and the vehicle's location information from near to far, and it is determined whether the planned parking path meets the safety conditions, until the planned parking path meets the safety conditions, and the parking space in the candidate parking area that meets the safety conditions is taken as the target parking space.

[0037] Optionally, before determining the free area of ​​the vehicle in the target map, the method further includes:

[0038] Obtain the vehicle's speed information, location information, and obstacle information;

[0039] Based on the vehicle speed information, determine the region of interest;

[0040] Based on the location information and the obstacle information, a drivable area is determined;

[0041] A local grid map is determined based on the region of interest and the drivable area;

[0042] The target map is obtained based on a series of multiple frames of local raster maps.

[0043] Optionally, based on the vehicle speed information, the region of interest is determined, including:

[0044] The vehicle speed information and the performance of the vehicle sensors are input into a deep learning model to determine the region of interest. The vehicle sensors are used to acquire the positioning information and obstacle information.

[0045] Optionally, a local raster map is determined based on the region of interest and the drivable area, including:

[0046] The drivable region of interest is obtained by the union of the region of interest and the drivable region.

[0047] Based on the accuracy of the vehicle sensors, the granularity information of the grid is determined, and the vehicle sensors are used to acquire the positioning information and obstacle information;

[0048] The drivable region of interest is divided into grids based on the granularity information to obtain the local grid map.

[0049] Optionally, the target map is obtained based on a series of consecutive local raster maps, including:

[0050] The local raster maps of the target frame number are stitched together to obtain the intermediate raster map;

[0051] The intermediate raster map is updated using the local raster map of the next frame of the target frame to obtain the target map.

[0052] A second aspect of this disclosure provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the parking space detection method provided in the first aspect of this disclosure.

[0053] A third aspect of this disclosure provides a vehicle comprising:

[0054] A storage device, wherein a computer program is stored in the storage device;

[0055] A processing device for executing the computer program to implement the parking space detection method provided in the first aspect of this disclosure.

[0056] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0057] This disclosure identifies a free area excluding obstacles in a target map, and then determines areas within that free area that meet parking space conditions for parking. By determining parking spaces within the free area based on these conditions, parking can be achieved even without identifying parking lines or reference obstacle corners, thus increasing the vehicle's automatic parking capability.

[0058] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0059] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0060] Figure 1 It is a grid map diagram illustrating an automatic parking scheme based on relevant technologies.

[0061] Figure 2 This is another grid map illustration of an automatic parking solution based on related technologies.

[0062] Figure 3 It is a schematic diagram showing the parking trajectory of a vehicle for a perpendicular parking space, based on relevant technologies.

[0063] Figure 4 This is a flowchart illustrating a parking space detection method according to an exemplary embodiment.

[0064] Figure 5 This is a schematic diagram illustrating a method for determining candidate parking areas according to an exemplary embodiment.

[0065] Figure 6 It is a schematic diagram showing the parking trajectory of a vehicle for a horizontal parking space, based on relevant technologies.

[0066] Figure 7 This is a schematic diagram illustrating a parking trajectory planning process according to an exemplary embodiment.

[0067] Figure 8This is a schematic diagram illustrating a local raster map stitching process according to an exemplary embodiment.

[0068] Figure 9 This is a block diagram illustrating a parking space detection device according to an exemplary embodiment.

[0069] Figure 10 This is a block diagram illustrating a vehicle according to an exemplary embodiment.

[0070] Figure 11 This is a block diagram illustrating an apparatus for a parking space detection method according to an exemplary embodiment. Detailed Implementation

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0072] In related technologies, the target area in the navigation map is projected; or the vehicle position and parking space corner coordinates, as well as obstacles, are projected onto the grid map; or scene images collected by the surround view device and detection information collected by the ultrasonic radar sensor are received, and a grid map after data fusion is obtained based on the scene image and the detection information.

[0073] like Figure 1 and Figure 2 As shown, current automatic parking solutions utilize ultrasonic radar installed in the vehicle and a surround-view perception fusion scheme to obtain information about parking spaces, parking space groups, and obstacles around the vehicle, resulting in a grid map. Further, based on the parking space and its size, parking space attributes (horizontal, vertical, diagonal), and surrounding obstacle information, one or more parking trajectories are obtained that satisfy vehicle kinematics, dynamics, and collision-free safety requirements. The vehicle is then controlled to follow these parking trajectories to complete the parking maneuver. For example... Figure 3 As shown, the dashed box represents the vehicle's position change during the parking process, and the solid box represents the vehicle's position after parking. The vehicle's parking trajectory is ABCDEFG in sequence, where O1 is the center of the arc in segment AB, O2 is the center of the arc in segment BC, O3 is the center of the arc in segment CD, and 0, 1, 2, and 3 are the four corner points of the parking space.

[0074] However, surround-view perception requires parking lines to identify parking spaces and parking space groups, while ultrasonic radar requires reference obstacle corners around the parking space to identify it. In the absence of parking lines or reference obstacle corners, the user can select the location and orientation of the virtual parking space based on HMI (Human Machine Interface) prompts. However, this interaction method can also introduce unnecessary difficulty in parking operations, causing automatic parking to lose its fully automatic parking attribute.

[0075] Figure 4 This is a flowchart illustrating a parking space detection method according to an exemplary embodiment, such as... Figure 4 As shown, the parking space detection method used in the terminal includes the following steps.

[0076] S401. Determine the free area of ​​the vehicle in the target map. The free area is the area in the target map that does not include obstacles.

[0077] For example, the target map can be a raster map, a point cloud map, a vector map, a semantic map, etc. A raster map is a raster data file obtained through computer processing, directly converted from a vector digital line map, thus maintaining consistency with the basic scale topographic map in content, geometric accuracy, and color. Obstacles include moving and immovable obstacles. For example, in a parking lot, there are fixed obstacles such as pillars, cones, padlocks, wheel stops, crash barriers, and reversing barriers, as well as moving obstacles such as pedestrians, animals, and other vehicles. The target map can be obtained through vehicle surround view and ultrasonic radar fusion.

