Method for constructing drivable area of vehicle in parking scenario and system thereof

Abstract

The present application relates to a vehicle drivable area construction method and system under a parking scenario. The method comprises the following steps: according to a preset parking scenario category, setting a default maximum drivable area and a default minimum drivable area, the default maximum drivable area and the default minimum drivable area being determined based on empirical values; establishing a vehicle model with double perception centers, based on the perception centers, emitting rays from the vehicle to detect obstacles and generate obstacle point clouds; taking two corner points of a parking space entrance line as the center, radiating to the surrounding to establish a local obstacle occupied grid map; using a Bayesian updating method to optimize the grid map in real time to fuse perception data and correct the occupancy probability of obstacles; based on the occupancy probability of the grid map, shrinking the parking boundary from the default maximum parking area to the default minimum parking area to the edge of the obstacle to generate a vehicle drivable area without obstacles.

Description

Technical Field

[0001] This invention relates to automatic parking technology, and more specifically, to a method and system for constructing a vehicle-accessible area in a parking scenario. Background Technology

[0002] In autonomous driving systems, the construction of the vehicle's drivable area serves as a crucial bridge between the perception, decision-making, and control modules. The accuracy and timeliness of drivable area modeling directly impact the safety and robustness of subsequent parking path planning, obstacle avoidance, and vehicle control.

[0003] Traditional methods typically rely on high-precision maps and rule bases to quickly determine drivable areas by matching vehicle positioning with the map. However, in autonomous driving applications, especially in unstructured parking scenarios, the frequent changes in the dynamic environment and the lag in map updates pose significant risks to constructing drivable areas solely based on map information. Therefore, real-time sensor perception of the surrounding environment and dynamic construction of drivable areas have become the mainstream approach, including methods based on grids, semantic segmentation, and the fusion of geometric and semantic information. These methods can alleviate the reliance on high-precision maps to some extent, but they still have the following shortcomings: Insufficient recognition accuracy in areas with weak features (such as open spaces and areas without obvious lane markings); It has limited detection capabilities for small obstacles and non-standard shaped obstacles; Multi-sensor data fusion and spatiotemporal alignment present real-time challenges, leading to delays in updating the achievable area. Summary of the Invention

[0004] To address the problems in the prior art, this invention aims to provide a method and system for constructing a vehicle drivable area in a parking scenario that balances the robustness, real-time performance, and fine-grained accuracy of algorithm design while also taking into account the computing power limitations of the algorithm deployment platform.

[0005] The method for constructing a vehicle-driving area in a parking scenario according to one aspect of the present invention includes the following steps: The default area setting step involves setting the default maximum drivable area and the default minimum drivable area based on preset parking scenario categories. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. The dual-sensor center detection step involves establishing a vehicle model with dual sensor centers, and emitting rays outward from around the vehicle based on the sensor centers to detect obstacles and generate obstacle point clouds. The local grid map construction steps involve taking the two corner points of the parking space entrance line as the center and radiating outwards to create a local grid map occupied by obstacles. The grid map only records obstacles that affect the parking trajectory. The grid map update step uses a Bayesian update method to optimize the grid map in real time to integrate sensing data and correct the occupancy probability of obstacles; The parking space shrinking step involves shrinking the parking boundary from the default maximum parking area to the default minimum parking area, based on the occupancy probability of the grid map, to the edge of the obstacle, in order to generate an obstacle-free drivable area for vehicles.

[0006] Optionally, after the parking space retraction step, the following may be further included: The safe redundancy area generation step involves further shrinking the effective parking space inward by a preset safe distance to form a safe redundancy area.

[0007] Optionally, each sensing center has a 300-degree sensing range and a 1-degree resolution.

[0008] Optionally, in the dual-sensing center detection step, a single-frame ray beam is emitted from the sensing center to the surrounding 360 degrees to search for the nearest obstacle point line by line. If no obstacle is detected, a preset default long-distance occupancy is used for compensation.

[0009] Optionally, in the local grid map construction step, the size and resolution of the grid map are dynamically adjusted according to the parking space size and obstacle density.

