Parking control method, electronic equipment and storage medium

Abstract

The invention provides a parking control method, electronic equipment and a storage medium, and the parking control method comprises the steps that a parking intention and surrounding image information of a vehicle are acquired, and the parking intention is used for indicating and controlling the parking demand of the vehicle through structured information; performing parking dynamic matching according to the parking intention and the surrounding image information of the vehicle to obtain a plurality of candidate parking areas; if the confidence coefficient of any candidate parking area meets a preset confidence coefficient threshold value, determining any candidate parking area as a target parking area; if the confidence coefficients corresponding to the multiple candidate parking areas do not meet a preset confidence coefficient threshold value, a target parking area is determined from the multiple candidate parking areas based on the surrounding image information, the parking intention, historical parking records of the vehicle and preset weight parameters; and parking planning is carried out based on the target parking area, and the vehicle is controlled to run to the target parking area according to a corresponding planning result. According to the invention, the accuracy of vehicle parking control can be improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a parking control method, electronic equipment, and storage medium. Background Art

[0002] As the number of motor vehicles increases, the parking environment is becoming increasingly complex. Drivers typically observe the vehicle's surroundings through rearview mirrors or side mirrors to ensure the vehicle can park in a suitable parking space based on the correct movement path. To reduce the impact of blind spots on parking safety, drivers can activate the vehicle's automatic parking function through the human-vehicle interaction interface, driving the vehicle to the appropriate parking space according to the driver's selection. However, this automatic parking method is too cumbersome to operate and suffers from low accuracy in parking space selection when controlling the vehicle to park. Summary of the Invention

[0003] In view of the above, it is necessary to propose a parking control method, electronic device and storage medium to solve the technical problem of low accuracy in selecting a parking space when controlling vehicle parking.

[0004] The present application provides a parking control method, the method comprising: obtaining a parking intention and surrounding image information of a vehicle, the parking intention being used to indicate a need to control the vehicle to park via structured information; performing dynamic parking matching based on the parking intention and the surrounding image information of the vehicle to obtain multiple candidate parking areas; determining any candidate parking area as a target parking area if a confidence level of any candidate parking area meets a preset confidence threshold; if the confidence levels corresponding to the multiple candidate parking areas do not meet the preset confidence threshold, determining a target parking area from the multiple candidate parking areas based on the surrounding image information, the parking intention, the vehicle's historical parking record, and a preset weight parameter; performing parking planning based on the target parking area, and controlling the vehicle to travel to the target parking area according to the corresponding planning result.

[0005] In some embodiments, determining the target parking area from the multiple candidate parking areas based on the surrounding image information, the parking intention information, the historical parking record of the vehicle, and preset weight parameters includes: determining a score of any one of the multiple candidate parking areas based on the surrounding image information, the parking intention information, the historical parking record of the vehicle, and preset weight parameters; and determining the target parking area from the multiple candidate parking areas based on the score.

[0006] In some embodiments, the method for determining the score of any one of the candidate parking areas includes: determining area information of the any one of the candidate parking areas in the surrounding image information and environmental information of the vehicle based on the surrounding image information; determining a first score of the any one of the candidate parking areas based on the area information, the semantic information of the parking intention, and a first parameter among the weight parameters; determining a second score of the any one of the candidate parking areas based on the semantic information of the parking intention, the environmental information, and a second parameter among the weight parameters; determining a third score of the any one of the candidate parking areas based on the historical parking records, the environmental information, and a third parameter among the weight parameters; and determining the score of the any one of the candidate parking areas based on the first score, the second score, and the third score.

[0007] In some embodiments, the method further includes: determining the weight parameters, including: determining a first parameter in the weight parameters based on the similarity between the semantic information of the parking intention and the area information of any one of the candidate parking areas; determining a second parameter in the weight parameters based on the similarity between the semantic information of the parking intention and the environmental information; and determining a third parameter in the weight parameters based on the similarity between the historical parking records and the environmental information.