[0078] For example, areas in the target map that include obstacle information can be designated as occupied areas, while areas that do not include obstacle information can be called free areas.

[0079] S402. In the free area, determine the area that meets the parking space conditions for parking.

[0080] For example, parking space conditions can be used to determine the area that meets the space requirements for vehicle parking, specifically, the minimum space required for parking. Based on the parking space conditions, an area within a free area that meets the minimum space requirements for parking can be used for parking; that is, a target parking space exists within the area that meets the parking space conditions, eliminating the need for surround-view perception or ultrasonic sensors to detect parking spaces. Therefore, whether it is a marked parking space, an unmarked parking space, or an irregular parking area, by determining an area within a free area that meets the parking space conditions, a parking space can be identified within that area, and the parking process can be completed.

[0081] For example, in a free area, after determining an area that meets the parking space conditions for parking, the vehicle can be controlled to park within the parking space of that area. Through parking path planning, parking speed planning, and obstacle avoidance planning, a global parking path from the vehicle's current position to the parking space can be generated based on the free area. Local parking trajectories are generated in real time by combining information on moving obstacles to complete the parking action. The parking path planning scheme can include circular arcs with straight lines, RS curves, Dubins curves, spiral curves, hybrid A* algorithms, etc., to generate a safe and collision-free global parking path for single-step or multi-step parking. On the global parking path, the recommended longitudinal speed and recommended acceleration information for each global parking point can be determined using a distance-relative speed map based on the parking key points and the parking target point.

[0082] This disclosure identifies a free area excluding obstacles in a target map, and then determines areas within that free area that meet parking space conditions for parking. By determining parking spaces within the free area based on these conditions, parking can be achieved even without identifying parking lines or reference obstacle corners, thus increasing the vehicle's automatic parking capability.

[0083] As an optional embodiment, in the free area, determining an area that meets the parking space conditions for parking includes:

[0084] Within the free area, identify candidate parking areas that meet the parking space requirements;

[0085] Within the candidate parking area, the target parking space is selected for parking.

[0086] For example, a candidate parking area is a parking area that meets the parking space requirements. A candidate parking area includes the target parking space and the movement space during the parking process. There can be one or more candidate parking areas. If there are multiple candidate parking areas, the target parking area can be determined from the candidate parking areas, and then the target parking space can be determined from the target parking area. When there are multiple candidate parking areas, each candidate parking area can be assigned an ID number according to its order, such as candidate parking area 1, candidate parking area 2, etc., to facilitate differentiation between candidate parking areas.

[0087] As an optional embodiment, in the free area, candidate parking areas that meet the parking space conditions are determined, including:

[0088] Within the free area, define the boundary of the area that meets the parking space requirements;

[0089] The area corresponding to the boundary of the area is identified as a candidate parking area.

[0090] For example, the area boundary can be a partial boundary of a free area, or it can be the boundary of a target area included within a free area. Based on the area boundary that meets the parking space conditions, the area corresponding to that boundary can be identified as a candidate parking area.

[0091] As an optional embodiment, in the free area, determining the area boundary that satisfies the parking space conditions includes:

[0092] Within the free region, define the region boundaries that at least satisfy the following conditions:

[0093] The first boundary, the second boundary, the third boundary, and the fourth boundary are all parallel to each other. The first boundary, the second boundary, and the third boundary are adjacent and all located on one side of the vehicle. The fourth boundary is located on the opposite side of the vehicle.

[0094] The lengths of the first boundary, the second boundary, the third boundary, and the fourth boundary satisfy the first threshold condition, and the distance between the target boundary in the first boundary, the second boundary, the third boundary, and the fourth boundary satisfies the second threshold condition.

[0095] For example, in the absence of parking lines or parking space groups, a parking area can be determined within a free area based on parking space conditions, such as... Figure 5 As shown, the candidate parking area can be a T-shaped area. In this case, the boundaries of the candidate parking area can include the upper boundary, left boundary, right boundary, upper left boundary, upper right boundary, and lower boundary. The first, second, and third boundaries correspond to the upper, upper left, and upper right boundaries, respectively, while the fourth boundary corresponds to the lower boundary. If the lengths of the first, second, third, and fourth boundaries satisfy a first threshold condition, and the distance between the target boundary (one of the first, second, and third boundaries) and the fourth boundary satisfies a second threshold condition, then the area formed by the first, second, third, and fourth boundaries can be considered a candidate parking area.

[0096] Understandably, once the upper boundary, upper left boundary, upper right boundary, and lower boundary of the area are determined, the left boundary of the area can be obtained by connecting the left end of the upper boundary and the right end of the upper left boundary, the right end of the right boundary can be obtained by connecting the right end of the upper boundary and the upper right boundary, and the left end of the upper left boundary and the lower boundary, as well as the right end of the upper right boundary and the lower boundary, to finally obtain a candidate parking area.

[0097] As an optional embodiment, the first threshold condition includes:

[0098] The sum of the lengths of the first boundary, the second boundary, and the third boundary is greater than or equal to the first threshold, and the length of the fourth boundary is greater than or equal to the first threshold;

[0099] The second threshold condition includes:

[0100] The distance between the first target boundary and the fourth boundary is greater than or equal to the second threshold, the length of the first target boundary is greater than or equal to the third threshold, and the distance between the second target boundary and the fourth boundary is greater than or equal to the fourth threshold. The first target boundary is the boundary with the largest distance from the fourth boundary among the first, second, and third boundaries. The second target boundary is any boundary among the first, second, and third boundaries except the first target boundary. The second threshold is greater than the fourth threshold, and the third threshold is less than the first threshold.

