An automatic parking path planning method, device, medium and equipment
By identifying and virtualizing ground markings to construct parking space boundaries, and using a scaled-down virtual vehicle silhouette for crash testing, the problem of large parking space requirements and low success rate in existing technologies is solved, achieving more efficient parking path planning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- MOMENTA (SUZHOU) TECHNOLOGY CO LTD
- Filing Date
- 2021-12-28
- Publication Date
- 2026-08-04
AI Technical Summary
Existing automatic parking path planning methods require a large amount of space when planning parking spaces, and have low success rates, low efficiency, and cannot effectively utilize space beyond the ground markings.
By identifying and virtualizing ground markings to construct virtual boundaries for parking spaces, and using the scaled-down virtual vehicle outlines to conduct collision tests against the virtual boundaries, vehicles are allowed to cross the ground markings, increasing available parking space and improving the success rate of path planning.
In real-world parking scenarios, the available space for path planning is increased, improving parking success rate and efficiency while reducing the space required for parking.
Smart Images
Figure CN116353620B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and in particular to an automatic parking path planning method, apparatus, medium and equipment. Background Technology
[0002] Automated parking is a crucial function in autonomous driving. Existing technologies include various automated parking path planning methods, including the real-time circular arc method. This method, based on the vehicle's and parking space's positions, uses multiple circular arcs to construct a pre-planned parking path, ensuring no collisions with the parking space defined by parking space boundaries, road edges, and obstacles, and adhering to kinematic constraints. Another automated parking path planning method utilizes the Hybrid A* search algorithm to calculate and plan the parking path. Both of these methods require a relatively large parking space for route planning, but are limited by the specific parking environment, resulting in low path planning success rates and low parking efficiency. Summary of the Invention
[0003] To address the problems existing in the prior art, this application mainly provides an automatic parking path planning method, device, medium, and equipment. By constructing a virtual boundary of the parking space and using a reduced virtual vehicle outline to conduct collision tests with the virtual boundary, the vehicle can cross the ground markings in actual parking scenarios, thereby increasing the available parking space during path planning, improving the success rate of path planning, reducing the space required for parking, and improving parking efficiency.
[0004] To achieve the above objectives, one technical solution adopted in this application is: providing an automatic parking path planning method, which includes:
[0005] The system extracts perception information from the vehicle within the predetermined parking space, including the target parking space, and identifies ground markings based on this information. It then virtualizes the ground markings to obtain virtual ground markings and uses these markings to determine the virtual parking space boundary of the target parking space. The system extracts the vehicle outline of the vehicle to be parked and reduces its size to obtain a virtual vehicle outline. Virtual collision detection is performed using the virtual parking space boundary and the virtual vehicle outline, and the system plans the parking trajectory of the vehicle based on the results of the virtual collision detection to obtain an automatic parking planning path.
[0006] Another technical solution adopted in this application is: providing an automatic parking path planning device, which includes:
[0007] The system comprises the following modules: a ground marking acquisition module, which extracts perceived information about the predetermined parking space, including the target parking space, and identifies the ground markings based on this information; a virtual parking space boundary acquisition module, which virtualizes the ground markings to obtain virtual ground markings and uses these virtual ground markings to obtain the virtual parking space boundary of the target parking space; a virtual vehicle outline acquisition module, which extracts the vehicle outline of the vehicle to be parked and reduces its size to obtain the virtual vehicle outline; and a collision detection and planning module, which performs virtual collision detection using the virtual parking space boundary and the virtual vehicle outline, and plans the parking trajectory of the vehicle to be parked based on the results of the virtual collision detection to obtain an automatic parking planning path.
[0008] Another technical solution adopted in this application is to provide a computer-readable storage medium storing computer instructions that are operated to execute the automatic parking path planning method in the above solution.
[0009] Another technical solution adopted in this application is: providing a computer device, which includes a processor and a memory, the memory storing computer instructions, which are operated to execute the automatic parking path planning method in the above solution.
[0010] The beneficial effects of the technical solution of this application are: an automatic parking path planning method, device, medium, and equipment. This application constructs a virtual boundary of the parking space using ground markings and conducts collision tests using a scaled-down virtual vehicle outline against the virtual boundary. This allows vehicles to cross the ground markings in actual parking scenarios, increasing the available parking space during path planning, improving the success rate of on-street parking path planning, reducing the space required for parking, and improving parking efficiency. Attached Figure Description
[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is a schematic diagram illustrating a use case of an automatic parking path planning method according to this application;
[0013] Figure 2 This is a diagram illustrating the vehicle crossing the side line of the ground during normal parking.
