Automatic parking control method, vehicle end, cloud end and electronic equipment

Abstract

The embodiment of the invention relates to the technical field of automatic parking, and discloses an automatic parking control method, a vehicle end, a cloud end and electronic equipment, and the method comprises the steps: responding to a parking space detection instruction, and determining a target parking space and a target scene element based on obtained sensor data; in response to a parking starting instruction, sending a scene matching request to a cloud based on the target scene element; the scene matching request is used for indicating the cloud to determine a reference scene matched with the target scene element in a plurality of preset parking scenes, and sending a parking difficulty coefficient corresponding to the reference scene to a vehicle end; the parking difficulty coefficient sent by the cloud end is received, default regulation and control parameters are adjusted based on the parking difficulty coefficient, target regulation and control parameters are obtained, and a target parking strategy is generated based on the target regulation and control parameters; and parking the vehicle into the target parking space by adopting the target parking strategy. By applying the technical scheme of the invention, the rationality of the parking strategy can be improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of automatic parking, and in particular, to an automatic parking control method, a vehicle terminal, a cloud end, and an electronic device. Background Art

[0002] The current automatic parking technology can automatically plan a parking path after the driver selects a parking space, and automatically park the vehicle into the selected parking space without the driver operating the vehicle.

[0003] In the related art, there is a situation where the automatically planned parking path is unreasonable, resulting in a collision during the automatic parking process. Summary of the Invention

[0004] In view of the above problems, the embodiments of the present invention provide an automatic parking control method, a vehicle terminal, a cloud end, and an electronic device, which are used to solve the problem that the parking strategy of automatic parking in the prior art is unreasonable.

[0005] According to one aspect of the embodiments of the present invention, an automatic parking control method is provided. The method includes:

[0006] In response to a parking space detection instruction, determining a target parking space and target scene elements based on the acquired sensor data; the target scene elements are in the parking scene area corresponding to the target parking space; in response to a parking start instruction, sending a scene matching request to the cloud end based on the target scene elements; the scene matching request is used to instruct the cloud end to determine a reference scene that matches the target scene elements among multiple preset parking scenes, and send the parking difficulty coefficient corresponding to the reference scene to the vehicle terminal; receiving the parking difficulty coefficient sent by the cloud end, adjusting the default regulation and control parameters based on the parking difficulty coefficient to obtain target regulation and control parameters, and generating a target parking strategy based on the target regulation and control parameters; parking the vehicle into the target parking space by using the target parking strategy.

[0007] According to another aspect of the embodiments of the present invention, an automatic parking control method is provided. The method includes:

[0008] In response to a scene matching request sent by the vehicle terminal, obtain the target scene elements included in the scene matching request; the scene matching request is sent by the vehicle terminal in response to a parking space detection instruction, based on the acquired sensor data to determine the target parking space and target scene elements, and in response to a parking start instruction for the target parking space, based on the target scene elements to the cloud; the target scene elements are in the parking scene area corresponding to the target parking space; determine a reference scene that matches the target scene elements among multiple preset parking scenes; obtain the parking difficulty coefficient corresponding to the reference scene, and send the parking difficulty coefficient to the vehicle terminal, so that the vehicle terminal adjusts the default regulation and control parameters based on the parking difficulty coefficient to obtain the target regulation and control parameters, and generates a target parking strategy based on the target regulation and control parameters, and uses the target parking strategy to park the vehicle into the target parking space.

[0009] According to another aspect of the embodiments of the present invention, a vehicle terminal is provided, including:

[0010] An identification module, configured to, in response to a parking space detection instruction, determine a target parking space and target scene elements based on the acquired sensor data; the target scene elements are in the parking scene area corresponding to the target parking space;

[0011] A request module, configured to, in response to a parking start instruction, send a scene matching request to the cloud based on the target scene elements; the scene matching request is used to instruct the cloud to determine a reference scene that matches the target scene elements among multiple preset parking scenes, and send the parking difficulty coefficient corresponding to the reference scene to the vehicle terminal;

[0012] A regulation and control parameter adjustment module, configured to receive the parking difficulty coefficient sent by the cloud, adjust the default regulation and control parameters based on the parking difficulty coefficient to obtain the target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters;

[0013] A parking execution module, configured to park the vehicle into the target parking space using the target parking strategy.

[0014] According to another aspect of the embodiments of the present invention, a cloud is provided, including:

[0015] An acquisition module, configured to, in response to a scene matching request sent by the vehicle terminal, obtain the target scene elements included in the scene matching request; the scene matching request is sent by the vehicle terminal in response to a parking space detection instruction, based on the acquired sensor data to determine the target parking space and target scene elements, and in response to a parking start instruction for the target parking space, based on the target scene elements to the cloud; the target scene elements are in the parking scene area corresponding to the target parking space;

[0016] A scene matching module, configured to determine a reference scene that matches the target scene elements among multiple preset parking scenes;

[0017] The difficulty coefficient sending module is configured to obtain the parking-in difficulty coefficient corresponding to the reference scenario and send the parking-in difficulty coefficient to the vehicle terminal, so that the vehicle terminal adjusts the default motion control parameters based on the parking-in difficulty coefficient to obtain the target motion control parameters, generates a target parking strategy based on the target motion control parameters, and parks the vehicle into the target parking space by using the target parking strategy.

[0018] According to another aspect of the embodiments of the present invention, there is provided an automatic parking control system, including:

[0019] The vehicle terminal is configured to, in response to a parking space detection instruction, determine a target parking space and target scenario elements based on the acquired sensor data; the target scenario elements are in the parking scenario area corresponding to the target parking space; and in response to a parking start instruction, send a scenario matching request to the cloud based on the target scenario elements.

[0020] The cloud is configured to, in response to the scenario matching request sent by the vehicle terminal, acquire the target scenario elements included in the scenario matching request; determine a reference scenario matching the target scenario elements from multiple preset parking scenarios; obtain the parking-in difficulty coefficient corresponding to the reference scenario, and send the parking-in difficulty coefficient to the vehicle terminal.

[0021] The vehicle terminal is further configured to receive the parking-in difficulty coefficient sent by the cloud, adjust the default motion control parameters based on the parking-in difficulty coefficient to obtain the target motion control parameters, generate a target parking strategy based on the target motion control parameters, and park the vehicle into the target parking space by using the target parking strategy.

[0022] According to another aspect of the embodiments of the present invention, there is provided an electronic device, including: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus; the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operations of the automatic parking control method.

[0023] According to yet another aspect of the embodiments of the present invention, there is provided a computer-readable storage medium, in which at least one executable instruction is stored, and the executable instruction causes an electronic device / device to execute the operations of the automatic parking control method.

[0024] According to yet another aspect of the embodiments of the present invention, there is provided a computer program product, which includes at least one executable instruction, and the executable instruction causes an electronic device / device to execute the operations of the automatic parking control method.

[0025] In an embodiment of the present invention, the vehicle end identifies target scene elements within the parking scene area corresponding to the target parking space, and jointly performs scene matching with the cloud to determine the reference scene where the target parking space is located. The default regulation and control parameters are adjusted by using the parking difficulty coefficient corresponding to the reference scene to obtain the target regulation and control parameters, and a target parking strategy is generated based on the target regulation and control parameters. Thus, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same. However, in different parking scenes, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficients of different parking scenes, and a target parking strategy is generated based on the target regulation and control parameters, so that the target parking strategy conforms to the actual parking scene, improving the rationality of the parking strategy. In high-difficulty parking scenes, the collision risk caused by using the default parking strategy for automatic parking is avoided, enhancing the safety of automatic parking. In low-difficulty parking scenes, the efficiency of automatic parking can be improved.