[0101] For example, the first target boundary is the boundary among the first, second, and third boundaries that has the largest vertical distance to the fourth boundary. The first target boundary can be the upper boundary of the region. The second target boundary is any two of the first, second, and third boundaries other than the first target boundary. The distance between either of these two boundaries and the fourth boundary is greater than or equal to a fourth threshold. The fourth boundary is the lower boundary of the region. Specifically, the fourth threshold can be used to determine whether the distance between the second and fourth target boundaries meets the longitudinal dimension requirements for the parking space; the third threshold can be used to determine whether the width of the first target boundary meets the lateral dimension requirements for the parking space; the second threshold can be used to determine whether the distance between the first and fourth target boundaries meets both the longitudinal dimension requirements for the parking space and the longitudinal dimension requirements for the parking space; the first threshold is used to determine whether the sum of the lengths of the first, second, and third boundaries, and the length of the fourth boundary, both meet the lateral dimension requirements for the parking space and the lateral dimension requirements for the parking space. (Refer to...) Figure 5 The horizontal direction can be understood as the X-axis of the vehicle coordinate system, and the vertical direction as the Y-axis. The vehicle coordinate system has its origin at the center of the rear axle, with the vehicle's direction of travel as the X-axis and the horizontal / vertical X-axis as the Y-axis. Since the horizontal and vertical dimension requirements may differ for different types of parking spaces, the relationships between the second and third thresholds, the second and fourth thresholds, or the first and third thresholds, and the first and fourth thresholds are not limited and can be set according to specific circumstances.

[0102] Specifically, refer to Figure 5The above solution can be expressed as: H+I+J≥N1, K≥N1, I≥N3, L≥N2, M≥N4, N2>N4, N3<N1. Wherein, I is the upper boundary of the region, H is the upper left boundary of the region, J is the upper right boundary of the region, I can be used as the first target boundary, H or J can be used as the second target boundary, K is the fourth boundary, L is the perpendicular line between the first target boundary and the fourth boundary, M is the perpendicular line between the second target boundary and the fourth boundary, N1 is the first threshold, N2 is the second threshold, N3 is the third threshold, and N4 is the fourth threshold.

[0103] As shown in Figure 5 , if the No.1 region does not meet the parking space condition and the No.2 region meets the parking space condition, the No.1 region will not be taken as a candidate parking region, and the No.2 region will be taken as a candidate parking region.

[0104] As an optional embodiment, determining the region boundary that meets the parking space condition in the free region includes:

[0105] Determining a region boundary in the free region that at least meets the following conditions:

[0106] The fifth boundary, the sixth boundary and the seventh boundary are all parallel to each other, the fifth boundary and the sixth boundary are adjacent and both located on one side of the vehicle, and the seventh boundary is located on the other side of the vehicle opposite to the side;

[0107] The lengths between the fifth boundary, the sixth boundary and the seventh boundary satisfy the third threshold condition, and the distance between the target boundary among the fifth boundary and the sixth boundary and the seventh boundary satisfies the fourth threshold condition.

[0108] For example, the parking space condition can be set without considering the upper left boundary or upper right boundary of the aforementioned candidate parking region, for example, the parking space is against a wall, or refer to Figure 6 , where L represents the lateral dimension of the horizontal parking space, only considering that the vehicle travels forward from a position horizontally parallel to the parking space, the parking motion space in the motion space and the region corresponding to the parking space are L-shaped regions. In this case, the candidate parking region may include a first upper region boundary, a second upper region boundary, a left region boundary, a first right region boundary, a second right region boundary, and a lower region boundary.

[0109] It can be understood that after the first upper region boundary, the second upper region boundary and the lower region boundary are determined, the left region boundary can be obtained by connecting the left end of the first upper region boundary and the lower region boundary, the first right region boundary can be obtained by connecting the right end of the first upper region boundary and the left end of the second upper region boundary, and the second right region boundary can be obtained by connecting the right end of the second upper region boundary and the lower region boundary, so as to finally obtain a candidate parking region.

[0110] As an optional embodiment, the third threshold condition includes:

[0111] The sum of the lengths of the fifth and sixth boundaries is greater than or equal to the fifth threshold, and the length of the seventh boundary is greater than or equal to the fifth threshold;

[0112] The fourth threshold condition includes:

[0113] The distance between the third target boundary and the seventh boundary is greater than or equal to the sixth threshold, the length of the third target boundary is greater than the seventh threshold, and the distance between the fourth target boundary and the seventh boundary is greater than or equal to the eighth threshold. The third target boundary is the boundary with the largest distance from the seventh boundary among the fifth and sixth boundaries, and the fourth target boundary is the other boundary among the fifth and sixth boundaries besides the third target boundary. The seventh threshold is less than the fifth threshold, and the eighth threshold is greater than the sixth threshold.

[0114] For example, the third target boundary is the boundary with the largest vertical distance to the seventh boundary among the fifth and sixth boundaries. The third target boundary can be the first upper boundary of the region. The fourth target boundary is the other boundary among the fifth and sixth boundaries besides the third target boundary. The fourth target boundary can be the second upper boundary of the region, and the seventh boundary is the lower boundary of the region. Specifically, the fifth threshold can be used to determine that the sum of the lengths of the fifth and sixth boundaries, and the length of the seventh boundary, both satisfy the lateral dimension requirements for the parking space and the lateral dimension requirements for the parking space. The sixth threshold can be used to determine that the distance between the third target boundary and the seventh boundary satisfies the longitudinal dimension requirements for the parking space and the longitudinal dimension requirements for the parking space. The seventh threshold can be used to determine that the length of the third target boundary satisfies the lateral dimension requirements for the parking space. The eighth threshold can be used to determine that the distance between the fourth target boundary and the seventh boundary satisfies the longitudinal dimension requirements for the parking space. Since the lateral and longitudinal dimension requirements may differ for different types of parking spaces, the relationships between the fifth and sixth thresholds, the fifth and eighth thresholds, or the seventh and eighth thresholds are not limited and can be set according to specific circumstances. (Refer to...) Figure 5 The horizontal direction can be understood as the X-axis direction of the vehicle coordinate system, and the vertical direction can be understood as the Y-axis direction of the vehicle coordinate system. The vehicle coordinate system takes the center of the rear axle of the vehicle as the origin, the direction of vehicle travel as the X-axis, and the horizontal and vertical X-axis direction as the Y-axis.