[0008] In some embodiments, performing dynamic parking matching based on the parking intention and the surrounding image information of the vehicle to obtain multiple candidate parking areas includes: performing image segmentation on the surrounding image information based on the parking intention to determine a reference target in the surrounding image information; constructing first prompt data based on a preset first prompt word and image information corresponding to the reference target in the surrounding image information; and determining multiple candidate parking areas and the confidence level corresponding to each candidate parking area using a pre-trained generative model based on the first prompt data.

[0009] In some embodiments, the method for determining the parking intention includes: parsing the semantics of pre-acquired parking requirements to obtain corresponding structured semantic information; determining, based on sensor data of the vehicle, position parameters corresponding to the orientation information in the structured semantic information; the position parameters are used to characterize the relative positional relationship between the parking area indicated by the parking requirement and the vehicle; and determining the parking intention based on the structured semantic information and the position parameters.

[0010] In some embodiments, the method further includes: when controlling the vehicle to move according to the moving path indicated by the planning result, if the vehicle meets preset conditions, determining the positional relationship between the vehicle and the target parking area based on real-time image information of the vehicle acquired in real time; updating the moving path based on the positional relationship to obtain an updated moving path; and controlling the vehicle to travel to the target parking area according to the updated moving path.

[0011] In some embodiments, the preset condition includes any one of the following: the speed of the vehicle is greater than or equal to a preset speed threshold; the distance between the vehicle and any target in the real-time image information is less than or equal to a preset distance threshold.

[0012] An embodiment of the present application further provides an electronic device, comprising a processor and a memory, wherein the processor implements the parking control method when executing a computer program stored in the memory.

[0013] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the parking control method is implemented.

[0014] As can be seen from the above technical solutions, the embodiments of the present application analyze the semantics of parking request information to determine the parking intention that controls the vehicle to execute a parking maneuver. The system then characterizes the visual characteristics of the vehicle's surrounding environment based on the vehicle's surrounding image information. Based on the parking intention and visual characteristics, the system determines multiple candidate parking areas around the vehicle, thereby narrowing the scope of the selected parking areas and improving parking efficiency. Furthermore, the system also quantitatively scores the candidate parking areas based on the vehicle's surrounding environmental information and historical parking records, and selects target parking areas based on the scores. This multi-layered screening mechanism determines the parking area for the vehicle, and controls the vehicle's movement to the target parking area based on the relative positional relationship between the target parking area and the vehicle. This improves the accuracy of the parking area determination and enhances the safety of the parking process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is an application scenario diagram of a parking control method provided in one embodiment of the present application.

[0016] Figure 2 This is a flowchart of a parking control method provided by one embodiment of the present application.

[0017] Figure 3 4 is a flowchart of a method for determining multiple candidate parking areas provided by an embodiment of the present application.

[0018] Figure 4This is a flowchart of a method for determining a target parking area provided by an embodiment of the present application.

[0019] Figure 5 4 is a flowchart of a method for determining a score of a candidate parking area provided in one embodiment of the present application.

[0020] Figure 6 This is a flowchart of a method for determining weight parameters provided in one embodiment of the present application.

[0021] Figure 7 This is a schematic diagram of a vehicle movement path provided in an embodiment of the present application.

[0022] Figure 8 This is a flowchart of a method for updating a vehicle's moving path provided in one embodiment of the present application.

[0023] Figure 9 Schematic diagram of an updated moving path provided in one embodiment of the present application.

[0024] Figure 10 2 is a schematic diagram of surrounding image information provided by an embodiment of the present application.

[0025] Figure 11 This is a flowchart of a method for determining parking intention provided by an embodiment of the present application.

[0026] Figure 12 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to more clearly understand the purpose, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other. In the following description, many specific details are set forth to facilitate a full understanding of the present application. The embodiments described are only a part of the embodiments of the present application, rather than all of the embodiments.

[0028] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the described features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein in the specification of this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0030] Embodiments of the present application provide a parking control method that can be applied to one or more electronic devices. An electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes but is not limited to a microprocessor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0031] An electronic device can be any electronic product that can interact with a user, such as a personal computer, tablet computer, smartphone, personal digital assistant (PDA), game console, Internet Protocol Television (IPTV), smart wearable device, etc.