[0014] Figure 3 This is a schematic diagram of a specific implementation of an automatic parking path planning method according to this application;
[0015] Figure 4 This is a schematic diagram of the parking environment range determined in a specific embodiment of an automatic parking path planning method of this application;
[0016] Figure 5 This is a specific embodiment of an automatic parking path planning method of this application, which utilizes a reduced outline diagram of the vehicle;
[0017] Figure 6 This is a schematic diagram of a specific parking scenario in a specific embodiment of an automatic parking path planning method of this application;
[0018] Figure 7 This is a schematic diagram of a specific parking scenario in a specific embodiment of an automatic parking path planning method of this application;
[0019] Figure 8 This is a schematic diagram of a specific implementation of an automatic parking path planning method according to this application;
[0020] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0021] The preferred embodiments of this application will now be described in detail with reference to the accompanying drawings, so that the advantages and features of this application can be more easily understood by those skilled in the art, thereby providing a clearer and more definite definition of the scope of protection of this application.
[0022] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.
[0023] Automated parking is an important function in autonomous driving, and its use cases include... Figure 1As shown, the Automatic Parking Assist (APA) system automatically parks the vehicle by controlling its acceleration, deceleration, and steering angle. The system uses sensors to detect collisions and perceive the parking environment, estimating the vehicle's attitude (position and direction of travel). Based on the driver's selection, it automatically or manually sets the target parking space. The system then calculates the automatic parking path and, through precise vehicle positioning and control, automates the parking process along the defined path until the final target parking space is reached.
[0024] Several automatic parking path planning methods already exist in the technology. When using sensors to perceive the environment and detect obstacles during the parking path planning process, these existing technologies rely on the vehicle's actual outline against the parking space boundaries and road markings for collision detection to avoid collisions between the vehicle and the parking space boundaries. This ensures that even if the parking space boundaries are marked on the ground, vehicles can still cross the ground edges during normal parking. Figure 2 As shown, when planning a route, it is also necessary to avoid vehicles hitting the ground edge lines, which would result in the actual planned route being less than optimal, requiring more parking space, having a low success rate in parking route planning, and low parking efficiency.
[0025] This application constructs a virtual boundary for parking space by identifying and virtualizing ground markings in the parking environment. It then uses a scaled-down virtual vehicle outline to conduct collision tests with the virtual boundary, allowing the vehicle body to collide with ground markings during path planning. This increases the available space during path planning, improves the success rate of path planning, and ensures that the final planned path is optimal, reducing the space required for parking and improving parking efficiency.
[0026] The technical solutions of this application will now be described in detail with reference to specific embodiments and accompanying drawings. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.
[0027] Figure 3 This application illustrates a specific implementation of an automatic parking path planning method.
[0028] exist Figure 3The specific implementation of the automatic parking path planning method of this application shown includes the following steps: Step S301: Extracting the perception information of the vehicle within the predetermined parking space, including the target parking space, and identifying the ground markings based on the perception information; Step S302: Virtualizing the ground markings to obtain virtual ground markings, and using the virtual ground markings to obtain the virtual parking space boundary of the target parking space; Step S303: Extracting the vehicle outline of the vehicle to be parked, and reducing the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked; and Step S304: Performing virtual collision detection using the virtual parking space boundary and the virtual vehicle outline, and planning the parking trajectory of the vehicle to be parked based on the results of the virtual collision detection to obtain the automatic parking planning path.
[0029] By identifying and virtualizing road markings, including parking lines, and then constructing virtual parking space boundaries, collision tests are conducted between the virtual outline of the vehicle (after reducing its size) and the virtual boundaries. This allows vehicles to cross the road markings in parking scenarios, thereby increasing the available space for parking path planning, improving the flexibility of the parking process, optimizing the final planned path, reducing the space required for parking, and increasing the parking success rate.
[0030] Process S301 represents the process of extracting perceived information from the vehicle within the predetermined parking space, including the target parking space, and identifying ground markings based on this perceived information. This facilitates the virtualization of ground markings, allowing subsequent use of the virtual ground markings to obtain the virtual parking space boundary. Specifically, the ground markings include the parking space boundary lines of the target parking space and road markings.
[0031] In one specific embodiment of this application, the process of extracting the perceived information of the vehicle within a predetermined parking space, including the target parking space, includes determining the area within the predetermined parking space based on the vehicle's dimensions and minimum turning radius, such as... Figure 4 As shown. For example, the above parking environment range can be determined as a 30m x 30m square based on empirical values, or the above parking environment range can be determined as a square with a side length of: vehicle length + minimum turning radius.