[0010] Optionally, in the grid map update step, the corrected obstacle occupancy probability includes: During the initialization phase, all grids are assigned a uniform prior occupancy probability; Once the sensing data arrives, the posterior probability is calculated based on the current observation results; If a grid cell is detected as occupied multiple times, the probability of occupancy is gradually increased; conversely, the probability of occupancy is gradually decreased. Once the set threshold is exceeded, the grid is officially marked as occupied or free.

[0011] The vehicle drivable area construction system in a parking scenario according to one aspect of the present invention includes: The default area setting module is used to set the default maximum drivable area and the default minimum drivable area according to the preset parking scenario category. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. The dual-sensor center detection module is used to establish a dual-sensor center vehicle model, and emit rays outward from the vehicle based on the sensor centers to detect obstacles and generate obstacle point clouds; The local grid map construction module is used to radiate outwards from the two corner points of the parking space entrance line to create a grid map occupied by local obstacles. The grid map only records obstacles that affect the parking trajectory. The grid map update module is used to optimize the grid map in real time using a Bayesian update system to fuse multi-frame perception data and correct the occupancy probability of obstacles. The parking space shrinking module is used to shrink the parking boundary from the default maximum parking area to the default minimum parking area based on the occupancy probability of the grid map, so as to generate an obstacle-free vehicle driving area.

[0012] Optionally, it further includes: The safety redundancy area generation module is used to further shrink the effective parking space inward by a preset safety distance to form a safety redundancy area.

[0013] Optionally, each sensing center has a 300-degree sensing range and a 1-degree resolution.

[0014] Optionally, in the dual-sensing center detection module, a single-frame ray beam is emitted from the sensing center to the surrounding 360 degrees to search for the nearest obstacle point line by line. If no obstacle is detected, a preset default long-distance occupancy is used for compensation.

[0015] The present invention provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing a vehicle drivable area in a parking scenario.

[0016] A computer device according to one aspect of the present invention includes a storage module, a processor, and a computer program stored on the storage module and executable on the processor. When the processor executes the computer program, it implements the method for constructing a vehicle drivable area in a parking scenario.

[0017] A computer program product according to one aspect of the present invention includes a computer program that, when executed by a processor, implements the method for constructing a vehicle drivable area in a parking scenario. Attached Figure Description

[0018] The described and other objects and advantages of the invention will become more fully clear from the following detailed description taken in conjunction with the accompanying drawings, wherein like or similar elements are denoted by the same reference numerals.

[0019] Figure 1 This is a flowchart illustrating an example of a multi-sensor-based parking system.

[0020] Figure 2 This is a flowchart illustrating a method for constructing a vehicle drivable area in a parking scenario according to one embodiment of the present invention.

[0021] Figure 3 This diagram illustrates the default parking spaces for three different parking scenarios.

[0022] Figure 4 A schematic diagram of a vehicle perception model and a grid map in a parking system is shown.

[0023] Figure 5 This is a schematic diagram illustrating the structure of a vehicle driving area construction system in a parking scenario according to an embodiment of the present invention. Detailed Implementation

[0024] The following are some embodiments of the present invention, intended to provide a basic understanding of the invention. They are not intended to identify key or decisive elements of the invention or to limit the scope of protection sought.

[0025] For purposes of brevity and illustrativeness, the principles of the invention are described herein primarily with reference to exemplary embodiments thereof. However, those skilled in the art will readily recognize that the same principles are equivalently applicable to all types of vehicle drivable area construction methods and systems of the present invention, and that these same principles can be implemented therein, with any such variations not departing from the true spirit and scope of the invention.

[0026] Furthermore, reference is made in the following description to the accompanying drawings, which illustrate specific exemplary embodiments. Electrical, mechanical, logical, and structural modifications may be made to these embodiments without departing from the spirit and scope of the invention. Moreover, while features of the invention are disclosed in conjunction with only one of several embodiments, this feature may be combined with one or more other features of other embodiments if desired and / or advantageous for any given or identifiable function. Therefore, the following description should not be considered limiting in any sense, and the scope of the invention is defined by the appended claims and their equivalents.

[0027] Terms such as “possessing” and “comprising” indicate that, in addition to having units (modules) and steps that are directly and explicitly stated in the specification and claims, the technical solution of the present invention does not exclude the presence of other units (modules) and steps that are not directly or explicitly stated.

[0028] Figure 1 This is a flowchart illustrating an example of a multi-sensor-based parking system.