[0032] The electronic device may also include a network device and / or a client device, wherein the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of hosts or network servers.

[0033] The networks where electronic devices are located include but are not limited to the Internet, wide area networks, metropolitan area networks, local area networks, and virtual private networks (VPNs).

[0034] like Figure 1The figure shows an application scenario diagram of a parking control method provided by an embodiment of the present application. The parking control method provided by the present application can be applied to an electronic device 100. The electronic device 100 is set in a vehicle 200, and the electronic device 100 is communicatively connected with other on-board devices (not shown) in the vehicle 200. For example, the other on-board devices may include an on-board camera device, an on-board display, an on-board audio device, a steering device of the vehicle 200, etc., which is not limited in this application. In other embodiments, the electronic device 100 may also be an on-board device of the vehicle 200, and the parking control method is directly applied to the on-board device in the vehicle, which is not limited in actual application.

[0035] like Figure 1 As shown, when the vehicle 200 drives to any area within the parking range, the driver can determine the parking area to be parked based on the location of the vehicle 200 and the surrounding environment information, and output parking requirement information based on natural language. Figure 1 As shown, the area surrounding vehicle 200 includes reference objects 101, 102, 103, parking area 104, and 105. The driver's parking request information may be "park in the space to the left of the red vehicle," or "park in the space to the right of the black vehicle." The red vehicle in the parking request information may be reference object 302, and the black vehicle in the parking request information may be reference object 303. It is understood that the reference object may also be any entity, such as a wall, a cone, or a load-bearing column, and this application does not limit the specific type of reference object.

[0036] In this application, the electronic device 100 determines the corresponding parking intention based on the parking demand information, and determines multiple candidate parking areas based on the surrounding image information of the vehicle 200 and the parking intention. For example, the candidate parking areas may include Figure 1 The electronic device 100 also determines a target parking area from multiple candidate parking areas based on the parking intention, and determines a movement path for the vehicle 200 based on the positional relationship between the vehicle 200 and the target parking area. Based on the movement path, the electronic device 100 can control the vehicle 200 to travel to the target parking area. In this way, using the parking control method of the present application, multiple candidate parking areas can be determined based on parking requirement information and visual information around the vehicle, and a target parking area that meets the requirements can be screened from the multiple candidate parking areas. The vehicle's movement can then be controlled based on the positional relationship between the vehicle's real-time position and the target parking area, thereby improving the accuracy of vehicle parking control.

[0037] like Figure 2FIG2 is a flow chart of a parking control method provided in one embodiment of the present application. The order of the steps in the flow chart may be changed and some steps may be omitted according to different requirements. The parking control method provided in one embodiment of the present application includes the following steps.

[0038] S20 , obtaining parking intention and surrounding image information of the vehicle, wherein the parking intention is used to indicate a need to control the vehicle to park through structured information.

[0039] In one embodiment of the present application, to improve the efficiency of the vehicle parking process, a parking intention can be determined based on acquired parking request information, and the vehicle can be controlled to park in a parking area based on the parking intention. The parking request information can be a user's request expressed via natural language commands received by the vehicle's electronic equipment. For example, the parking request information can be "park in the space next to the red car on the left" or "park in the space directly in front."

[0040] In one embodiment of the present application, due to the presence of ambiguous semantics in parking requirement information, a parking intention can be determined based on the parking requirement to ensure that the parking requirement information used to control the vehicle's parking maneuver is understandable. The parking intention is used to instruct the vehicle to perform the parking maneuver. Specifically, the parking requirement information can be parsed using a model equipped with natural language processing capabilities to obtain the parking intention. Exemplarily, the model equipped with natural language processing capabilities can be a pre-trained generative model or a transfer learning model, which is not limited in this application. Specifically, based on the parking requirement information, the model equipped with natural language processing capabilities outputs the semantics of the parking requirement information and converts the semantics into structured information, thereby indicating the parking requirement to control the vehicle through this structured information. For example, if the parking requirement information is "park in the parking space to the left of the red car," the corresponding structured semantic information can be [reference object: red car, positional relationship: adjacent to the left, action: park]. When the parking requirement information is "park in the parking space to the right of the black SUV", the corresponding structured semantic information may be [reference object: black SUV, position relationship: adjacent to the right, action: parking].