[0032] In one specific embodiment of this application, at least one sensor configured on the vehicle acquires an image of the physical parking environment where the parking space is located, and objects related to parking are extracted to facilitate further identification of ground markings in the parking environment. Specifically, the vehicle can be any type of vehicle that carries people and / or objects and moves via a power system such as an engine or battery, including but not limited to cars, trucks, buses, electric vehicles, motorcycles, RVs, trains, etc. The image includes objects related to parking (e.g., parking space boundaries, including information such as pillars, cones, and edge lines) as well as objects unrelated to parking (e.g., other vehicles and scenery), requiring the identification of parking-related objects from a large number of objects.
[0033] Optionally, the aforementioned at least one sensor may include an AVM (around-view fisheye camera) and a USS (ultrasonic radar).
[0034] In a specific example of this application, the process of extracting parking-related objects includes parking space detection, actual obstacle detection, and grounding wire detection. Specifically, the parking space detection process includes parking space boundary detection; the actual obstacle detection process includes detecting other vehicles, people, and traffic cones within a predetermined range; and the grounding wire detection process includes detecting the boundaries of walls, pillars, and other building structures that contact the ground within a predetermined range.
[0035] In one specific embodiment of this application, the process of extracting parking-related objects further includes an object fusion process. This involves fusing object information perceived by multiple cameras into a single object information; for example, fusing information about the same obstacle captured by multiple cameras into a single obstacle information for output; fusing camera perception results with ultrasonic perception results; for example, fusing parking space information captured by cameras and parking space information obtained by ultrasonic radar into a single parking space information for output; or fusing multiple frames of perception information into a single information; for example, fusing multiple frames of images of the same grounding wire captured by a single camera into a single obstacle information for output. This results in more stable and accurate object information.
[0036] In one specific embodiment of this application, the process of identifying ground markings based on perceived information includes identifying the ground boundary lines of the target parking space. Specifically, after completing the parking space detection, it is determined whether the parking space boundary is a ground marking, and the obtained ground parking space boundary lines are virtualized to obtain virtual parking space boundary lines, so as to subsequently use the virtual parking space boundary lines to construct a virtual parking space boundary.
[0037] In one specific embodiment of this application, the process of identifying ground markings based on perception information includes identifying road markings within the predetermined parking space of the target parking space so as to subsequently virtualize the road markings and construct a virtual parking space boundary using the virtual road markings.
[0038] Process S302 represents the process of virtualizing ground markings to obtain virtual ground markings, and using virtual ground markings to obtain the virtual parking space boundary of the target parking space, which facilitates collision detection using the virtual parking space boundary.
[0039] In one specific embodiment of this application, the parking space edge line is virtualized to obtain a virtual parking space edge line, and the virtual parking space boundary of the target parking space is obtained using the virtual parking space edge line.
[0040] In one specific embodiment of this application, road markings are virtualized and the virtual parking space boundary of the target parking space is obtained using the virtual road markings.
[0041] In one specific embodiment of this application, the parking space edge lines and road markings are virtualized to obtain virtual parking space edge lines and virtual road markings, and the virtual parking space boundary of the target parking space is obtained using the virtual parking space edge lines and virtual road markings.
[0042] Process S303 represents the extraction of the vehicle outline of the vehicle to be parked, and the reduction of the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked, as shown in the figure. Figure 5 As shown, it is possible to facilitate collision detection using the virtual vehicle outline and the virtual parking space boundary, and to conduct collision tests using the scaled-down virtual vehicle outline and the virtual boundary, thereby increasing the available parking space during path planning, improving the success rate of path planning, reducing the space required for parking, and improving parking efficiency.
[0043] In one specific embodiment of this application, the process of reducing the vehicle outline to obtain a virtual vehicle outline of the vehicle to be parked includes reducing the vehicle outline to obtain the virtual vehicle outline. Optionally, the vehicle outline is reduced to no more than 0.8 times the actual vehicle outline, centered on its center, to obtain the virtual vehicle outline. Reducing the vehicle outline to 0.8 times significantly improves the success rate of parking path planning and yields a more ideal parking path.
[0044] In one specific embodiment of this application, the process of reducing the vehicle outline to obtain a virtual vehicle outline of the vehicle to be parked includes reducing the vehicle outline to no more than 0.5 times the actual vehicle outline, using the center of the vehicle outline as the center, to obtain a virtual vehicle outline. Reducing the vehicle outline to 0.5 times significantly improves the success rate of parking path planning and results in a more ideal parking path.