[0029] like Figure 1 As shown, the parking system first receives data from multiple sensors, such as cameras, radar, and ultrasonic sensors, to collect information about the vehicle's surroundings.

[0030] Next, near-field obstacle fusion mapping is carried out, which mainly includes: using sensor data to construct a local obstacle grid map for parking space guidance; establishing a vehicle perception model with dual perception centers; and fusing historical perception results from multiple sensors.

[0031] Next, parking trajectory planning is performed, mainly including: using traditional geometric regular trajectory planning methods; a trajectory planning method combining regularization and parametric search; and a trajectory planning method based on Hybrid A* search optimization.

[0032] Then, path tracking control is performed, which mainly includes: during the path tracking process, tracking of the heading preview point to ensure that the tracking heading matches the target heading; at the same time, tracking of the lateral preview point to ensure that the lateral tracking position matches the target position.

[0033] Finally, the parking system sends control commands to downstream actuators, such as the electric power steering (EPS) system and electronic stability program (ESP), to actually control the movement of the vehicle and complete the parking operation.

[0034] The purpose of this invention is to provide a method for constructing a vehicle's drivable area in parking scenarios. This method can stably output the vehicle's drivable area based on parking space information, obstacle information, and positioning information. The main vehicle can drive freely within this drivable area without any risk of collision.

[0035] Figure 2 This is a flowchart illustrating a method for constructing a vehicle drivable area in a parking scenario according to one embodiment of the present invention.

[0036] like Figure 2 As shown, a method for constructing a vehicle drivable area in a parking scenario according to one embodiment of the present invention includes the following steps: In the default area setting step S100, the default maximum drivable area and the default minimum drivable area are set according to the preset parking scenario category. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. Dual-sensory-center detection step S200: Establish a vehicle model with dual-sensory-centers, and emit rays outward from around the vehicle based on the sensory centers to detect obstacles and generate obstacle point clouds. In the local grid map construction step S300, a grid map radiating outwards from the two corner points of the parking space entrance line is established to cover the local obstacles. The grid map only records obstacles that affect the parking trajectory. In the grid map update step S400, the Bayesian update method is used to optimize the grid map in real time to fuse sensing data and correct the occupancy probability of obstacles. The parking space shrinking step S500 involves shrinking the parking boundary from the default maximum parking area to the default minimum parking area based on the occupancy probability of the grid map, down to the edge of the obstacle, to generate an obstacle-free drivable area for vehicles; and In the safety redundancy area generation step S600, the effective parking space is further contracted inward according to a preset safety distance to form a safety redundancy area.

[0037] Next, these steps will be explained in detail.

[0038] In the default area setting step, firstly, a default maximum maneuverable area is constructed based on the scenario category. Specifically, according to different parking scenario categories (such as horizontal, perpendicular, and angled parking), based on extensive measured data and engineering experience, a default maximum space range that the vehicle can utilize during parking is defined. This default maximum space aims to ensure that the vehicle can smoothly complete parking with a reasonable trajectory under unobstructed conditions, typically covering the maximum space required for typical parking maneuvers. Figure 3 This diagram illustrates the default parking spaces for three different parking scenarios. Figure 3 The left side indicates perpendicular parking, the middle indicates angled parking, and the right side indicates parallel parking.

[0039] Next, a default minimum drivable area is constructed based on the scenario category. Specifically, a default minimum drivable area is further set based on the default maximum drivable area. This minimum area is the minimum space required for the vehicle to complete the parking maneuver, and its parameters are also set based on extensive simulation and real-world road testing experience. If obstacles (such as guardrails, parking locks, or debris) are detected within the minimum parking area, it indicates that there are insurmountable obstacles on the path the vehicle must take to complete the parking process. In this case, the mapping module will proactively output an abnormal signal to the path planning module, indicating that the current parking task cannot be executed, thus avoiding wasting system computing resources and time, improving the response speed of the decision-making module, and enhancing the overall intelligence level of the system.

[0040] In the dual-sensing-center detection step, a vehicle perception model based on dual-sensing-centers is established. Since the actual vehicles are equipped with different sensor layouts (such as millimeter-wave radar, ultrasonic radar, lidar, surround-view cameras, etc.) and different perception characteristics (such as detection distance, detection angle, resolution, etc.), this invention proposes a vehicle perception model based on dual-sensing-centers to improve the integrity and reliability of environmental perception.