[0041] In one embodiment of the present application, to mitigate the negative impact of ambiguous semantics in parking request information on the parking control process, the fuzzy semantics can be converted into quantified information based on the vehicle's position information, and an executable parking intention can be generated. For example, in the case of structured semantic information such as [reference object: red car, positional relationship: adjacent to the left, action: park], the fuzzy semantic information can be "adjacent to the left." Based on the spatial relationship mapping between the vehicle and the parking space, "adjacent to the left" can be converted into specific orientation parameters (e.g., within 1.5 meters laterally and within 0.3 meters longitudinally), resulting in a corresponding parking intention. For example, the parking intention can be "park within a lateral distance of 1.5 meters and a longitudinal distance of 0.3 meters to the left of the red car." This embodiment converts natural language information into clear quantitative features, which can then be matched with visual information captured by the vehicle's camera to determine the parking location.

[0042] S21 , performing parking dynamic matching based on the parking intention and surrounding image information of the vehicle to obtain a plurality of candidate parking areas.

[0043] In one embodiment of the present application, the surrounding image information may be an image captured by a camera device of the vehicle, which is used to indicate the environmental information surrounding the vehicle. After determining the parking intention, multiple candidate parking areas around the vehicle may be determined based on the quantitative information in the parking intention and the surrounding image information. For example, the candidate parking areas may include: Figure 1 Parking area 104 and parking area 105 are shown.

[0044] In one embodiment of the present application, surrounding image information can be determined based on video data captured by a vehicle's camera. Specifically, image data corresponding to multiple first timestamps can be acquired from the video data at a preset sampling frequency, and based on the image data corresponding to any first timestamp, surrounding image information of the vehicle corresponding to any one of the timestamps can be generated. For example, when the preset sampling frequency is 10 Hz, image data can be acquired from the real-time video data captured by the camera at 100 ms intervals, and all acquired image data can be spliced ​​together to obtain surrounding image information.

[0045] In one embodiment of the present application, the parking intention can characterize the relative position between the parking area indicated in the parking requirement and the vehicle based on the reference targets around the vehicle. Therefore, the reference targets in the surrounding image information can be determined based on the parking intention, and the corresponding candidate parking areas can be determined based on the reference targets. Each reference object determined based on the surrounding image information corresponds to a candidate parking area. The parking intention can also characterize the relative position between the parking area indicated in the parking requirement and the vehicle based on the environmental information around the vehicle. Therefore, the image features in the surrounding image information and the parking intention can be directly compared to determine the candidate parking areas in the surrounding image information. For a specific method of determining multiple candidate parking areas, please refer to Figure 3 Corresponding detailed description.

[0046] For example, the information of a candidate parking area determined based on reference targets around the vehicle may be [Vehicle ID: 001, Reference target attribute: Red, Vehicle coordinates: (X1, Y1), Adjacent slot: (X2, Y2)]. The vehicle ID indicates the vehicle's identification, the reference target attribute indicates information describing the attributes of the reference target determined based on surrounding image information (e.g., the reference target's color, type, or brand), the vehicle coordinates indicate the vehicle's position in the corresponding vehicle coordinate system, and the adjacent slot indicates the position of the candidate parking area corresponding to the reference target in the vehicle coordinate system.

[0047] At step S22, if the confidence level of any candidate parking area satisfies a preset confidence threshold, the candidate parking area is determined as the target parking area. If the confidence levels corresponding to the multiple candidate parking areas do not satisfy the preset confidence threshold, the target parking area is determined from the multiple candidate parking areas based on the surrounding image information, the parking intention, the vehicle's historical parking record, and preset weight parameters.