[0115] Specifically, the first, second, third, fourth, fifth, sixth, seventh, and eighth thresholds can all be set according to actual needs, such as based on parking function specifications and vehicle dimensions. Parking function specifications define the shape and dimensions of parking spaces with different attributes. Specifically, the minimum width of a perpendicular parking space is vehicle width + 0.6m, and the minimum depth is greater than or equal to 3.5m; the minimum length of a horizontal parking space is vehicle length + 0.7m, and the minimum depth is 2m; the angle of an angled parking space with parking lines ranges from 45° to 60°, with a minimum width of 2.8m and a minimum depth of 5m. (Refer to...) Figure 3 and Figure 6 , Figure 3 The parking spaces shown are perpendicular parking spaces. Figure 6 The parking spaces shown are level parking spaces.

[0116] As an optional embodiment, determining a target parking space for parking within a candidate parking area includes:

[0117] Obtain vehicle location information and obstacle information;

[0118] Based on the location information and obstacle information, a target parking space that meets the safety conditions is determined in the candidate parking area. The safety conditions are used to ensure that the vehicle does not collide with the obstacle corresponding to the obstacle information during the parking process.

[0119] For example, obstacle information refers to the parameters of an obstacle. When the obstacle is a moving obstacle, the obstacle information includes obstacle coding information, obstacle position information, heading angle information, attribute information, size information, confidence level, lateral and longitudinal velocities, lateral and longitudinal acceleration, etc. When the obstacle is a stationary obstacle, the obstacle information includes obstacle position information, obstacle size information, etc. Based on the vehicle's positioning information and obstacle information, a target parking space meeting safety conditions can be determined within the candidate parking area for parking. These safety conditions ensure that the vehicle does not collide with the obstacle corresponding to the obstacle information during parking; that is, the vehicle either does not collide with the obstacle during parking, or may collide with the obstacle but can avoid it.

[0120] For example, if there are no obstacles in the candidate parking area, a safe and collision-free global parking path can be generated using common parking path planning algorithms, including circular arc plus straight line, RS curve, Dubins curve, spiral curve, and hybrid A* algorithm. On the global parking path, the longitudinal recommended speed and recommended acceleration information of each global parking point are determined by the distance-relative speed map based on the parking key points and parking target points.

[0121] For example, if an obstacle moves at a certain speed and heading angle, it indicates that the obstacle is a moving obstacle. By predicting the kinematic trajectory of the obstacle, we can obtain the points on the global parking path that the obstacle will pass through in the future. If it is predicted that the vehicle will not collide with the moving obstacle at the recommended speed of the global parking path, the vehicle can park normally. If it is predicted that the vehicle will collide with the moving obstacle, the longitudinal speed of the global parking path is replanned according to the future collision time and collision point.

[0122] For example, if an obstacle is positioned on the global parking path, the global parking path and longitudinal velocity are replanned based on the location of the moving obstacle. Figure 7 As shown, global parking trajectory 1 can be updated to global parking trajectory 2, and local parking planning can be further completed. If an obstacle is in front of the candidate parking area and blocks all possible global parking paths, parking in the candidate parking area is abandoned, and the search for the next candidate parking area continues.

[0123] As an optional embodiment, determining a target parking space that meets safety conditions within a candidate parking area based on location information and obstacle information includes:

[0124] Based on the location information, target parking area and obstacle information, parking path planning is performed to obtain the target planned path. The target parking area is a candidate parking area whose distance from the vehicle is less than a distance threshold.

[0125] If the planned path meets safety requirements, parking spaces within the target parking area will be designated as target parking spaces.

[0126] For example, the distance threshold can be set according to the actual situation; for instance, it can be set to 2m. The parking route and speed of the vehicle in the candidate parking area are planned, and it is determined whether the planned path will collide with obstacles. If no collision occurs, the safety conditions can be met, and the parking space in the candidate parking area can be identified as the target parking space. The parking route planning here includes planning both the parking route and the parking speed.

[0127] As an optional embodiment, determining a target parking space that meets safety conditions within a candidate parking area based on location information and obstacle information includes:

[0128] Based on the location information and obstacle information, parking path planning is repeatedly performed on the candidate parking areas in order of distance from the vehicle's location information, from near to far. The planned parking path is then used to determine whether it meets the safety conditions. This process continues until the planned parking path meets the safety conditions, at which point the parking space in the candidate parking area that meets the safety conditions is taken as the target parking space.

[0129] For example, based on the distance between candidate parking areas and the vehicle's location information, from closest to furthest, path planning is performed from the candidate parking area closest to the vehicle's current location. It is then determined whether the planned parking path meets safety conditions, i.e., whether a collision with an obstacle will occur. For instance, if candidate parking areas are sorted from shortest to longest distance from the vehicle, and all planned parking paths for the first candidate parking area (ranked first in distance) are blocked by obstacles, a second candidate parking area (ranked second in distance) can be searched. It is then determined whether the planned parking path for the second candidate parking area is blocked by an obstacle and whether it can be avoided. If it is not blocked by an obstacle, or if a parking path is blocked but can be replanned to avoid the obstacle, the parking space in the second candidate parking area can be used as the target parking space.