[0045] Process S304 represents the use of virtual parking space boundaries and virtual vehicle outlines to perform virtual collision detection, and based on the results of virtual collision detection, to plan the parking trajectory of the vehicle to be parked and obtain an automatic parking planning path. It can ultimately plan a parking path that intersects with the ground markings with a high success rate, so that parking can be carried out efficiently.
[0046] In a specific example of this application, a collision test is conducted using a virtual vehicle outline and virtual parking space lines representing the boundaries of a virtual parking space. Specifically, during actual vehicle parking, this results in the vehicle being able to cross the edge lines of the parking space on the ground. This not only increases the available space during path planning but also reduces the parking space required during actual parking, thereby improving the success rate of path planning and increasing parking efficiency.
[0047] In a specific example of this application, a collision test is conducted using a virtual vehicle outline and virtual road markings representing the boundaries of a virtual space. Specifically, in real-world vehicle parking scenarios, the area outside the road markings often contains other parking spaces or available space, such as… Figure 5 As shown, existing technologies employ a collision test method between road markings and the actual vehicle outline, rendering the space outside the road markings unusable. In this embodiment, during parking, the vehicle can utilize the road markings and borrow available space outside the road markings for parking. This increases the available space during path planning, improves the success rate of path planning, and enhances parking efficiency.
[0048] In a specific instance of this application, in Figure 6 In the parking environment shown, where the parking boundary is only ground markings, the parking space boundary constructed by virtual parking space lines and virtual road markings is used to facilitate collision testing with virtual vehicle outlines and increase the available space in the parking path planning process.
[0049] In one specific embodiment of this application, such as Figure 7 When the parking environment shown includes not only ground markings but also actual obstacles at the parking boundary, the automatic parking path planning method of this application further includes: identifying the outline of the actual obstacle based on perception information; performing actual collision detection using the outline of the actual obstacle and the actual vehicle outline of the vehicle to be parked; and when there are actual obstacles in the actual parking scenario, performing collision detection using the actual vehicle outline and the actual obstacle to avoid collision between the vehicle and the actual obstacle.
[0050] In one specific embodiment of this application, the automatic parking path planning method further includes: identifying the ground side lines and road markings of the target parking space based on perception information; trimming the ground side lines and the road markings intersecting with the ground side lines, and using the remaining ground side lines and road markings after trimming as the corresponding boundaries of the available parking space of the target parking space; and performing actual collision detection using the corresponding boundaries of the available parking space and the actual vehicle outline of the vehicle to be parked.
[0051] In a specific example of this application, the process of trimming the ground side line and the road markings intersecting with the ground side line includes: starting from the portion of the ground side line closest to the road edge line, trimming off 0.3 to 1 times the length of the ground side line; and starting from the portion of the road edge line intersecting with the ground side line, trimming off 0.5 to 1.5 times the length of the trimmed ground side line. When the ground side line and road markings are trimmed to this extent, the success rate and efficiency of automatic parking path planning are significantly improved, and the resulting path is more ideal.
[0052] In one specific embodiment of this application, the automatic parking path planning method further includes planning a parking trajectory for the vehicle to be parked based on the actual collision detection results to obtain an automatic parking planning path. When actual obstacles exist in a real parking scenario, collision detection is performed between the actual vehicle outlines. This ensures a high success rate in planning an optimal parking path, enabling efficient parking while safely avoiding obstacles.
[0053] This embodiment can plan an optimal parking path with a high success rate, enabling efficient parking while safely avoiding obstacles. Specifically, the outlines of the actual obstacles include the actual outlines of objects such as walls, pillars, and cones.
[0054] In one specific embodiment of this application, the process of planning the parking trajectory of the vehicle to be parked based on the results of virtual collision detection to obtain an automatic parking planning path includes planning the path using the real-time circular arc method based on the results of collision detection.
[0055] In one specific embodiment of this application, the process of planning the parking trajectory of the vehicle to be parked based on the results of virtual collision detection to obtain an automatic parking planning path includes calculating the planned parking path using the HybridA* search algorithm based on the results of collision detection.
[0056] In one specific embodiment of this application, in parking scenarios where vehicles need to be parked frequently, such as in a home garage, the automatic parking path planning method of this application further includes storing the parking path obtained by performing the above-mentioned automatic parking path planning, so that it can be directly called and followed according to the stored path when parking in the future.