[0048] In one embodiment of the present application, to improve the accuracy of controlling a vehicle's parking in a parking area, a target parking area can be determined from candidate parking areas based on the parking intention and the image features of the surrounding image information, and the vehicle can be controlled to move to the target parking area. In this application, the first condition includes any one or more of the following: the similarity between the image features of the surrounding image information and the semantic information of the parking intention is greater than a preset first threshold; and the parking intention includes descriptive information of a reference target. The similarity can be the cosine similarity between the image features of the surrounding image information and the semantic information of the parking intention, or the inverse of the Euclidean distance between the image features of the surrounding image information and the semantic information of the parking intention. This application does not limit the specific method for determining the similarity. Among them, when the surrounding image information and the parking intention meet a preset first condition, indicating that the information in the parking intention can match the image features in the surrounding image information, the target parking area can be determined from the candidate parking areas based on the parking intention and the image features in the surrounding image information; when the surrounding image information and the parking intention do not meet the preset first condition, indicating that the degree of match between the information in the parking intention and the image features in the surrounding image information is low, the degree to which any candidate parking area meets the parking requirements can also be jointly determined based on the vehicle's historical parking records, and the influence of the surrounding image information, the parking intention, and the historical parking records on the determination of the target parking area can be adjusted based on preset weight parameters, thereby improving the accuracy of controlling vehicle parking.

[0049] In one embodiment of the present application, when the similarity is greater than a first threshold, indicating a high degree of similarity between the image features of the surrounding image information and the semantic information of the parking intention, the candidate parking area indicated by the parking intention has a high degree of match with the image features in the surrounding image information. Therefore, the target parking area can be directly determined from multiple candidate parking areas based on the surrounding image information and the parking intention. For example, the first threshold can be 0.8 or 0.9, and this application does not limit the specific value of the first threshold. When the parking intention includes descriptive information of a reference target, it indicates that the parking intention can indicate the specific location of the parking area based on the descriptive information of the reference target. Therefore, the target parking area can be determined from multiple candidate parking areas based on the surrounding image information and the parking intention. Because the surrounding image information can represent the visual information surrounding the vehicle from multiple dimensions, the higher the similarity between the image features in the surrounding image information and the descriptive information corresponding to the reference target in the parking intention, the higher the degree of match between the candidate parking area corresponding to the reference target and the parking requirement.

[0050] For example, a parking intention may be [reference object: red car, positional relationship: adjacent area 1.5 meters horizontally and 0.3 meters vertically on the left, action: parking]. Image features of the surrounding image information may be used to indicate information about candidate parking areas. For example, the candidate parking area information indicated by the image features may be [vehicle ID: 001, reference object attribute: red, vehicle coordinates: (X1, Y1), adjacent space: (X2, Y2)]. Because the parking intention includes information describing the reference object, the target parking area can be directly determined from multiple candidate parking areas based on the surrounding image information and the parking intention. If a preset first threshold is 0.99 and the similarity between the parking intention and the image features of the surrounding image information is 100%, indicating a high degree of match between the candidate parking areas corresponding to the image features in the surrounding image information and the reference object in the parking intention and the parking requirement, the target parking area can be directly determined from the multiple candidate parking areas based on the surrounding image information and the parking intention.

[0051] In one embodiment of the present application, if the surrounding image information and the parking intention information do not satisfy the first condition, indicating that the candidate parking area indicated by the parking intention has a low degree of match with the image features in the surrounding image information, or the parking intention does not include descriptive information corresponding to the reference target, and therefore the specific location of the candidate parking area corresponding to the reference target cannot be determined based on the parking intention, the degree to which any candidate parking area meets the parking requirement may be jointly determined based on the vehicle's historical parking records, and the influence of the surrounding image information, the parking intention, and the historical parking records on the determination of the target parking area may be adjusted based on preset weight parameters, thereby improving the accuracy of controlling the vehicle's parking.

[0052] Specifically, based on the surrounding image information, the parking intention information, the historical parking record of the vehicle and the preset weight parameters, the specific method of determining the target parking area from the multiple candidate parking areas can be found in Figure 4 、 Figure 5 and Figure 6 Corresponding detailed description.