[0130] In related technologies, when a vehicle encounters a potential collision obstacle in the parking path, the vehicle stops moving and suspends parking. If the waiting time exceeds a certain threshold, the vehicle exits automatic parking and prompts for manual intervention, making it impossible to continue the automatic parking process even when the parking space meets the vehicle's obstacle avoidance planning requirements. However, the embodiments of this disclosure consider different scenarios that may occur between obstacles and the vehicle, and design corresponding obstacle avoidance strategies to minimize the likelihood of the vehicle exiting automatic parking and prompting for manual intervention due to obstacles in the parking path. This allows the automatic parking process to continue even when the parking space meets the vehicle's obstacle avoidance planning requirements.

[0131] As an optional embodiment, before determining the free area of ​​the vehicle in the target map, the method further includes:

[0132] Obtain vehicle speed information, location information, and obstacle information;

[0133] Determine the area of ​​interest based on vehicle speed information;

[0134] Based on location information and obstacle information, determine the drivable area;

[0135] Determine the local grid map based on the region of interest and the drivable area;

[0136] The target map is obtained from a series of multiple frames of local raster maps.

[0137] For example, vehicle speed information can be obtained from the vehicle's speed sensor. Vehicle positioning information can be calculated using four-wheel wheel speed pulses and steering wheel angle based on Ackermann steering kinematics. Higher-spec vehicles can also integrate automotive-grade high-precision IMUs (Inertial Measurement Units) to obtain even more precise positioning information, including the vehicle's position and heading angle. Obstacle information can be obtained through surround-view perception and ultrasonic radar fusion. This obstacle information includes obstacle coding, location, heading angle, attribute information, size information, confidence level, lateral and longitudinal velocities, lateral and longitudinal accelerations, etc. Based on the obstacle information and the performance of the vehicle's sensors, the drivable area is further determined. This drivable area is defined as the range between the maximum range that the sensors can perceive at the vehicle's current position and the perceived boundaries of surrounding obstacles (walls, railings, etc.). The smaller of these two ranges constitutes the drivable area. Figure 8 As shown, the drivable area boundary is the obstacle boundary. The region of interest (ROI) is a selected area of ​​the image that is the focus of image analysis; this area is delineated for further processing. In machine vision and image processing, outlining the area to be processed from the image using rectangles, circles, ellipses, irregular polygons, etc., can reduce processing time and increase accuracy.

[0138] For example, the target map is obtained as a raster map through multiple consecutive frames of local raster maps. One method of raster division is to define a raster as occupied if it includes obstacle information; another method is to define a raster as occupied if an obstacle occupies more than half of its area, and as free if an obstacle occupies less than half. The area composed of free rasters is the free region, and the area composed of occupied rasters is the region occupied by obstacles. Since the vehicle's image acquisition distance based on surround-view perception and ultrasonic radar is limited, a local raster map is determined based on the region of interest and drivable area of ​​a single frame image. Multiple consecutive frames of local raster maps can be stitched and / or fused to obtain the target map.

[0139] As an optional embodiment, determining the region of interest based on vehicle speed information includes:

[0140] Vehicle speed information and the performance of vehicle sensors are input into a deep learning model to determine the region of interest. The vehicle sensors are used to obtain location information and obstacle information.

[0141] For example, the region of interest (ROI) is determined based on a combination of the performance of surround-view perception, ultrasonic radar, and vehicle speed. Here, a deep learning model, such as a DCNN deep convolutional neural network model, is used to determine the ROI. Specifically, the better the performance of the surround-view perception and ultrasonic radar, the larger the determined ROI; the worse the performance of the surround-view perception and ultrasonic radar, the smaller the determined ROI. Similarly, the greater the vehicle's lateral and longitudinal speeds, the larger the ROI; the smaller the lateral and longitudinal speeds, the smaller the ROI.

[0142] As an optional embodiment, a local grid map is determined based on the region of interest and the drivable area, including:

[0143] The drivable region of interest is obtained by the union of the region of interest and the drivable region.

[0144] Based on the accuracy of the vehicle sensors, the granularity of the grid is determined. The vehicle sensors are used to acquire positioning information and obstacle information.

[0145] The drivable region of interest is divided into grids based on granularity information to obtain a local grid map.

[0146] For example, by finding the union of the drivable area and the region of interest, the drivable region of interest can be obtained. Then, based on the obstacle perception accuracy of the surround view and ultrasonic radar, and the fusion positioning accuracy, the granularity of the grid is determined to divide the drivable region of interest into grids. Figure 8 As shown, on the boundary of the drivable area, if the area occupied by a unit grid within the drivable area boundary is greater than the area outside the drivable area boundary, it is assigned to the local grid map; otherwise, it is assigned to the area outside the drivable area boundary. Then, after dividing the drivable region of interest into grids, obstacle information is projected onto the grid map to form occupied grids, while grids without obstacle information are free grids, resulting in the local grid map. Specifically, within the drivable region of interest, if the area occupied by the obstacle boundary within a unit grid is greater than the unoccupied area within that unit grid, it is assigned to an occupied grid; otherwise, it is assigned to a free grid.

[0147] As an optional embodiment, obtaining the target raster map based on consecutive multi-frame local raster maps includes:

[0148] The local raster maps of the target frame number are stitched together to obtain the intermediate raster map;

[0149] The intermediate raster map is updated using the local raster map of the next frame of the target frame to obtain the target raster map.

[0150] For example, the target number of frames can be set according to the actual situation, such as 2 frames. For each frame of image data acquired by the vehicle, information such as the drivable region of interest, obstacle information, drivable area boundaries, local grid map, occupied grid and free grid is determined to obtain a local grid map. By stitching the local grid maps of the target number of frames together in time sequence, an intermediate grid map can be obtained.