[0026] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to be able to understand the technical means of the embodiment of the present invention more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and understandable, the following specifically gives the specific implementation manners of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings are only used to illustrate the embodiments and are not considered as a limitation to the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0028] Figure 1 The flowchart of the first embodiment of the automatic parking control method provided by the present invention is shown;

[0029] Figure 2 The schematic diagram of the parking scene provided by the present invention is shown;

[0030] Figure 3 The schematic diagram of the default parking trajectory determined in the related art is shown;

[0031] Figure 4 The schematic diagram of the target parking trajectory determined in the embodiment of the present invention is shown;

[0032] Figure 5 The flowchart of the second embodiment of the automatic parking control method provided by the present invention is shown;

[0033] Figure 6 The schematic diagram of adjusting the difficulty weight of the element and the difficulty coefficient of the scene provided by the present invention is shown;

[0034] Figure 7It shows a schematic flowchart of an automatic parking control method executed by the automatic parking control system provided by the present invention;

[0035] Figure 8 It shows a schematic diagram of the execution units at the vehicle end and the cloud end provided by the present invention;

[0036] Figure 9 It shows a schematic diagram of the structure of the vehicle end provided by the present invention;

[0037] Figure 10 It shows a schematic diagram of the structure of the cloud end provided by the present invention;

[0038] Figure 11 It shows a schematic diagram of the structure of an embodiment of the electronic device provided by the present invention. Detailed Embodiments

[0039] Hereinafter, exemplary embodiments of the present invention will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0040] Figure 1 It shows a flowchart of the first embodiment of the automatic parking control method of the present invention, and this method is executed by the vehicle end. As Figure 1 shown, this method includes the following steps:

[0041] Step 110: In response to a parking space detection instruction, determine a target parking space and target scene elements based on the acquired sensor data; the target scene elements are in the parking scene area corresponding to the target parking space.

[0042] Among them, the parking space detection instruction is used to search for available parking spaces and identify target scene elements;

[0043] The sensor data may include but is not limited to: image data acquired by an in-vehicle camera, point cloud data acquired by an in-vehicle lidar; the target parking space is the parking space where the vehicle parks.

[0044] The target scene elements are elements in the parking scene area corresponding to the target parking space. For example, the target scene elements may include: flower beds, lane lines, ground markings, pillars, gate rods, steps; the target scene elements are used to determine the parking scene where the target parking space is located. For example, according to flower beds, ground markings, and steps, it can be determined that the parking scene where the target parking space is located is a sidewalk parking scene.

[0045] The parking scenario area is a partial area in the parking scenario where the target parking space is located. The parking scenario area covers the first area corresponding to the current position of the vehicle and the second area corresponding to the target parking space. The first area can be an area determined with the current position as the center and a preset distance as the radius, and the second area can be an area determined with the target parking space as the center and a preset distance as the radius; the parking scenario area includes: a drivable geographical area and a non-drivable geographical area.

[0046] The parking scenario area can be far from the target parking space and is not limited to the area near the target parking space and the area near the current position of the vehicle. Furthermore, the target scenario elements can include elements that do not have a direct impact on the parking path.

[0047] Optionally, the user triggers a parking space detection instruction by performing a parking space detection operation. The vehicle terminal responds to the parking space detection instruction, performs parking space recognition through the acquired sensor data, obtains available parking spaces and displays them. The user performs a selection operation on one of the available parking spaces, and the vehicle terminal uses the available parking space pointed to by the selection operation as the target parking space. The vehicle terminal determines the parking scenario area based on the target parking space, performs element recognition through the sensor data to obtain scenario elements, and selects target scenario elements from the scenario elements according to the parking scenario area.

[0048] Optionally, the user triggers a parking space detection instruction by performing a parking space detection operation. The vehicle terminal responds to the parking space detection instruction, performs parking space recognition and element recognition on the acquired sensor data to obtain available parking spaces and scenario elements. The vehicle terminal displays the available parking spaces so that the user can perform a selection operation on one of the available parking spaces. The vehicle terminal uses the available parking space pointed to by the selection operation as the target parking space, determines the parking scenario area corresponding to the target parking space, and selects target scenario elements belonging to the parking scenario area from the scenario elements.

[0049] Among them, identifying scenario elements through sensor data can be to splice images acquired by multiple vehicle-mounted cameras with different perspectives to obtain a panoramic image, perform feature extraction on the panoramic image to obtain image features; perform feature extraction on the point cloud data acquired by the vehicle-mounted lidar to obtain radar features, fuse the image features and radar features to obtain fused features, and perform element recognition on the fused features to obtain scenario elements.

[0050] Identifying available parking spaces through sensor data can be to splice images acquired by multiple vehicle-mounted cameras with different perspectives to obtain a panoramic image, perform feature extraction on the panoramic image to obtain image features; perform feature extraction on the point cloud data acquired by the vehicle-mounted lidar to obtain radar features, fuse the image features and radar features to obtain fused features, and perform parking space recognition on the fused features to obtain available parking spaces.

[0051] Step 120: In response to a parking start instruction, send a scene matching request to the cloud based on target scene elements; the scene matching request is used to instruct the cloud to determine a reference scene that matches the target scene elements among multiple preset parking scenes and send the parking difficulty coefficient corresponding to the reference scene to the vehicle terminal.

[0052] Among them, the parking start instruction is used to indicate the execution of a parking operation; in the application of leaving the vehicle and parking in, when the user selects a target parking space, the user leaves the vehicle, and when the vehicle terminal detects that the user has left the vehicle and closed the door, the parking start instruction is triggered; in the application of automatic parking in, when the user selects a target parking space, a parking start operation can be triggered (such as clicking the parking start control in the in-vehicle display or the vehicle control application), and when the vehicle terminal detects the parking start operation, the parking start instruction is triggered.

[0053] The reference scene is one of the multiple preset parking scenes, and is a scene determined by the cloud that is similar to the parking scene where the target parking space is located.

[0054] The parking difficulty coefficient is used to characterize the difficulty of automatic parking in the reference scene. The larger the parking difficulty coefficient, the greater the difficulty of automatic parking in the reference scene. Conversely, the smaller the parking difficulty coefficient, the smaller the difficulty of automatic parking in the reference scene.

[0055] Each of the multiple preset parking scenes has its own corresponding parking difficulty coefficient; the parking difficulty coefficient is determined in advance, and the parking difficulty coefficient of the preset parking scene can be determined according to the difficulty weights of the preset elements included in the preset parking scene; the difficulty weights of the preset elements can be set according to experience, or can be obtained by updating the difficulty weights set according to experience multiple times based on the actual automatic parking situation.

[0056] Specifically, in response to the parking start instruction for the target parking space, the vehicle terminal generates a scene matching request according to the target scene elements and sends the scene matching request to the cloud.

[0057] The cloud receives the scene matching request, obtains the target scene elements included in the scene matching request; determines a reference scene that matches the target scene elements among the multiple preset parking scenes, obtains the parking difficulty coefficient corresponding to the reference scene, and sends the parking difficulty coefficient to the vehicle terminal.

[0058] Among them, determining a reference scene that matches the target scene elements among the multiple preset parking scenes includes: for each preset parking scene, obtaining the preset elements in the preset parking scene targeted, determining the matching degree of the preset parking scene targeted based on the similarity between the preset elements and the target scene elements; selecting a reference scene among the multiple preset parking scenes according to the matching degrees corresponding to the multiple preset parking scenes.

[0059] Specifically, the number of target scenario elements can be multiple; for each preset parking scenario, the preset parking scenario includes multiple preset elements. For each target scenario element, the first candidate similarity between the target scenario element and each preset element is determined respectively, and the average of the multiple first candidate similarities is calculated to obtain the second candidate similarity corresponding to the target scenario element. In the same way, the second candidate similarities corresponding to all target scenario elements can be determined, and then the average of the multiple second candidate similarities is calculated to obtain the matching degree between the target scenario elements and the preset parking scenario.

[0060] In the same way as above, the matching degrees between the target scenario elements and multiple preset parking scenarios can be obtained respectively; the preset parking scenario corresponding to the highest matching degree among the matching degrees corresponding to the multiple preset parking scenarios is used as the reference scenario; the cloud obtains the parking-in difficulty coefficient corresponding to the reference scenario and sends the parking-in difficulty coefficient to the vehicle terminal.

[0061] Step 130: Receive the parking-in difficulty coefficient sent by the cloud, adjust the default regulation and control parameters based on the parking-in difficulty coefficient to obtain the target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters.