[0057] Figure 8 This application illustrates a specific embodiment of an automatic parking path planning device.
[0058] exist Figure 8 The specific implementation of the automatic parking path planning device of this application shown includes a ground marking acquisition module 801, which extracts the perception information of the vehicle within the predetermined parking space, including the target parking space, and identifies the ground markings based on the perception information; a virtual parking space boundary acquisition module 802, which virtualizes the ground markings to obtain virtual ground markings and uses the virtual ground markings to obtain the virtual parking space boundary of the target parking space; a virtual vehicle outline acquisition module 802, which extracts the vehicle outline of the vehicle to be parked and reduces the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked; and a collision detection and planning module 804, which performs virtual collision detection using the virtual parking space boundary and the virtual vehicle outline, and plans the parking trajectory of the vehicle to be parked based on the results of the virtual collision detection to obtain an automatic parking planning path.
[0059] The path planning device of this application can identify and virtualize the markings on the ground, including parking space lines, and then construct a virtual parking space boundary. By using the virtual outline of the vehicle (after reducing its size) to conduct a collision test with the virtual boundary, the vehicle can cross the ground markings in the parking scenario. This increases the available space during parking path planning, improves the flexibility of the parking process, optimizes the final planned path, reduces the space required for parking, and increases the parking success rate.
[0060] The ground marking acquisition module 801 is used to extract perception information within a predetermined parking space, including the target parking space, as perceived by the vehicle, and to identify ground markings based on the perception information. This facilitates the virtualization of ground markings, allowing subsequent use of the virtual ground markings to obtain the boundaries of the virtual parking space. Specifically, the ground markings include the parking space boundary lines of the target parking space and road markings.
[0061] In one specific embodiment of this application, the virtual ground marking acquisition module 801 includes at least one sensor configured on the vehicle, such as AVM (Around View Fisheye Camera) and USS (Ultrasonic Radar).
[0062] The virtual parking space boundary acquisition module 802 is used to virtualize ground markings to obtain virtual ground markings and use virtual ground markings to obtain the virtual parking space boundary of the target parking space, which facilitates subsequent collision detection using the virtual parking space boundary.
[0063] The virtual vehicle outline acquisition module 803 is used to extract the vehicle outline of the vehicle to be parked and reduce the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked. It can facilitate collision detection using the virtual vehicle outline and the virtual parking space boundary, and use the reduced virtual vehicle outline and the virtual boundary for collision testing. This increases the available parking space when performing path planning, improves the success rate of path planning, reduces the space required for parking, and improves parking efficiency.
[0064] The collision detection and planning module 804 is used to perform virtual collision detection using the virtual parking space boundary and virtual vehicle outline, and to plan the parking trajectory of the vehicle to be parked based on the results of the virtual collision detection to obtain an automatic parking planning path. It can ultimately plan a parking path that intersects with the ground markings with a high success rate, so that parking can be carried out efficiently.
[0065] In one specific embodiment of this application, the automatic parking path planning device of this application further includes: an actual obstacle contour line recognition module, used to recognize the actual obstacle contour line based on perception information.
[0066] In one specific embodiment of this application, the automatic parking path planning device further includes: a corresponding boundary acquisition module for available parking space, used to identify the ground side lines and road markings of the target parking space based on perception information; to trim the ground side lines and the road markings intersecting with the ground side lines, and to use the remaining ground side lines and road markings after trimming as the corresponding boundaries of the available parking space of the target parking space.
[0067] In this specific embodiment, the collision detection and planning module 804 is further configured to perform actual collision detection using the actual obstacle outline and the corresponding boundary of the available parking space and the actual vehicle outline of the vehicle to be parked; and to obtain an automatic parking planning path by planning the parking trajectory of the vehicle to be parked based on the results of the actual collision detection.
[0068] In real-world parking scenarios, when actual obstacles exist, collision detection is required between the vehicle's outline and the obstacle to prevent collisions. This embodiment can plan an optimal parking path with a high success rate, enabling efficient parking while safely avoiding obstacles.
[0069] In one specific embodiment of this application, in parking scenarios where vehicles need to be parked frequently, such as in a home garage, the automatic parking path planning device of this application further includes a path storage device, which is used to store the parking path obtained by performing the above-mentioned automatic parking path planning, so that it can be directly called and followed according to the stored path when parking in the future.
[0070] The extended automatic parking path planning device provided in this application can be used to execute the extended automatic parking path planning method described in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be repeated here.