[0041] As an example, each perception center covers a 300-degree detection range with a perception angle resolution of 1 degree, ensuring detailed and accurate environmental mapping. Based on the perception model settings, scattering lines are emitted outwards from the vehicle's center, and the first hit point is recorded when an obstacle is encountered. If no obstacle is detected in a certain direction, the default maximum detection distance is used as a constraint to avoid blind spots affecting subsequent processing.

[0042] After processing, each frame of input obstacle data can generate, for example, a local point cloud containing 600 discrete points centered on the vehicle. This high-density point cloud can accurately reflect the distribution of obstacles around the vehicle, providing a reliable basis for subsequent construction of local grid maps and dynamic updates of obstacles.

[0043] In the local grid map construction step, a local obstacle occupancy grid map is established based on the corner points of parking spaces. Specifically, the two corner points of the parking space entrance line output by the parking space detection module are used as reference points for local mapping, and a local obstacle occupancy grid map is generated radiating outwards. The coverage of this local map focuses on areas where vehicles are highly likely to pass during parking trajectory planning, and it uses dynamic size and resolution, which can be adaptively adjusted according to the size of the parking space and the density of obstacles. By processing only local areas, the computational load for mapping and updating is significantly reduced, improving the real-time performance of the system. In addition, since obstacles around parking spaces have the greatest impact on parking path planning, this invention further improves the accuracy of the planning module in judging drivability by focusing on mapping this area, reducing the number of path planning failures or path replanning.

[0044] Figure 4 It is a schematic diagram showing a vehicle perception model and a grid map in a parking system.

[0045] Figure 4 The left side shows a vehicle perception model with dual perception centers. There are two intersecting ellipses around the vehicle, representing the vehicle's perception boundary, which represents the perception range of the vehicle's surrounding environment. Figure 4 The right side of the screen displays two grid maps, one above the target parking space. These maps divide the vehicle's surroundings into small grids, each representing a specific area and containing information about that area (such as the presence of obstacles). Below the grid maps, the location of the target parking space is marked, indicating the specific spot where the vehicle needs to be parked. The information in the grid maps helps the vehicle determine its relative position to the target parking space and plan the optimal parking route.

[0046] In the grid map update step, to ensure the real-time performance and accuracy of environmental mapping, this invention employs a grid map update method based on Bayes' theorem. Specifically, the occupancy probability of each grid cell is continuously updated based on the obstacle point cloud input. By fusing information from multiple heterogeneous sensors, misjudgments caused by perceptual noise, occlusion interference, and differences in obstacle height are corrected, thereby improving the stability and robustness of obstacle detection.

[0047] As an example, the specific implementation process is as follows: During the initialization phase, all grids are assigned a uniform prior occupancy probability; Once the sensing data arrives, the posterior probability is calculated based on the current observation results; If a grid cell is detected as occupied multiple times, its occupancy probability is gradually increased; conversely, it is gradually decreased if the cell is detected as occupied less frequently. Once the set threshold is exceeded, the grid is officially marked as occupied or free.

[0048] This dynamic update mechanism not only enhances the system's adaptability in complex dynamic environments, but also effectively mitigates the fluctuations in perception performance caused by the types of obstacles (such as highly reflective objects, low obstacles, etc.).

[0049] In the parking space shrinking step, after completing the local grid map construction, this invention partitions the complementary area between the default minimum parking space and the default maximum parking space based on the obstacle point set. Subsequently, according to the partitioning results, the boundary of the default maximum parking space is dynamically shrunk from the outside in until the boundary is close to the obstacles, forming an actual usable parking space, ensuring that there are no obstacle points inside the final parking area.

[0050] This space contraction method has the following advantages: it can preserve parking space to the maximum extent and improve the parking success rate; it has good adaptability to abnormal obstacle shapes (such as non-rectangular ones); and it supports real-time updates in dynamic scenarios, such as adaptive space adjustment during the entry and exit of other vehicles.

[0051] Figure 5 This is a schematic diagram illustrating the structure of a vehicle driving area construction system in a parking scenario according to an embodiment of the present invention.