[0062] Among them, the default regulation and control parameters are the default parameters of the planning and control model, and the planning and control model is used to generate a parking strategy.

[0063] The default regulation and control parameters include the default trajectory segment length threshold and the default parking speed; the default trajectory segment length threshold is used to constrain the maximum length of a single trajectory in the parking trajectory, and the default parking speed is used to constrain the maximum vehicle speed during the parking process.

[0064] The target regulation and control parameters include the target trajectory segment length threshold and the target parking speed; the default trajectory segment length threshold is different from the target trajectory segment length threshold, and / or the default parking speed is different from the target parking speed.

[0065] It should be noted that changing the trajectory segment length threshold may cause the length of a single trajectory in the parking strategy to change. For example, the default trajectory segment length threshold is 5 meters, and the parking strategy generated under this constraint may include a 5-meter trajectory segment. The target trajectory segment length threshold is 3 meters, and the parking strategy generated under this constraint does not include a trajectory segment longer than 3 meters; changing the parking speed means changing the speed of the vehicle during automatic parking.

[0066] The target parking strategy is a strategy generated by the planning and control model under the constraint of the target regulation and control parameters; the target parking strategy includes the target parking trajectory; according to the target parking trajectory, the vehicle can be parked from the current position into the target parking space.

[0067] Specifically, the vehicle end receives the parking-in difficulty coefficient sent by the cloud, determines whether the parking-in difficulty coefficient is the same as the default coefficient. If the parking-in difficulty coefficient is different from the default coefficient, the default control parameters are adjusted according to the parking-in difficulty coefficient. When the parking-in difficulty coefficient is greater than the default coefficient, the default trajectory segment length threshold and the default parking speed are reduced to obtain the target control parameters. When the parking-in difficulty coefficient is less than the default coefficient, the default trajectory segment length threshold and the default parking speed are increased to obtain the target control parameters. The planning and control model performs parking planning based on the current position of the vehicle, the target parking space, and the sensor data under the constraint of the target control parameters to obtain the target parking strategy.

[0068] Optionally, the automatic parking control method further includes: the vehicle end generates a default parking strategy according to the current position of the vehicle, the target parking space, and the sensor data in response to a parking start instruction for the target parking space; and adjusts the default parking strategy based on the parking-in difficulty coefficient to obtain the target parking strategy.

[0069] Specifically, the current position of the vehicle, the target parking space, and the sensor data are input into the planning and control model, and parking planning is performed under the constraint of the default control parameters to obtain the default parking strategy. The vehicle end receives the parking-in difficulty coefficient sent by the cloud. When the parking-in difficulty coefficient is different from the default coefficient, the parking trajectory and / or the parking speed included in the default parking strategy are adjusted according to the parking-in difficulty coefficient and the target scenario elements to obtain the target parking strategy.

[0070] When the parking-in difficulty coefficient is higher than the default coefficient, the length of each single segment of the parking trajectory can be reduced, and at the same time, the curvature of the single segment of the trajectory is adaptively adjusted. The parking speed can also be reduced to obtain the target parking strategy. It should be noted that if only the length of the single segment of the trajectory is reduced, it may cause the vehicle to be unable to park in the target parking space according to the adjusted parking trajectory. Therefore, while reducing the length of the single segment of the trajectory, the curvature of the single segment of the trajectory is adaptively adjusted to ensure that the vehicle can park in the target parking space according to the adjusted parking trajectory.

[0071] When the parking-in difficulty coefficient is lower than the default coefficient, the length of each single segment of the parking trajectory can be increased, and at the same time, the curvature of the single segment of the trajectory is adaptively adjusted. The parking speed can also be increased to obtain the target parking strategy. Similarly, if only the length of the single segment of the trajectory is increased, it may cause the vehicle to be unable to park in the target parking space according to the adjusted parking trajectory. Therefore, while increasing the length of the single segment of the trajectory, the curvature of the single segment of the trajectory is adaptively adjusted to ensure that the vehicle can park in the target parking space according to the adjusted parking trajectory.

[0072] Optionally, when the parking-in difficulty coefficient is equal to the default coefficient, the target parking strategy is generated based on the default control parameters.

[0073] Step 140: Park the vehicle in the target parking space using the target parking strategy.

[0074] Specifically, the vehicle end controls the vehicle to perform automatic parking based on the target parking strategy, so that the vehicle parks into the target parking space.

[0075] It should be noted that in the related art, in different parking scenarios, when the relative position between the target parking space and the current position is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the same regulation and control parameters (i.e., default regulation and control parameters) are adopted in different parking scenarios, and thus the default parking strategy obtained is the same; for example, when the relative position between the target parking space and the current position is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategies in the parking scenarios of the underground parking lot, the sidewalk parking scenario, and the open-air parking lot are the same.

[0076] However, in actual situations, there may be elements such as flower beds, steps, street lights, and temporarily placed items in the sidewalk parking scenario, resulting in a very complex situation for automatic parking; in the parking scenario of the underground parking lot, there may be columns and walls, resulting in a relatively complex situation for automatic parking. In the parking scenario of the open-air parking lot, there are fewer static obstacles, so the situation for automatic parking is relatively simple. That is to say, in practical applications, the difficulty of automatic parking in the parking scenarios of the underground parking lot, the sidewalk parking scenario, and the open-air parking lot is different, but the related art does not distinguish between parking scenarios. Using the default parking strategy for automatic parking in a high-difficulty parking scenario will lead to a collision risk.

[0077] The automatic parking control method provided by the embodiment of the present application, the vehicle end identifies the target scenario elements in the parking scenario area corresponding to the target parking space, and jointly performs scenario matching with the cloud to determine the reference scenario where the target parking space is located, adjusts the default regulation and control parameters by using the parking difficulty coefficient corresponding to the reference scenario to obtain the target regulation and control parameters, and generates a target parking strategy based on the target regulation and control parameters; in this way, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same, but in different parking scenarios, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficients of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, so that the target parking strategy conforms to the actual parking scenario, improving the rationality of the parking strategy. In a high-difficulty parking scenario, the collision risk caused by using the default parking strategy for automatic parking is avoided, improving the safety of automatic parking. In a low-difficulty parking scenario, the efficiency of automatic parking can be improved.

[0078] In an alternative manner, determining a target parking space and target scene elements based on the acquired sensor data includes: performing target recognition on the acquired sensor data to obtain available parking spaces and candidate scene elements; in response to a selection operation, determining a target parking space from the available parking spaces; determining a parking scene area corresponding to the target parking space; and based on the parking scene area, determining target scene elements from the candidate scene elements.

[0079] Among them, an available parking space is a parking space for the user to select; the target parking space is one of the available parking spaces, that is, the parking space selected by the user; the parking scene area corresponding to the target parking space is a partial area in the parking scene where the target parking space is located; the parking scene area is larger than the area occupied by the target parking space, the current position of the vehicle, and the passage area from the current position to the target parking space; the target scene elements are a part of the candidate scene elements.

[0080] Specifically, the sensor data includes: image data respectively acquired by multiple in-vehicle cameras from different perspectives, and point cloud data acquired by the in-vehicle lidar; stitching the multiple image data to obtain a panoramic image, performing feature extraction on the panoramic image to obtain image features; performing feature extraction on the point cloud data to obtain radar features, fusing the image features and radar features to obtain fused features, performing target recognition on the fused features to obtain a recognition result, taking the targets of the type of parking space in the recognition result as available parking spaces, and taking the targets of the type of scene elements in the recognition result as candidate scene elements.

[0081] Optionally, after identifying the available parking spaces, display the available parking spaces on the parking space detection page; for example, the user initiates a parking space detection instruction through the parking space detection page displayed on the vehicle terminal display, and after sub-identifying the available parking spaces, display the available parking spaces on the parking space detection page.

[0082] The user performs a selection operation on the available parking spaces displayed on the parking space detection page, and the vehicle terminal, in response to the selection operation, takes the available parking space pointed to by the selection operation as the target parking space.