[0071] In one specific embodiment of this application, the functional modules of the automatic parking path planning device of this application can be directly in hardware, in software modules executed by a processor, or in a combination of both.
[0072] Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in this art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium.
[0073] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor, but alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration. Alternatively, the storage medium can be integrated with the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in the user terminal. Alternatively, the processor and storage medium can reside as discrete components in the user terminal.
[0074] In another specific embodiment of this application, a computer-readable storage medium stores computer instructions that are operated to perform the extended automatic parking path planning method described above.
[0075] In another specific embodiment of this application, a computer device includes a processor and a memory, the memory storing computer instructions that are operated to execute the automatic parking path planning method in the above scheme.
[0076] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0077] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0078] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. An automatic parking path planning method characterized by comprising: include: Extract the perception information of the vehicle within the predetermined parking space, including the target parking space, and identify the ground markings based on the perception information; The ground markings are virtualized to obtain virtual ground markings, and the virtual parking space boundary of the target parking space is obtained using the virtual ground markings. Extract the vehicle outline of the vehicle that needs to be parked, and reduce the vehicle outline to obtain the virtual vehicle outline of the vehicle that needs to be parked. Virtual collision detection is performed using the virtual parking space boundary and the virtual vehicle outline. Based on the results of the virtual collision detection, parking trajectory planning is performed on the vehicle to be parked to obtain an automatic parking planning path.
2. The automatic parking path planning method according to claim 1, characterized in that, The ground markings include the parking space edge lines of the target parking space and the road markings; The process of virtualizing the ground markings to obtain virtual ground markings, and using the virtual ground markings to obtain the virtual parking space boundary of the target parking space includes, The parking space edge lines and road markings are virtualized to obtain virtual parking space edge lines and virtual road markings, and the virtual parking space boundary of the target parking space is obtained using the virtual parking space edge lines and the virtual road markings.
3. The automatic parking path planning method according to claim 1, characterized in that, The process of reducing the size of the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked includes: The virtual vehicle outline is obtained by shrinking the vehicle outline with the center of the vehicle outline as the center.
4. The automatic parking path planning method according to claim 1, characterized in that, The process of reducing the size of the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked includes: Using the center of the vehicle outline as the center, the vehicle outline is reduced to no more than 0.5 times the size of the actual vehicle outline to obtain the virtual vehicle outline.
5. The automatic parking path planning method according to claim 1, characterized in that, It also includes, The actual obstacle outline is identified based on the perceived information; Actual collision detection is performed using the actual obstacle outline and the actual vehicle outline of the vehicle to be parked. as well as Based on the actual collision detection results, an automatic parking planning path is obtained by planning the parking trajectory of the vehicle that needs to be parked.
6. The automatic parking path planning method according to claim 4, characterized in that, It also includes, The ground side lines and road markings of the target parking space are identified based on the perceived information. The ground side lines and the road markings that intersect with the ground side lines are cut off, and the remaining ground side lines and road markings after cutting are used as the corresponding boundaries of the available parking space of the target parking space. as well as The actual collision detection is performed using the corresponding boundaries of the available parking space and the actual vehicle outline of the vehicle to be parked.
7. The automatic parking path planning method according to claim 5, characterized in that, It also includes the process of trimming the ground side line and the road edge line intersecting with the ground side line. Starting from the portion closest to the road edge, trim off 0.3 to 1 times the length of the ground side line; as well as Starting from the portion closest to the ground side line, cut off 0.5 to 1.5 times the length of the cut-off ground side line.
8. An automatic parking path planning device, characterized in that, include, The ground marking acquisition module is used to extract the perception information of the vehicle within the predetermined parking space, including the target parking space, and to identify the ground markings based on the perception information. The virtual parking space boundary acquisition module is used to virtualize the ground markings to obtain virtual ground markings, and use the virtual ground markings to obtain the virtual parking space boundary of the target parking space. The virtual vehicle outline acquisition module is used to extract the vehicle outline of the vehicle to be parked and reduce the vehicle outline to obtain the virtual vehicle outline of the vehicle to be parked. The collision detection and planning module is used to perform virtual collision detection using the virtual parking space boundary and the virtual vehicle outline, and to plan the parking trajectory of the vehicle to be parked based on the results of the virtual collision detection to obtain an automatic parking planning path.
9. A computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are operated to perform the automatic parking path planning method according to any one of claims 1-7.
10. A computer device comprising a processor and a memory storing computer instructions, wherein the processor operates the computer instructions to perform the automatic parking path planning method according to any one of claims 1-7.