[0151] For example, when the target map is a raster map, the intermediate raster maps can be updated using the next local raster map corresponding to the target frame number. Specifically, the first frame of the local raster map for the target frame number can be deleted, and the next local raster map for the target frame number can be stitched together with the last frame of the local raster map for the target frame number to obtain the target map. For example, if the target frame number is 2 frames, after the vehicle obtains the 3rd local raster map, the first frame of the first 2 local raster maps is deleted, and the 3rd local raster map is stitched together with the 2nd local raster map to obtain the target raster map. Alternatively, the last local raster map for the target frame number can be directly stitched together with the next local raster map for the target frame number to obtain a more complete target raster map. For example, if the target frame number is 3 frames, after the vehicle obtains the 4th local raster map, the 4th local raster map is stitched together with the 3rd local raster map.

[0152] See Figure 8 The process involves stitching together the global raster map from the previous frame and the local raster map from the current frame. The global raster map from the previous frame is the intermediate raster map obtained by stitching together the local raster maps of the target frame, and the local raster map of the current frame is the local raster map of the next frame of the target frame. Furthermore, the stitching process considers the possibility of overlapping stitching areas. That is, if a complete raster is not formed during the local raster division of the previous frame, the length of the incomplete raster can be stored and used as the basis for raster division and stitching in the current frame.

[0153] Figure 9 This is a block diagram illustrating a parking space detection device according to an exemplary embodiment, such as... Figure 1 As shown, the parking space detection device includes the following modules.

[0154] The first determining module 910 is used to determine the free area of ​​the vehicle in the target map, where the free area is the area in the target map that does not include obstacles.

[0155] The second determining module 920 is used to determine, within the free area, an area that meets the parking space conditions for parking.

[0156] As an optional embodiment, the second determining module 920 includes:

[0157] The first determining submodule is used to determine candidate parking areas that meet the parking space conditions in the free area;

[0158] The second determination submodule is used to determine the target parking space for parking within the candidate parking area.

[0159] As an optional embodiment, the first determining submodule includes:

[0160] The third determination submodule is used to determine the boundary of the area that meets the parking space conditions in the free area;

[0161] The fourth determination submodule is used to determine the area corresponding to the area boundary as a candidate parking area.

[0162] As an optional embodiment, the third determining submodule is specifically used for:

[0163] Within the free region, define the region boundaries that at least satisfy the following conditions:

[0164] The first boundary, the second boundary, the third boundary, and the fourth boundary are all parallel to each other. The first boundary, the second boundary, and the third boundary are adjacent and all located on one side of the vehicle. The fourth boundary is located on the opposite side of the vehicle.

[0165] The lengths of the first boundary, the second boundary, the third boundary, and the fourth boundary satisfy the first threshold condition, and the distance between the target boundary in the first boundary, the second boundary, the third boundary, and the fourth boundary satisfies the second threshold condition.

[0166] As an optional embodiment, the first threshold condition includes:

[0167] The sum of the lengths of the first boundary, the second boundary, and the third boundary is greater than or equal to the first threshold, and the length of the fourth boundary is greater than or equal to the first threshold;

[0168] The second threshold condition includes:

[0169] The distance between the first target boundary and the fourth boundary is greater than or equal to the second threshold, the length of the first target boundary is greater than or equal to the third threshold, and the distance between the second target boundary and the fourth boundary is greater than or equal to the fourth threshold. The first target boundary is the boundary with the largest distance from the fourth boundary among the first, second, and third boundaries. The second target boundary is any boundary among the first, second, and third boundaries except the first target boundary. The second threshold is greater than the fourth threshold, and the third threshold is less than the first threshold.

[0170] As an optional embodiment, the third determining submodule is specifically used for:

[0171] Within the free region, define the region boundaries that at least satisfy the following conditions:

[0172] The fifth, sixth, and seventh boundaries are all parallel to each other. The fifth and sixth boundaries are adjacent and both are located on one side of the vehicle, while the seventh boundary is located on the opposite side of the vehicle.

[0173] The lengths between the fifth, sixth, and seventh boundaries satisfy the third threshold condition, and the distances between the target boundaries in the fifth and sixth boundaries and the seventh boundary satisfy the fourth threshold condition.

[0174] As an optional embodiment, the third threshold condition includes:

[0175] The sum of the lengths of the fifth and sixth boundaries is greater than or equal to the fifth threshold, and the length of the seventh boundary is greater than or equal to the fifth threshold;

[0176] The fourth threshold condition includes:

[0177] The distance between the third target boundary and the seventh boundary is greater than or equal to the sixth threshold, the length of the third target boundary is greater than the seventh threshold, and the distance between the fourth target boundary and the seventh boundary is greater than or equal to the eighth threshold. The third target boundary is the boundary with the largest distance from the seventh boundary among the fifth and sixth boundaries, and the fourth target boundary is the other boundary among the fifth and sixth boundaries besides the third target boundary. The seventh threshold is less than the fifth threshold, and the eighth threshold is greater than the sixth threshold.

[0178] As an optional embodiment, the second determining submodule includes:

[0179] The information acquisition module is used to acquire the vehicle's location information and obstacle information;

[0180] The fifth determination submodule is used to determine the target parking space that meets the safety conditions in the candidate parking area based on the positioning information and obstacle information. The safety conditions are used to ensure that the vehicle does not collide with the obstacle corresponding to the obstacle information during the parking process.

[0181] As an optional embodiment, the fifth determining submodule is specifically used for:

[0182] Based on the location information, target parking area and obstacle information, parking path planning is performed to obtain the target planned path. The target parking area is a candidate parking area whose distance from the vehicle is less than a distance threshold.