[0083] Optionally, the vehicle terminal takes the central position point of the target parking space as the center of a circle and a preset distance as the radius to determine the parking scene area corresponding to the target parking space; the preset distance can be set according to actual requirements.

[0084] Optionally, the vehicle terminal takes the central position point of the target parking space as the first center of a circle and a first preset distance as the radius to determine a first parking scene area, takes the current position of the vehicle as the second center of a circle and a second preset distance as the radius to determine a second parking scene area, and the parking scene area includes the first parking scene area and the second parking scene area; where the specific values of the first preset distance and the second preset distance can be set according to actual requirements.

[0085] After determining the parking scenario area corresponding to the target parking space, obtain the positions of candidate scenario elements. If the positions of the candidate scenario elements belong to the parking scenario area, then use the candidate scenario elements as the target scenario elements.

[0086] In the above embodiment, the sensor data used to search for parking spaces is reused to identify candidate scenario elements. After selecting the target parking space, the parking scenario area is determined based on the target parking space, and then the target scenario elements within the parking scenario area are selected, which can narrow down the range of target scenario elements, reduce the amount of data used to identify scenario elements, and improve the efficiency of the automatic parking control method.

[0087] In an optional manner, the default control parameters include a default trajectory segment length threshold and a default parking speed; adjust the default control parameters based on the parking difficulty coefficient to obtain the target control parameters, including: when the parking difficulty coefficient is different from the default coefficient, determine an adjustment factor based on the parking difficulty coefficient and the default coefficient; adjust the default trajectory segment length threshold and / or the default parking speed according to the adjustment factor to obtain the target control parameters.

[0088] Among them, the adjustment factor can be the absolute value of the difference between the parking difficulty coefficient and the default coefficient.

[0089] Specifically, the vehicle terminal calculates the absolute value of the difference between the parking difficulty coefficient and the default coefficient to obtain the adjustment factor. If the adjustment factor belongs to the first preset interval, then adjust the default parking speed according to the adjustment factor to obtain the target control parameters; if the adjustment factor belongs to the second preset interval, then adjust the default trajectory segment length threshold according to the adjustment factor to obtain the target control parameters; if the adjustment factor belongs to the third preset interval, then adjust the default trajectory segment length threshold and the default parking speed according to the adjustment factor to obtain the target control parameters; among them, the values in the first preset interval are less than the values in the second preset interval, and the values in the second preset interval are less than the values in the third preset interval.

[0090] In the above embodiment, adjust the default control parameters according to the parking difficulty coefficient to obtain the target control parameters, so as to subsequently generate a target parking strategy based on the target control parameters, making the target parking strategy conform to the actual parking scenario and improving the rationality of the parking strategy.

[0091] In a possible way, adjusting the default trajectory segment length threshold and / or the default parking speed according to the adjustment factor to obtain the target control parameters includes: when the parking difficulty coefficient is greater than the default coefficient, reduce the default trajectory segment length threshold and / or the default parking speed according to the preset interval to which the adjustment factor belongs to obtain the target control parameters; when the parking difficulty coefficient is less than the default coefficient, increase the default trajectory segment length threshold and / or the default parking speed according to the preset interval to which the adjustment factor belongs to obtain the target control parameters.

[0092] Specifically, when the parking-in difficulty coefficient is greater than the default coefficient, if the preset interval to which the adjustment factor belongs is the first preset interval, the default parking speed is reduced according to the adjustment factor to obtain the target parking speed; the target control parameter includes the target parking speed and the default trajectory segment length threshold.

[0093] If the adjustment factor belongs to the second preset interval, the default trajectory segment length threshold is reduced according to the adjustment factor to obtain the target trajectory segment length threshold; the target control parameter includes: the default parking speed and the target trajectory segment length threshold.

[0094] If the adjustment factor belongs to the third preset interval, the default trajectory segment length threshold and the default parking speed are reduced according to the adjustment factor to obtain the target control parameter; the target control parameter includes: the target parking speed and the target trajectory segment length threshold.

[0095] When the parking-in difficulty coefficient is less than the default coefficient, if the preset interval to which the adjustment factor belongs is the first preset interval, the default parking speed is increased according to the adjustment factor to obtain the target parking speed; the target control parameter includes the target parking speed and the default trajectory segment length threshold.

[0096] If the adjustment factor belongs to the second preset interval, the default trajectory segment length threshold is increased according to the adjustment factor to obtain the target trajectory segment length threshold; the target control parameter includes: the default parking speed and the target trajectory segment length threshold.

[0097] If the adjustment factor belongs to the third preset interval, the default trajectory segment length threshold and the default parking speed are increased according to the adjustment factor to obtain the target control parameter; the target control parameter includes: the target parking speed and the target trajectory segment length threshold.

[0098] Among them, reducing (or increasing) the default trajectory segment length threshold according to the adjustment factor is to obtain the length adjustment value corresponding to the adjustment factor, and reducing (or increasing) the default trajectory segment length threshold according to the length adjustment value; reducing (or increasing) the default parking speed according to the adjustment factor is to obtain the speed adjustment value corresponding to the adjustment factor, and reducing (or increasing) the default trajectory segment length threshold according to the speed adjustment value.

[0099] Exemplarily, the parking scenario is as Figure 2 shown. The identified target scenario elements include: the left closed wall and the right column. Combining with the cloud for scenario matching, the parking-in difficulty coefficient is determined to be 7, and the default coefficient is 5. Since the parking-in difficulty coefficient is greater than the default coefficient, the default parking speed and the target trajectory segment length threshold are reduced according to the parking-in difficulty coefficient.

[0100] Exemplarily, the parking scenario is as Figure 3As shown, the target parking space is adjacent to an external road, which is a drivable space, and there is no obstruction between the area where the target parking space is located and the external road. The default parking trajectory obtained under the constraint of default regulation parameters includes trajectory segment 1 and trajectory segment 2. It can be seen that parts of trajectory segment 1 and trajectory segment 2 belong to the external road, which may cause the vehicle to collide with objects on the external road during the automatic parking process.

[0101] As Figure 4 shown, the embodiment of the present application identifies that the target scenario element includes an external road, and jointly obtains a relatively high parking-in difficulty coefficient from the cloud. According to the parking-in difficulty coefficient, the default parking speed and the length threshold of the target trajectory segment are reduced, thereby shortening the length of a single trajectory segment in the target parking strategy. It can be seen that neither trajectory segment 1 nor trajectory segment 2 included in the target parking trajectory is on the external road, avoiding the situation of collision with objects on the external road during the automatic parking process and improving the safety of automatic parking.

[0102] In the above embodiment, the default regulation parameters are adjusted according to the parking-in difficulty coefficient to obtain the target regulation parameters, which improves the rationality of the parking strategy. In high-difficulty parking scenarios, the safety of automatic parking is improved, and in low-difficulty parking scenarios, the efficiency of automatic parking can be improved.

[0103] In a possible implementation manner, the target parking strategy includes a target parking trajectory; after generating the target parking strategy based on the target regulation parameters, it further includes: when the target scenario element and the target parking trajectory meet the intersection condition, adjusting the target parking trajectory included in the target parking strategy based on the target scenario element to obtain a reference parking strategy; the reference parking strategy includes a reference parking trajectory; correspondingly, parking the vehicle into the target parking space using the target parking strategy includes: parking the vehicle into the target parking space using the reference parking strategy.

[0104] Among them, the target scenario element and the target parking trajectory meeting the intersection condition means that the position of the target scenario element is the same as the position of the trajectory point of the target parking trajectory; that is, when the position of the target scenario element is the same as the position of a certain trajectory point in the target parking trajectory, it is determined that the target scenario element and the target parking trajectory meet the intersection condition.

[0105] Specifically, the vehicle terminal obtains the position of the target scenario element and the position of the trajectory point of the target parking trajectory, and determines whether the target scenario element and the target parking trajectory meet the intersection condition according to the position of the target scenario element and the position of the trajectory point. When the target scenario element and the target parking trajectory meet the intersection condition, the target parking trajectory included in the target parking strategy is adjusted based on the target scenario element to obtain a reference parking strategy.