[0052] like Figure 5 As shown, a vehicle drivable area construction system 100 in a parking scenario according to an embodiment of the present invention includes: The default area setting module 110 is used to set the default maximum drivable area and the default minimum drivable area according to the preset parking scenario category. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. The dual-sensing-center detection module 120 is used to establish a dual-sensing-center vehicle model, and to emit rays outward from around the vehicle based on the sensing centers to detect obstacles and generate obstacle point clouds. The local grid map construction module 130 is used to radiate outwards from the two corner points of the parking space entrance line to establish a grid map occupied by local obstacles, wherein the grid map only records obstacles that affect the parking trajectory. The grid map update module 140 is used to optimize the grid map in real time using a Bayesian update system to fuse multi-frame perception data and correct the occupancy probability of obstacles. A parking space shrinkage module 150 is used to shrink the parking boundary from the default maximum parking area to the default minimum parking area based on the occupancy probability of the grid map, to the edge of the obstacle, so as to generate an obstacle-free drivable area for vehicles; and The safety redundancy area generation module 160 is used to further shrink the effective parking space inward according to a preset safety distance to form a safety redundancy area.

[0053] Each sensing center has a 300-degree sensing range and a 1-degree resolution. In the dual-sensing-center detection module 120, a single-frame ray beam is emitted from the sensing center to the surrounding 360 degrees, searching for the nearest obstacle point line by line. If no obstacle is detected, a preset default long-distance occupancy is used for compensation.

[0054] In the safety redundancy area generation module 160, a set safety distance threshold (e.g., about 15cm to 25cm) is shrunk inward from the final parking boundary to leave an obstacle buffer zone.

[0055] As described above, the method for constructing a vehicle drivable area in a parking scenario and the system for constructing a vehicle drivable area in a parking scenario according to one aspect of the present invention can achieve the following technical effects: Based on the characteristics of different parking scenarios, a dual setting of default maximum drivable area and default minimum drivable area is proposed. Through empirical modeling, the initial boundary definition of parking space is given, providing basic condition judgment for parking decisions in different scenarios, thereby improving parking safety. A vehicle modeling method is proposed, which establishes dual perception centers with the two sides of the vehicle center as the base points, forming an ultra-wide field-of-view perception network covering 300 degrees and with a resolution of 1 degree. This method can improve the integrity and redundancy of environmental mapping and adapt to various vehicle sensor layouts. Starting from the two corners of the parking space entrance line, local mapping is carried out radiating outwards. The perceived obstacle point cloud is divided into zones according to its spatial relationship with the parking space area. Obstacle points close to the parking space entrance and the necessary path are given high priority, while the interference of distant and irrelevant obstacles on mapping and decision-making is reduced. The system uses Bayesian inference to fuse multi-frame perception results in real time and dynamically updates the grid map occupancy probability. This overcomes single-frame perception errors caused by obstacle material, shape, height, etc., and can achieve a robust improvement in environmental perception. By detecting the distribution of obstacles within the complement area of ​​the maximum and minimum parking spaces, adaptive boundary shrinkage processing of the parking space is performed to ensure that the parking space closely matches the actual obstacle distribution, thus avoiding wasted space or misjudgment. Based on the parking boundary after it has been shrunk to be free of obstacles, it is further shrunk inward according to the set safety distance threshold to ensure that even if the vehicle has a slight deviation or position drift during the parking process, it can avoid the risk of contact with obstacles. Starting from the perception center, a single-frame ray beam is emitted in a 360-degree radius around the sensor to search for the nearest obstacle point line by line. If no obstacle is detected, the default long-distance occupancy compensation is used to ensure that areas without information can be processed correctly. The planning module and the mapping module work together dynamically, adjusting the parking trajectory online based on real-time mapping results to ensure that the trajectory is updated synchronously with the drivable area, effectively responding to moving obstacles or sudden environmental changes in the site.

[0056] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art can conceive of other feasible variations or substitutions based on the technical scope disclosed in the present invention, and such variations or substitutions are all covered within the scope of protection of the present invention. In the absence of conflict, the embodiments of the present invention and the features thereof can also be combined with each other. The scope of protection of the present invention is determined by the claims.