[0053] In one embodiment of the present application, information about candidate parking areas and target parking areas can be simultaneously output to improve the accuracy of target parking area determination. Specifically, visual information about the candidate parking areas and target parking areas can be displayed on the vehicle's display device, with the target parking area labeled (for example, by setting the brightness of the target parking area image to be higher than that of the candidate parking areas). Simultaneously, the vehicle's audio device can output voice data to prompt the driver and passengers to select the target parking area or switch to another candidate parking area. For example, the voice data output by the vehicle's audio device could be, "Please confirm whether the highlighted target parking area is correct."

[0054] In one embodiment of the present application, after receiving feedback information from voice data, a parking operation can be executed based on the semantics of the feedback information. For example, if the semantics of the feedback information indicate that the target parking area is correct, the vehicle can be controlled to move to the target parking area based directly on the relative positional relationship between the vehicle and the target parking area. If the feedback information contains the message "Switch to the right parking space," the semantics of the feedback information indicate that the target parking area is inaccurate and has changed. In this case, the target parking area is re-determined from the candidate parking areas based on the semantics of the feedback information, and the vehicle is controlled to move to the target parking area based on the relative positional relationship between the vehicle and the target parking area.

[0055] S23: Perform parking planning based on the target parking area, and control the vehicle to travel to the target parking area according to the corresponding planning result.

[0056] In one embodiment of the present application, after a target parking area is determined among candidate parking areas, a vehicle movement path can be determined based on the positional relationship between the vehicle and the target parking area. The vehicle is then controlled to move to the target parking area based on the movement path. The movement path indicates the path the vehicle will take from its current position to the target parking area.

[0057] In one embodiment of the present application, in order to determine the positional relationship between the vehicle and the target parking area, a vehicle coordinate system can be constructed based on the position of the vehicle, and the first coordinate of the target parking area in the vehicle coordinate system can be determined based on the internal parameters and external parameters of the vehicle's camera device and the image information corresponding to the target parking area in the surrounding image information. The positional relationship is determined based on the first coordinate and the second coordinate of the vehicle in the vehicle coordinate system. Since the vehicle must comply with the vehicle's own motion constraints during movement, the vehicle's moving path can be determined based on the vehicle's motion constraints and the positional relationship. Among them, the vehicle's motion constraints include the vehicle's turning radius, the vehicle's maximum angular velocity, the vehicle's maximum speed in the area where it is located, etc. This application does not limit the specific content of the motion constraints.

[0058] like Figure 7 As shown, the area where vehicle 800 is located includes target parking area 801, candidate parking area 802, reference target 803, reference target 804, and reference target 805. When determining the movement path of vehicle 800, movement path 806 can be determined based on the motion constraints of vehicle 800 and the relative positional relationship between vehicle 800 and target parking area 801, ensuring that vehicle 800 complies with the motion constraints of vehicle 800 when moving along movement path 806 to target parking area 801.

[0059] In one embodiment of the present application, the method for determining a vehicle's moving path based on a positional relationship between a vehicle and a target parking area specifically includes: determining a first coordinate of the target parking area in a vehicle coordinate system of the vehicle based on image information corresponding to the target parking area in the surrounding image information; determining the positional relationship between the vehicle and the target parking area based on the first coordinate and a second coordinate of the vehicle in the vehicle coordinate system; and determining the vehicle's moving path based on the positional relationship and motion constraints of the vehicle.

[0060] In one embodiment of the present application, the process of determining the target parking area based on surrounding image information involves analyzing the vehicle's surrounding environment based solely on two-dimensional information. To improve the accuracy of determining the vehicle's path to the target parking area, the first coordinates of the target parking area in the vehicle coordinate system are also determined based on the corresponding image information of the target parking area in the surrounding image information. This converts the two-dimensional information of the target parking area into three-dimensional information, and the vehicle's path can then be determined based on this three-dimensional information.