[0106] Optionally, adjusting the target parking trajectory included in the target parking strategy based on the target scenario element may be to input the position of the target scenario element, the current position of the vehicle, the target parking space, and the sensor data into the planning and control model, and obtain a reference parking trajectory through the planning and control model; input the position of the target scenario element into the planning and control model, so that the planning and control model re-determines the drivable area according to the position of the target scenario element, and then performs parking planning in the drivable area, so that the output reference parking trajectory does not intersect with the target scenario element.

[0107] Optionally, adjusting the target parking trajectory included in the target parking strategy based on the target scenario element may be to determine the first trajectory segment that intersects with the target element in the target parking trajectory according to the position of the target scenario element, use a trajectory segment after the first trajectory segment in the target parking trajectory as the second trajectory segment, adjust the length of the first trajectory segment according to the position of the target scenario element to obtain a first reference trajectory segment, so that the first reference trajectory segment does not intersect with the target scenario element; determine a second reference trajectory segment according to the end point of the first reference trajectory segment and the end point of the second trajectory segment; replace the first trajectory segment in the target parking trajectory with the first reference trajectory segment and the second trajectory segment with the second reference trajectory segment to obtain a reference parking trajectory.

[0108] In the above embodiment, when the target scenario element and the target parking trajectory meet the intersection condition, the target parking trajectory is adjusted so that the reference parking trajectory does not intersect with the target scenario element, reducing the collision risk of automatic parking.

[0109] In a possible implementation manner, after the vehicle is parked in the target parking space by adopting the target parking strategy, it further includes: generating a target parking record corresponding to the reference scenario, and sending the target parking record corresponding to the reference scenario to the cloud, so that the cloud determines the target difficulty coefficient of the reference scenario based on the target parking record; when the target difficulty coefficient is different from the parking difficulty coefficient, obtaining candidate scenarios based on the difficulty interval to which the target difficulty coefficient belongs; determining the target elements in the reference scenario based on the candidate scenarios and the reference scenario; adjusting the difficulty weight of the target elements based on the target difficulty coefficient and the parking difficulty coefficient to obtain a target difficulty weight; updating the parking difficulty coefficient of the reference scenario based on the target difficulty weight.

[0110] Among them, the target parking record includes the actual parking strategy for parking the vehicle in the target parking space; it should be noted that during the automatic parking process, re-planning may be performed due to obstacle avoidance requirements, resulting in the actual parking strategy being different from the target parking strategy.

[0111] The cloud can use a scoring model to estimate the difficulty of the target parking record of the reference scenario, obtain the target difficulty coefficient of the reference scenario. When the target difficulty coefficient is different from the parking-in difficulty coefficient, if the target difficulty coefficient belongs to the high-difficulty interval, candidate scenarios with an actual difficulty coefficient belonging to the high-difficulty interval are obtained; if the target difficulty coefficient belongs to the low-difficulty interval, candidate scenarios with an actual difficulty coefficient belonging to the low-difficulty interval are obtained; the actual difficulty coefficient is determined according to the parking records of the candidate scenarios; the number of candidate scenarios can be multiple.

[0112] The cloud analyzes the elements included in the candidate scenarios and the elements included in the reference scenario, and takes the elements with a frequency of occurrence greater than the preset frequency as target elements; it can be understood that, for example, for a reference scenario and candidate scenarios belonging to the high-difficulty interval, if a certain element is included in both the reference scenario and most of the candidate scenarios, then this element may be the target element that causes the high parking difficulty.

[0113] The cloud calculates the difference between the target difficulty coefficient and the parking-in difficulty coefficient, determines the adjustment amount corresponding to the difference. When the target difficulty coefficient is greater than the parking-in difficulty coefficient, the difficulty weight of the target element is increased according to the adjustment amount to obtain the target difficulty weight; when the target difficulty coefficient is less than the parking-in difficulty coefficient, the difficulty weight of the target element is decreased according to the adjustment amount to obtain the target difficulty weight; there can be a positive correlation between the difference and the adjustment amount.

[0114] The cloud determines the target parking-in difficulty coefficient according to the target difficulty weight of the target element in the reference scenario and the difficulty weights of other preset elements except the target element, and replaces the parking-in difficulty coefficient of the reference scenario with the target parking-in difficulty coefficient to update the parking-in difficulty coefficient of the reference scenario.

[0115] In the above embodiment, scenarios in the same difficulty interval are combined, the difficulty weights of the elements in the scenarios are updated, and then the parking-in difficulty coefficients of the scenarios are dynamically updated, so as to associate the difficulty weights of the elements with the parking-in difficulty coefficients of the scenarios, improve the comprehensive reasoning ability of the elements, scenarios, and parking difficulty, make the parking-in difficulty coefficients of the scenarios more and more accurate, and thus make the subsequent obtained target parking strategies more accurate and reliable.

[0116] The automatic parking control method provided by the embodiment of the present application enables the vehicle end to identify target scene elements within the parking scene area corresponding to the target parking space, and jointly perform scene matching with the cloud to determine the reference scene where the target parking space is located. The default regulation and control parameters are adjusted using the parking difficulty coefficient corresponding to the reference scene to obtain the target regulation and control parameters, and a target parking strategy is generated based on the target regulation and control parameters. In this way, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same. However, in different parking scenarios, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficients of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, making the target parking strategy conform to the actual parking scenario, improving the rationality of the parking strategy, avoiding the collision risk caused by using the default parking strategy for automatic parking in high-difficulty parking scenarios, enhancing the safety of automatic parking, and improving the efficiency of automatic parking in low-difficulty parking scenarios.

[0117] Figure 5 The flowchart of the second embodiment of the automatic parking control method of the present invention is shown, and this method is executed by the cloud. As Figure 5 shown, this method includes the following steps:

[0118] Step 510: In response to the scene matching request sent by the vehicle end, obtain the target scene elements included in the scene matching request; the scene matching request is sent by the vehicle end in response to the parking space detection instruction, based on the acquired sensor data to determine the target parking space and target scene elements, and in response to the parking start instruction for the target parking space, and is sent to the cloud based on the target scene elements; the target scene elements are within the parking scene area corresponding to the target parking space.

[0119] Step 520: Determine the reference scene that matches the target scene elements among multiple preset parking scenes;

[0120] Step 530: Obtain the parking difficulty coefficient corresponding to the reference scene, and send the parking difficulty coefficient to the vehicle end, so that the vehicle end adjusts the default regulation and control parameters based on the parking difficulty coefficient to obtain the target regulation and control parameters, generates a target parking strategy based on the target regulation and control parameters, and parks the vehicle into the target parking space using the target parking strategy.

[0121] Specifically, the user triggers a parking space detection instruction by performing a parking space detection operation. In response to the parking space detection instruction, the vehicle terminal obtains sensor data for parking space recognition and element recognition, obtaining available parking spaces and scene elements. The vehicle terminal displays the available parking spaces so that the user can perform a selection operation on one of the available parking spaces. The vehicle terminal takes the available parking space pointed to by the selection operation as the target parking space, determines the parking scene area corresponding to the target parking space, and selects the target scene elements belonging to the parking scene area from the scene elements; in response to a parking start instruction, the vehicle terminal generates a scene matching request based on the target scene elements and sends the scene matching request to the cloud.

[0122] The cloud receives the scene matching request and obtains the target scene elements included in the scene matching request; determines a reference scene that matches the target scene elements among multiple preset parking scenes, obtains the parking difficulty coefficient corresponding to the reference scene, and sends the parking difficulty coefficient to the vehicle terminal.

[0123] The vehicle terminal receives the parking difficulty coefficient sent by the cloud, determines whether the parking difficulty coefficient is the same as the default coefficient. If the parking difficulty coefficient is different from the default coefficient, the default regulation and control parameters are adjusted according to the parking difficulty coefficient; when the parking difficulty coefficient is greater than the default coefficient, the default trajectory segment length threshold and the default parking speed are reduced to obtain the target regulation and control parameters; when the parking difficulty coefficient is less than the default coefficient, the default trajectory segment length threshold and the default parking speed are increased to obtain the target regulation and control parameters; the planning and control model performs parking planning based on the current position of the vehicle, the target parking space, and the sensor data under the constraint of the target regulation and control parameters, obtaining a target parking strategy, and parks the vehicle into the target parking space using the target parking strategy.