Claims

1. A method for constructing a vehicle-driving area in a parking scenario, characterized in that, Includes the following steps: The default area setting step involves setting the default maximum drivable area and the default minimum drivable area based on preset parking scenario categories. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. The dual-sensor center detection step involves establishing a vehicle model with dual sensor centers, and emitting rays outward from around the vehicle based on the sensor centers to detect obstacles and generate obstacle point clouds. The local grid map construction steps involve taking the two corner points of the parking space entrance line as the center and radiating outwards to create a local grid map occupied by obstacles. The grid map only records obstacles that affect the parking trajectory. The grid map update step uses a Bayesian update method to optimize the grid map in real time to integrate sensing data and correct the occupancy probability of obstacles; as well as The parking space shrinking step involves shrinking the parking boundary from the default maximum parking area to the default minimum parking area based on the occupancy probability of the grid map, to the edge of the obstacle, in order to generate an obstacle-free drivable area for vehicles.

2. The method for constructing a vehicle-drivable area in a parking scenario as described in claim 1, characterized in that, Following the parking space retraction step, the following is further included: The safety redundancy area generation step further shrinks the effective parking space inward according to a preset safety distance to form a safety redundancy area.

3. The method for constructing a vehicle-driving area in a parking scenario as described in claim 1, characterized in that, Each of the aforementioned sensing centers has a 300-degree sensing range and a 1-degree resolution.

4. The method for constructing a vehicle-drivable area in a parking scenario as described in claim 1, characterized in that, In the dual-sensing center detection step, a single-frame ray beam is emitted from the sensing center to the surrounding 360 degrees to search for the nearest obstacle point line by line. If no obstacle is detected, a preset default long-distance occupancy is used for compensation.

5. The method for constructing a vehicle-drivable area in a parking scenario as described in claim 1, characterized in that, In the local grid map construction step, the size and resolution of the grid map are dynamically adjusted according to the parking space size and obstacle density.

6. The method for constructing a vehicle-drivable area in a parking scenario as described in claim 1, characterized in that, In the grid map update step, the corrected obstacle occupancy probability includes: During the initialization phase, all grids are assigned a uniform prior occupancy probability; Once the sensing data arrives, the posterior probability is calculated based on the current observation results; If a grid cell is detected as occupied multiple times, the probability of occupancy is gradually increased; conversely, the probability of occupancy is gradually decreased. Once the set threshold is exceeded, the grid is officially marked as occupied or free.

7. A system for constructing a vehicle-accessible area in a parking scenario, characterized in that, include: The default area setting module is used to set the default maximum drivable area and the default minimum drivable area according to the preset parking scenario category. The default maximum drivable area and the default minimum drivable area are determined based on empirical values. The dual-sensor center detection module is used to establish a dual-sensor center vehicle model, and emit rays outward from the vehicle based on the sensor centers to detect obstacles and generate obstacle point clouds; The local grid map construction module is used to radiate outwards from the two corner points of the parking space entrance line to create a grid map occupied by local obstacles. The grid map only records obstacles that affect the parking trajectory. The grid map update module is used to optimize the grid map in real time using a Bayesian update system to fuse multi-frame perception data and correct the occupancy probability of obstacles. The parking space shrinking module is used to shrink the parking boundary from the default maximum parking area to the default minimum parking area based on the occupancy probability of the grid map, so as to generate an obstacle-free vehicle driving area.

8. The vehicle drivable area construction system in a parking scenario as described in claim 7, characterized in that, Further includes: The safety redundancy area generation module is used to further shrink the effective parking space inward according to a preset safety distance to form a safety redundancy area.

9. The vehicle drivable area construction system in a parking scenario as described in claim 7, characterized in that, Each of the aforementioned sensing centers has a 300-degree sensing range and a 1-degree resolution.

10. The vehicle drivable area construction system in a parking scenario as described in claim 7, characterized in that, In the dual-sensing center detection module, a single-frame ray beam is emitted from the sensing center to the surrounding 360 degrees to search for the nearest obstacle point line by line. If no obstacle is detected, a preset default long-distance occupancy is used for compensation.

11. A computer-readable medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for constructing a vehicle drivable area in a parking scenario as described in any one of claims 1 to 6.

12. A computer device, comprising a storage module, a processor, and a computer program stored on the storage module and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for constructing a vehicle drivable area in a parking scenario as described in any one of claims 1 to 6.

13. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for constructing a vehicle drivable area in a parking scenario as described in any one of claims 1 to 6.

Images