[0183] If the planned path meets safety requirements, parking spaces within the target parking area will be designated as target parking spaces.

[0184] As an optional embodiment, the fifth determining submodule is specifically used for:

[0185] Based on the location information and obstacle information, parking path planning is repeatedly performed on the candidate parking areas in order of distance from the vehicle's location information, from near to far. The planned parking path is then used to determine whether it meets the safety conditions. This process continues until the planned parking path meets the safety conditions, at which point the parking space in the candidate parking area that meets the safety conditions is taken as the target parking space.

[0186] As an optional embodiment, the parking space detection device further includes:

[0187] The information acquisition module is used to acquire vehicle speed information, location information, and obstacle information;

[0188] The region of interest determination module is used to determine the region of interest based on vehicle speed information;

[0189] The drivable area determination module is used to determine the drivable area based on location information and obstacle information;

[0190] The local raster map determination module is used to determine a local raster map based on the region of interest and the drivable area;

[0191] The target map acquisition module is used to obtain the target map based on multiple consecutive frames of local raster maps.

[0192] As an optional embodiment, the region of interest determination module is specifically used for:

[0193] Vehicle speed information and the performance of vehicle sensors are input into a deep learning model to determine the region of interest. The vehicle sensors are used to obtain location information and obstacle information.

[0194] As an optional embodiment, the local grid map determination module is specifically used for:

[0195] The drivable region of interest is obtained by the union of the region of interest and the drivable region.

[0196] Based on the accuracy of the vehicle sensors, the granularity of the grid is determined. The vehicle sensors are used to acquire positioning information and obstacle information.

[0197] The drivable region of interest is divided into grids based on granularity information to obtain a local grid map.

[0198] As an optional embodiment, the target map acquisition module is specifically used for:

[0199] The local raster maps of the target frame number are stitched together to obtain the intermediate raster map;

[0200] The intermediate raster map is updated using the local raster map of the next frame of the target frame to obtain the target map.

[0201] Regarding the parking space detection device in the above embodiments, the specific way in which each module performs its operation has been described in detail in the embodiments related to the parking space detection method, and will not be elaborated here.

[0202] This disclosure also provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the parking space detection method provided in the embodiments of this disclosure.

[0203] This disclosure also provides a vehicle, including:

[0204] A storage device containing computer programs;

[0205] A processing device is provided to execute a computer program to implement the parking space detection method provided in the embodiments of this disclosure.

[0206] Figure 10 This is a block diagram illustrating a vehicle 1000 according to an exemplary embodiment. For example, vehicle 1000 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicle. Vehicle 1000 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0207] Reference Figure 10 The vehicle 1000 may include various subsystems, such as an infotainment system 1010, a perception system 1020, a decision control system 1030, a drive system 1040, and a computing platform 1050. The vehicle 1000 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and component of the vehicle 1000 can be interconnected via wired or wireless means.

[0208] In some embodiments, the infotainment system 1010 may include a communication system, an entertainment system, and a navigation system, etc.

[0209] The perception system 1020 may include several types of sensors for sensing information about the environment surrounding the vehicle 1000. For example, the perception system 1020 may include a global positioning system (which may be a GPS system, a BeiDou system, or another positioning system), an inertial measurement unit (IMU), a lidar, a millimeter-wave radar, an ultrasonic radar, and a camera device.

[0210] The decision control system 1030 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0211] The drive system 1040 may include components that provide powered motion to the vehicle 1000. In one embodiment, the drive system 1040 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0212] Some or all of the functions of the vehicle 1000 are controlled by a computing platform 1050. The computing platform 1050 may include at least one processor 1051 and a first memory 1052, the processor 1051 being able to execute instructions 1053 stored in the first memory 1052.

[0213] The processor 1051 can be any conventional processor, such as a commercially available CPU. The processor may also include, for example, a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), a System on Chip (SOC), an Application Specific Integrated Circuit (ASIC), or a combination thereof.

[0214] The first memory 1052 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0215] In addition to instruction 1053, the first memory 1052 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in the first memory 1052 can be used by the computing platform 1050.

[0216] In this embodiment of the disclosure, the processor 1051 may execute instruction 1053 to complete all or part of the steps of the above-described parking space detection method.

[0217] Figure 11 This is a block diagram illustrating an apparatus 1100 for a parking space detection method according to an exemplary embodiment. For example, apparatus 1100 may be provided as a server. (Refer to...) Figure 11The device 1100 includes a processing component 1122, which further includes one or more processors, and memory resources represented by a second memory 1132 for storing instructions, such as application programs, that can be executed by the processing component 1122. The application programs stored in the second memory 1132 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1122 is configured to execute instructions to perform the parking space detection method described above.

[0218] Device 1100 may also include a power supply component 1126 configured to perform power management of device 1100, a wired or wireless network interface 1150 configured to connect device 1100 to a network, and an input / output interface 1158. Device 1100 can operate on an operating system stored in a second memory 1132, such as Windows Server™, Mac OSX™, Unix™, Linux™, FreeBSD™, or similar.

[0219] In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable device, the computer program having a code portion for performing the parking space detection method described above when executed by the programmable device.