[0124] For the specific processes of the above steps 510 to 530, reference can be made to the descriptions of the above steps 110 to 140.

[0125] The automatic parking control method provided by the embodiment of the present application enables the vehicle end to identify target scene elements within the parking scene area corresponding to the target parking space, and jointly perform scene matching with the cloud to determine the reference scene where the target parking space is located. The default regulation and control parameters are adjusted using the parking difficulty coefficient corresponding to the reference scene to obtain the target regulation and control parameters, and a target parking strategy is generated based on the target regulation and control parameters. In this way, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same. However, in different parking scenarios, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficient of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, making the target parking strategy conform to the actual parking scenario, improving the rationality of the parking strategy. In high-difficulty parking scenarios, the collision risk caused by using the default parking strategy for automatic parking is avoided, enhancing the safety of automatic parking. In low-difficulty parking scenarios, the efficiency of automatic parking can be improved.

[0126] In an alternative manner, determining a reference scene that matches the target scene elements among multiple preset parking scenarios includes: for each preset parking scenario, obtaining the preset elements in the targeted preset parking scenario, and determining the matching degree of the targeted preset parking scenario based on the similarity between the preset elements and the target scene elements; selecting a reference scene from among the multiple preset parking scenarios according to the matching degrees of the multiple preset parking scenarios.

[0127] Specifically, the number of target scene elements can be multiple; for each preset parking scenario, the preset parking scenario includes multiple preset elements. For each target scene element, the first candidate similarity between the target scene element and each preset element is determined respectively, and the average of the multiple first candidate similarities is obtained to get the second candidate similarity corresponding to the target scene element. In the same way, the second candidate similarities corresponding to all target scene elements can be determined, and then the average of the multiple second candidate similarities is obtained to get the matching degree between the target scene elements and the preset parking scenario.

[0128] In the same way as above, the matching degrees between the target scene elements and multiple preset parking scenarios can be obtained; the preset parking scenario corresponding to the highest matching degree among the matching degrees corresponding to the multiple preset parking scenarios is used as the reference scene; the cloud obtains the parking difficulty coefficient corresponding to the reference scene and sends the parking difficulty coefficient to the vehicle end.

[0129] In the above embodiment, by performing scene matching at the element level, a reference scene is determined among multiple preset parking scenarios, improving the accuracy of scene matching. Furthermore, the parking difficulty coefficient of the reference scene matches the parking scenario of the target parking space, enhancing the reliability of automatic parking.

[0130] In an alternative approach, the automatic parking control method further includes: after parking the vehicle into the target parking space, receiving the target parking record corresponding to the reference scenario sent by the vehicle end; determining the target difficulty coefficient of the reference scenario based on the target parking record; when the target difficulty coefficient is different from the parking-in difficulty coefficient, obtaining candidate scenarios based on the difficulty interval to which the target difficulty coefficient belongs; determining the target elements in the reference scenario based on the candidate scenarios and the reference scenario; adjusting the difficulty weight of the target elements based on the target difficulty coefficient and the parking-in difficulty coefficient to obtain the target difficulty weight; and updating the parking-in difficulty coefficient of the reference scenario based on the target difficulty weight.

[0131] Among them, the target parking record includes the actual parking strategy for parking the vehicle into the target parking space; it should be noted that during the automatic parking process, replanning may be performed due to obstacle avoidance requirements, which may lead to differences between the actual parking strategy and the target parking strategy.

[0132] The cloud can use a scoring model to estimate the difficulty of the target parking record of the reference scenario to obtain the target difficulty coefficient of the reference scenario. When the target difficulty coefficient is different from the parking-in difficulty coefficient, when the target difficulty coefficient belongs to the high-difficulty interval, candidate scenarios with an actual difficulty coefficient belonging to the high-difficulty interval are obtained; when the target difficulty coefficient belongs to the low-difficulty interval, candidate scenarios with an actual difficulty coefficient belonging to the low-difficulty interval are obtained; the actual difficulty coefficient is determined according to the parking records of the candidate scenarios; and the number of candidate scenarios can be multiple.

[0133] As Figure 6 shown, the cloud analyzes the environmental elements included in multiple candidate scenarios and the environmental elements included in the reference scenario, and takes the environmental elements with a frequency of occurrence greater than the preset frequency as the target elements; it can be understood that, for example, for a reference scenario and candidate scenarios belonging to the high-difficulty interval, if a certain element is included in both the reference scenario and most of the candidate scenarios, then this element may be the target element that causes the high parking difficulty.

[0134] The cloud calculates the difference between the target difficulty coefficient and the parking-in difficulty coefficient, determines the adjustment amount corresponding to the difference. When the target difficulty coefficient is greater than the parking-in difficulty coefficient, the difficulty weight of the target elements is increased according to the adjustment amount to obtain the target difficulty weight; when the target difficulty coefficient is less than the parking-in difficulty coefficient, the difficulty weight of the target elements is decreased according to the adjustment amount to obtain the target difficulty weight; and there may be a positive correlation between the difference and the adjustment amount.

[0135] The cloud determines the target parking-in difficulty coefficient based on the target difficulty weight of the target elements in the reference scenario and the difficulty weights of other preset elements except the target elements, and replaces the parking-in difficulty coefficient of the reference scenario with the target parking-in difficulty coefficient to achieve the update of the parking-in difficulty coefficient of the reference scenario.

[0136] In the above embodiments, by combining scenarios in the same difficulty range, the difficulty weights of the elements in the scenario are updated, and then the parking-in difficulty coefficient of the scenario is dynamically updated, so as to associate the difficulty weights of the elements with the parking-in difficulty coefficient of the scenario, improving the comprehensive reasoning ability of the difficulty of elements, scenarios, and parking, making the parking-in difficulty coefficient of the scenario more and more accurate, and thus also making the subsequent obtained target parking strategy more accurate and reliable.

[0137] The automatic parking control method provided by the embodiments of the present application is such that the vehicle end identifies target scenario elements within the parking scenario area corresponding to the target parking space, and jointly performs scenario matching with the cloud to determine the reference scenario where the target parking space is located. The default regulation and control parameters are adjusted using the parking-in difficulty coefficient corresponding to the reference scenario to obtain the target regulation and control parameters, and a target parking strategy is generated based on the target regulation and control parameters. In this way, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same. However, in different parking scenarios, the default regulation and control parameters are adaptively adjusted according to the parking-in difficulty coefficient of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, making the target parking strategy conform to the actual parking scenario, improving the rationality of the parking strategy, avoiding the collision risk caused by using the default parking strategy for automatic parking in high-difficulty parking scenarios, improving the safety of automatic parking, and improving the efficiency of automatic parking in low-difficulty parking scenarios.

[0138] The present invention also provides an automatic parking control system. Figure 7 The flowchart showing the execution of the automatic parking control method by the automatic parking control system is as follows; the automatic parking control system includes:

[0139] The vehicle end 710 is configured to, in response to a parking space detection instruction, determine a target parking space and target scenario elements based on the acquired sensor data; the target scenario elements are in the parking scenario area corresponding to the target parking space; in response to a parking start instruction, send a scenario matching request to the cloud based on the target scenario elements.

[0140] The cloud 720 is configured to, in response to the scenario matching request sent by the vehicle end, acquire the target scenario elements included in the scenario matching request; determine a reference scenario that matches the target scenario elements among multiple preset parking scenarios; acquire the parking-in difficulty coefficient corresponding to the reference scenario, and send the parking-in difficulty coefficient to the vehicle end.

[0141] The vehicle end 710 is further configured to receive the parking-in difficulty coefficient sent by the cloud, adjust the default regulation and control parameters based on the parking-in difficulty coefficient to obtain the target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters; park the vehicle into the target parking space using the target parking strategy.

[0142] Exemplarily, such as Figure 8As shown in the figure, the vehicle end includes an automatic parking activation unit and a perception fusion unit; the parking activation unit is used to obtain sensor data in response to a parking space detection instruction; the perception fusion unit is used to determine a target parking space and target scene elements based on the obtained sensor data.

[0143] The cloud end includes a scene matching unit. The scene matching unit can determine a reference scene that matches the target scene elements from the preset scenes included in the scene library, and send the parking difficulty coefficient corresponding to the reference scene to the vehicle end.

[0144] The vehicle end further includes a regulation and control parameter adjustment unit and a parking control unit, which are used to adjust the default regulation and control parameters according to the parking difficulty coefficient to obtain target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters; the parking control unit is used to park the vehicle into the target parking space by adopting the target parking strategy.

[0145] In this embodiment, the specific process executed by the vehicle end can refer to the description of steps 110 to 140 above; the specific process executed by the cloud end can refer to the description of steps 510 to 530 above.

[0146] The automatic parking control system provided by the embodiment of the present application enables the vehicle end to identify the target scene elements in the parking scene area corresponding to the target parking space, and jointly perform scene matching with the cloud end to determine the reference scene where the target parking space is located, adjust the default regulation and control parameters by using the parking difficulty coefficient corresponding to the reference scene to obtain the target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters; thus, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same, but in different parking scenes, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficient of different parking scenes, and a target parking strategy will be generated based on the target regulation and control parameters, so that the target parking strategy conforms to the actual parking scene, improving the rationality of the parking strategy, avoiding the collision risk caused by using the default parking strategy for automatic parking in high-difficulty parking scenes, enhancing the safety of automatic parking, and improving the efficiency of automatic parking in low-difficulty parking scenes.

[0147] Figure 9 The figure shows a schematic structural diagram of an embodiment of the vehicle end provided by the present invention. As Figure 9 shown, the vehicle end 900 includes: an identification module 910, a request module 920, a regulation and control parameter adjustment module 930, and a parking-in execution module 940.

[0148] The identification module 910 is used to determine a target parking space and target scene elements based on the obtained sensor data in response to a parking space detection instruction; the target scene elements are in the parking scene area corresponding to the target parking space;

[0149] A request module 920, configured to send a scene matching request to the cloud based on target scene elements in response to a parking start instruction; the scene matching request is used to instruct the cloud to determine a reference scene that matches the target scene elements among multiple preset parking scenes and send the parking difficulty coefficient corresponding to the reference scene to the vehicle terminal;

[0150] A regulation and control parameter adjustment module 930, configured to receive the parking difficulty coefficient sent by the cloud, adjust the default regulation and control parameters based on the parking difficulty coefficient to obtain target regulation and control parameters, and generate a target parking strategy based on the target regulation and control parameters;

[0151] A parking execution module 940, configured to park the vehicle into a target parking space by using the target parking strategy.

[0152] In an optional manner, the recognition module 910 is further configured to perform target recognition based on the acquired sensor data to obtain available parking spaces and candidate scene elements; in response to a selection operation, determine a target parking space among the available parking spaces; determine a parking scene area corresponding to the target parking space; and determine target scene elements among the candidate scene elements based on the parking scene area.

[0153] In an optional manner, the default regulation and control parameters include a default trajectory segment length threshold and a default parking speed; the regulation and control parameter adjustment module 930 is further configured to, when the parking difficulty coefficient is different from the default coefficient, determine an adjustment factor based on the parking difficulty coefficient and the default coefficient; and adjust the default trajectory segment length threshold and / or the default parking speed according to the adjustment factor to obtain the target regulation and control parameters.

[0154] In an optional manner, the regulation and control parameter adjustment module 930 is further configured to, when the parking difficulty coefficient is greater than the default coefficient, reduce the default trajectory segment length threshold and / or the default parking speed by using the adjustment factor according to the preset interval to which the adjustment factor belongs to obtain the target regulation and control parameters; when the parking difficulty coefficient is less than the default coefficient, increase the default trajectory segment length threshold and / or the default parking speed by using the adjustment factor according to the preset interval to which the adjustment factor belongs to obtain the target regulation and control parameters. When the parking difficulty coefficient is less than the default coefficient, increase the default trajectory segment length threshold based on the length adjustment value corresponding to the adjustment factor to obtain the target trajectory segment length threshold, and increase the default parking speed based on the speed adjustment value corresponding to the adjustment factor to obtain the target parking speed.

[0155] The vehicle end provided by the embodiment of the present application identifies target scene elements within the parking scenario area corresponding to the target parking space, and jointly performs scene matching with the cloud to determine the reference scene where the target parking space is located. The default regulation and control parameters are adjusted by using the parking difficulty coefficient corresponding to the reference scene to obtain the target regulation and control parameters, and a target parking strategy is generated based on the target regulation and control parameters. In this way, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same. However, in different parking scenarios, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficients of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, so that the target parking strategy conforms to the actual parking scenario, improving the rationality of the parking strategy. In high-difficulty parking scenarios, the collision risk caused by using the default parking strategy for automatic parking is avoided, improving the safety of automatic parking. In low-difficulty parking scenarios, the efficiency of automatic parking can be improved.

[0156] Figure 10 The structural schematic diagram of the embodiment of the cloud provided by the present invention is shown. As Figure 10 shown, the cloud 1000 includes: an acquisition module 1010, a scene matching module 1020, and a difficulty coefficient sending module 1030.

[0157] The acquisition module 1010 is configured to acquire the target scene elements included in the scene matching request in response to the scene matching request sent by the vehicle end; the scene matching request is sent by the vehicle end based on the acquired sensor data to determine the target parking space and the target scene elements in response to the parking space detection instruction, and in response to the parking start instruction for the target parking space, and is sent to the cloud based on the target scene elements; the target scene elements are within the parking scenario area corresponding to the target parking space.

[0158] The scene matching module 1020 is configured to determine the reference scene that matches the target scene elements among multiple preset parking scenarios.

[0159] The difficulty coefficient sending module 1030 is configured to acquire the parking difficulty coefficient corresponding to the reference scene, and send the parking difficulty coefficient to the vehicle end, so that the vehicle end adjusts the default regulation and control parameters based on the parking difficulty coefficient to obtain the target regulation and control parameters, and generates a target parking strategy based on the target regulation and control parameters, and parks the vehicle into the target parking space by using the target parking strategy.

[0160] In an optional manner, the scene matching module 1020 is configured to, for each preset parking scenario, acquire the preset elements in the preset parking scenario targeted, determine the matching degree of the preset parking scenario targeted based on the similarity between the preset elements and the target scene elements; and select the reference scene among multiple preset parking scenarios according to the matching degrees of multiple preset parking scenarios.

[0161] In an alternative manner, the cloud further includes: a difficulty coefficient update module, configured to receive a target parking record corresponding to a reference scenario sent by the vehicle terminal after parking the vehicle into a target parking space; determine a target difficulty coefficient of the reference scenario based on the target parking record; when the target difficulty coefficient is different from the parking difficulty coefficient, obtain candidate scenarios based on the difficulty interval to which the target difficulty coefficient belongs; determine target elements in the reference scenario based on the candidate scenarios and the reference scenario; adjust the difficulty weights of the target elements based on the target difficulty coefficient and the parking difficulty coefficient to obtain target difficulty weights; and update the parking difficulty coefficient of the reference scenario based on the target difficulty weights.

[0162] The cloud provided in the embodiment of the present application determines a reference scenario that matches the target scenario elements in a preset scenario according to a scenario matching request sent by the vehicle terminal, and sends the parking difficulty coefficient corresponding to the reference scenario to the vehicle terminal, so that the vehicle terminal adjusts the default regulation and control parameters by using the parking difficulty coefficient corresponding to the reference scenario to obtain target regulation and control parameters, and generates a target parking strategy based on the target regulation and control parameters; thus, when the relative position between the current position of the vehicle and the target parking space is the same, the size of the target parking space is the same, and the width of the driving lane is the same, the default parking strategy is the same, but in different parking scenarios, the default regulation and control parameters will be adaptively adjusted according to the parking difficulty coefficients of different parking scenarios, and a target parking strategy is generated based on the target regulation and control parameters, so that the target parking strategy conforms to the actual parking scenario, improving the rationality of the parking strategy. In high-difficulty parking scenarios, the collision risk caused by using the default parking strategy for automatic parking is avoided, improving the safety of automatic parking. In low-difficulty parking scenarios, the efficiency of automatic parking can be improved.