[0220] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0221] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A parking space detection method, characterized in that, include: Determine the free area of ​​the vehicle in the target map, where the free area is a region in the target map that does not include obstacles; Within the free area, areas that meet the parking space requirements are designated for parking. Within the free area, areas that meet the parking space requirements are designated for parking, including: Within the free area, candidate parking areas that meet the parking space conditions are identified; Within the candidate parking area, a target parking space is determined for parking; In the free area, candidate parking areas that meet the parking space conditions are determined, including: Within the free area, the boundary of the area that satisfies the parking space conditions is determined; The area corresponding to the boundary of the area is determined as the candidate parking area; In the free area, determining the area boundary that satisfies the parking space conditions includes: Within the free region, define a region boundary that satisfies at least the following conditions: The first boundary, the second boundary, the third boundary, and the fourth boundary are all parallel to each other. The first boundary, the second boundary, and the third boundary are adjacent and all located on one side of the vehicle. The fourth boundary is located on the other side of the vehicle, opposite to the first boundary. The lengths of the first boundary, the second boundary, the third boundary, and the fourth boundary satisfy a first threshold condition, and the distance between the target boundary among the first boundary, the second boundary, and the third boundary and the fourth boundary satisfies a second threshold condition; Alternatively, within the free area, determining the boundary of the area that satisfies the parking space conditions includes: Within the free region, define a region boundary that satisfies at least the following conditions: The fifth, sixth, and seventh boundaries are all parallel to each other. The fifth and sixth boundaries are adjacent and both are located on one side of the vehicle. The seventh boundary is located on the other side of the vehicle, opposite to the first boundary. The lengths between the fifth boundary, the sixth boundary, and the seventh boundary satisfy the third threshold condition, and the distances between the target boundary in the fifth and sixth boundaries and the seventh boundary satisfy the fourth threshold condition.

2. The method according to claim 1, characterized in that, The first threshold condition includes: The sum of the lengths of the first boundary, the second boundary, and the third boundary is greater than or equal to a first threshold, and the length of the fourth boundary is greater than or equal to the first threshold; The second threshold condition includes: The distance between the first target boundary and the fourth boundary is greater than or equal to a second threshold, the length of the first target boundary is greater than or equal to a third threshold, and the distance between the second target boundary and the fourth boundary is greater than or equal to a fourth threshold. The first target boundary is the boundary with the largest distance from the fourth boundary among the first boundary, the second boundary, and the third boundary. The second target boundary is any boundary among the first boundary, the second boundary, and the third boundary except for the first target boundary. The second threshold is greater than the fourth threshold, and the third threshold is less than the first threshold.

3. The method according to claim 1, characterized in that, The third threshold condition includes: The sum of the lengths of the fifth boundary and the sixth boundary is greater than or equal to the fifth threshold, and the length of the seventh boundary is greater than or equal to the fifth threshold; The fourth threshold condition includes: The distance between the third target boundary and the seventh boundary is greater than or equal to the sixth threshold, the length of the third target boundary is greater than the seventh threshold, and the distance between the fourth target boundary and the seventh boundary is greater than or equal to the eighth threshold. The third target boundary is the boundary with the largest distance from the seventh boundary among the fifth and sixth boundaries, the fourth target boundary is the other boundary among the fifth and sixth boundaries besides the third target boundary, the seventh threshold is less than the fifth threshold, and the eighth threshold is greater than the sixth threshold.

4. The method according to any one of claims 1-3, characterized in that, In the candidate parking area, determining the target parking space for parking includes: Obtain vehicle location information and obstacle information; Based on the positioning information and the obstacle information, a target parking space that meets the safety conditions is determined in the candidate parking area. The safety conditions are used to ensure that the vehicle does not collide with the obstacle corresponding to the obstacle information during the parking process.

5. The method according to claim 4, characterized in that, Based on the location information and the obstacle information, determining the target parking space that meets the safety conditions within the candidate parking area includes: Based on the location information, the target parking area, and the obstacle information, a parking path is planned to obtain the target planned path. The target parking area is the candidate parking area whose distance from the vehicle is less than a distance threshold. If the target planned path meets the safety conditions, the parking space in the target parking area shall be used as the target parking space.

6. The method according to claim 4, characterized in that, Based on the location information and the obstacle information, determining the target parking space that meets the safety conditions within the candidate parking area includes: Based on the location information and the obstacle information, parking path planning is repeatedly performed on the candidate parking areas according to the distance between the candidate parking areas and the vehicle's location information from near to far, and it is determined whether the planned parking path meets the safety conditions, until the planned parking path meets the safety conditions, and the parking space in the candidate parking area that meets the safety conditions is taken as the target parking space.

7. The method according to any one of claims 1-3, characterized in that, Before determining the free area of ​​the vehicle in the target map, the method further includes: Obtain the vehicle's speed information, location information, and obstacle information; Based on the vehicle speed information, determine the region of interest; Based on the location information and the obstacle information, a drivable area is determined; A local grid map is determined based on the region of interest and the drivable area; The target map is obtained based on a series of multiple frames of local raster maps.

8. The method according to claim 7, characterized in that, Based on the vehicle speed information, the region of interest is determined, including: The vehicle speed information and the performance of the vehicle sensors are input into a deep learning model to determine the region of interest. The vehicle sensors are used to acquire the positioning information and obstacle information.

9. The method according to claim 7, characterized in that, Based on the region of interest and the drivable area, a local raster map is determined, including: The drivable region of interest is obtained by the union of the region of interest and the drivable region. Based on the accuracy of the vehicle sensors, the granularity information of the grid is determined, and the vehicle sensors are used to acquire the positioning information and obstacle information; The drivable region of interest is divided into grids based on the granularity information to obtain the local grid map.

10. The method according to claim 7, characterized in that, The target map is obtained based on a series of multiple frames of local raster maps, including: The local raster maps of the target frame number are stitched together to obtain the intermediate raster map; The intermediate raster map is updated using the local raster map of the next frame of the target frame to obtain the target map.

11. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the parking space detection method according to any one of claims 1-10.

12. A vehicle, characterized in that, include: A storage device, wherein a computer program is stored in the storage device; A processing device for executing the computer program to implement the parking space detection method according to any one of claims 1-10.

Citation Information

Patent Citations

  • Parking space detection method and device, electronic equipment, vehicle and storage medium

    CN111311925A