[0163] Figure 11 The structural schematic diagram of the embodiment of the electronic device of the present invention is shown. The specific implementation of the electronic device in the specific embodiment of the present invention is not limited.

[0164] As Figure 11 shown, the electronic device may include: a processor 1102, a communication interface 1104, a memory 1106, and a communication bus 1108.

[0165] Among them: the processor 1102, the communication interface 1104, and the memory 1106 communicate with each other through the communication bus 1108. The communication interface 1104 is used to communicate with network elements of other devices such as clients or other servers. The processor 1102 is used to execute the program 1110, and specifically may execute the relevant steps in the embodiment of the automatic parking control method described above.

[0166] Specifically, the program 1110 may include program code that includes computer-executable instructions.

[0167] The processor 1102 may be a central processing unit (CPU), or a specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the electronic device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0168] The memory 1106 is used to store the program 1110. The memory 1106 may include high-speed RAM memory and may also include non-volatile memory, such as at least one disk memory.

[0169] The program 1110 can specifically be called by the processor 1102 to cause the electronic device to perform the operations of the automatic parking control method in any of the above method embodiments.

[0170] The embodiments of the present invention provide a computer-readable storage medium storing at least one executable instruction, which, when running on an electronic device, causes the electronic device to perform the operations of the automatic parking control method in any of the above method embodiments.

[0171] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. In addition, the embodiments of the present invention are not directed to any specific programming language.

[0172] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. Similarly, in order to streamline the present invention and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. Among them, the claims following the specific implementation manners are hereby expressly incorporated into the specific implementation manners, and each claim itself serves as a separate embodiment of the present invention.

[0173] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and set in one or more devices different from those of the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive.

[0174] It should be noted that the above embodiments illustrate the present invention rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. An automatic parking control method, characterized in that: The method comprises: In response to the parking space detection instruction, determining a target parking space and a target scene element based on the acquired sensor data; the target scene element is located in a parking scene area corresponding to the target parking space; In response to the parking start instruction, a scene matching request is sent to the cloud based on the target scene element; the scene matching request is used to instruct the cloud to determine a reference scene matching the target scene element from a plurality of preset parking scenes, and send a parking difficulty coefficient corresponding to the reference scene to the vehicle end; receiving the parking difficulty coefficient sent by the cloud, adjusting a default regulation and control parameter based on the parking difficulty coefficient to obtain a target regulation and control parameter, and generating a target parking strategy based on the target regulation and control parameter; The target parking strategy is adopted to park the vehicle into the target parking space.

2. The method according to claim 1, characterized in that The determining of the target parking space and the target scene element based on the acquired sensor data includes: Perform target recognition based on the acquired sensor data to obtain available parking spaces and candidate scene elements; In response to the selection operation, determining a target parking space among the available parking spaces; Determine a parking scene area corresponding to the target parking space; Based on the parking scene area, a target scene element is determined from among the candidate scene elements.

3. The method according to claim 1 or 2, characterized in that: The default control parameters include a default trajectory segment length threshold and a default parking speed; The adjusting the default regulatory control parameters based on the parking difficulty coefficient to obtain the target regulatory control parameters includes: When the parking difficulty coefficient is different from a default coefficient, determining an adjustment factor based on the parking difficulty coefficient and the default coefficient; The default trajectory segment length threshold and / or the default parking speed are adjusted according to the adjustment factor to obtain a target regulation parameter.

4. The method according to claim 3, characterized in that The step of adjusting the default trajectory segment length threshold and / or the default parking speed according to the adjustment factor to obtain a target regulation parameter includes: When the parking difficulty coefficient is greater than the default coefficient, according to the preset interval to which the adjustment factor belongs, the default trajectory segment length threshold and / or the default parking speed are reduced by using the adjustment factor to obtain a target regulation parameter; When the parking difficulty coefficient is less than the default coefficient, the default trajectory segment length threshold and / or the default parking speed are increased by the adjustment factor according to the preset interval to which the adjustment factor belongs, so as to obtain a target regulation parameter.

5. An automatic parking control method, characterized in that: include: In response to a scene matching request sent by the vehicle end, obtaining a target scene element included in the scene matching request; The scene matching request is sent by the vehicle end to the cloud based on the target scene element in response to the parking space detection instruction, determining the target parking space and the target scene element based on the acquired sensor data, and responding to the parking start instruction for the target parking space; the target scene element is in the parking scene area corresponding to the target parking space; Determining a reference scene matching the target scene element from a plurality of preset parking scenes; A parking difficulty coefficient corresponding to the reference scenario is obtained, and the parking difficulty coefficient is sent to a vehicle end, so that the vehicle end adjusts a default regulation and control parameter based on the parking difficulty coefficient to obtain a target regulation and control parameter, and generates a target parking strategy based on the target regulation and control parameter, and adopts the target parking strategy to park the vehicle in the target parking space.

6. The method according to claim 5, characterized in that The determining a reference scene matching the target scene element from a plurality of preset parking scenes includes: For each preset parking scene, obtaining a preset element in the preset parking scene, and determining a matching degree of the preset parking scene based on a similarity between the preset element and the target scene element; According to the matching degree of the plurality of preset parking scenes, a reference scene is selected from the plurality of preset parking scenes.

7. The method according to claim 5 or 6, characterized in that: The method further comprises: After parking the vehicle in the target parking space, receiving a target parking record corresponding to the reference scene sent by the vehicle end; determining a target difficulty coefficient of the reference scenario based on the target parking record; When the target difficulty coefficient is different from the parking difficulty coefficient, acquiring a candidate scene based on the difficulty interval to which the target difficulty coefficient belongs; Based on the candidate scene and the reference scene, determining a target element in the reference scene; Based on the target difficulty coefficient and the parking difficulty coefficient, adjusting the difficulty weight of the target element to obtain a target difficulty weight; The parking difficulty coefficient of the reference scene is updated based on the target difficulty weight.

8. A vehicle end, characterized in that: include: A recognition module, for determining a target parking space and a target scene element based on acquired sensor data in response to a parking space detection instruction; The target scene element is in a parking scene area corresponding to the target parking space; A request module, configured to send a scene matching request to the cloud based on the target scene element in response to a parking start instruction; The scene matching request is used to instruct the cloud to determine a reference scene matching the target scene element from a plurality of preset parking scenes, and send the parking difficulty coefficient corresponding to the reference scene to the vehicle end; a regulatory parameter adjustment module, configured to receive the parking difficulty coefficient sent by the cloud, adjust default regulatory parameters based on the parking difficulty coefficient to obtain target regulatory parameters, and generate a target parking strategy based on the target regulatory parameters; The parking execution module is used to park the vehicle into the target parking space by adopting the target parking strategy.

9. A cloud, characterized in that: include: An acquisition module, configured to respond to a scene matching request sent by the vehicle end and acquire a target scene element included in the scene matching request; The scene matching request is sent by the vehicle end to the cloud based on the target scene element in response to the parking space detection instruction, determining the target parking space and the target scene element based on the acquired sensor data, and responding to the parking start instruction for the target parking space; the target scene element is in the parking scene area corresponding to the target parking space; A scene matching module, used to determine a reference scene matching the target scene element from a plurality of preset parking scenes; The difficulty coefficient sending module is used to obtain the parking difficulty coefficient corresponding to the reference scene, and send the parking difficulty coefficient to the vehicle end, so that the vehicle end adjusts the default regulation and control parameters based on the parking difficulty coefficient to obtain the target regulation and control parameters, and generates a target parking strategy based on the target regulation and control parameters, and adopts the target parking strategy to park the vehicle in the target parking space.

10. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the automatic parking control method as described in any one of claims 1-7.

Images