[0061] Specifically, the coordinates of any pixel in the surrounding image information belonging to the target parking area can be transformed based on the intrinsic parameters of the vehicle's camera device to obtain the corresponding position of any pixel in the vehicle coordinate system. For example, if the coordinates of a pixel in the surrounding image information belonging to the target parking area are (u, v) in the surrounding image information, and the intrinsic parameter of the vehicle's camera device is K, the method for determining the corresponding three-dimensional coordinates of the pixel in the vehicle coordinate system satisfies the following relationship: ; Wherein, u represents the horizontal coordinate of a pixel point belonging to the target parking area in the surrounding image information, v represents the vertical coordinate of the pixel point in the surrounding image information; X, Y, and Z represent the three-dimensional coordinates corresponding to the pixel point in the vehicle coordinate system.

[0062] In one embodiment of the present application, the second coordinate indicates the position of the vehicle in a vehicle coordinate system. When the vehicle coordinate system is a coordinate system with the vehicle as its origin, the second coordinate can be determined as (0, 0, 0). The positional relationship between the vehicle and the target parking area can be determined based on the difference between the first coordinate and the second coordinate.

[0063] In one embodiment of the present application, since the vehicle must comply with its own motion constraints during movement, the vehicle's movement path can be determined based on the vehicle's motion constraints and positional relationships. The vehicle's motion constraints include, for example, the vehicle's turning radius, maximum angular velocity, and maximum speed in the area it is located. The present application does not limit the specific content of the motion constraints.

[0064] In one embodiment of the present application, a vehicle's movement path can be determined based on a fast exploration random tree algorithm based on positional relationships and the vehicle's motion constraints. Specifically, a first sampling point is randomly determined in the vehicle coordinate system; the point closest to the first sampling point is determined as a second sampling point; a third sampling point is determined based on the direction of the first sampling point toward the second sampling point and the vehicle's motion constraints; and if there are no obstacles between the second and third sampling points, other sampling points are continuously expanded in the vehicle coordinate system based on the position of the third sampling point until the distance between the sampling point and the target parking area is less than a preset distance threshold. The sampling point and the target parking area are then connected to obtain the vehicle's movement path. In this way, the feasible space in the area surrounding the vehicle can be explored through random sampling, and safe nodes can be generated in combination with the vehicle's motion constraints. The starting point and the end point are gradually connected, and the movement path is optimized to meet actual driving needs.

[0065] In one embodiment of the present application, in the process of controlling the vehicle to move to the target parking area according to the moving path, since the position of the vehicle changes in real time, the relative position relationship between the vehicle and the target parking area also changes in real time. Since other targets around the vehicle (for example, pedestrians, other vehicles, etc.) may be located in the moving path, in order to avoid safety accidents caused by collisions between the vehicle and other targets, the moving path of the vehicle to the target parking area can be dynamically updated based on the real-time position of the vehicle to improve the safety of the parking process. In this way, based on the dynamic decision-making mechanism, the moving path of the vehicle can be dynamically adjusted according to environmental changes to ensure the success rate of the parking operation. Specifically, in the process of controlling the movement of the vehicle according to the moving path, the method of dynamically updating the moving path according to the real-time position of the vehicle can be seen. Figure 8 Corresponding detailed description.

[0066] like Figure 9 As shown, during the process of vehicle 800 moving to target parking area 801 along moving path 806, dynamic target 807 is moving simultaneously, so dynamic target 807 may appear on moving path 806. Therefore, the moving path of vehicle 800 needs to be updated to ensure that the updated moving path can avoid the dynamic target. Specifically, the moving path can be updated based on the real-time position of vehicle 800, the real-time position of dynamic target 807, the relative positional relationship between vehicle 800 and target parking area 801, and the motion constraints of vehicle 800 to obtain an updated moving path 808. In this way, driving safety can be guaranteed during the process of vehicle 800 moving to target parking area 801 along updated moving path 808, and the corresponding motion constraints can be ensured during the movement of vehicle 800.

[0067] In one embodiment of the present application, after a vehicle parks in a target parking area, corresponding natural language description information can be generated based on the vehicle's movement time and parking result. For example, if the parking requirement information is "parking space to the left of the red car" and the vehicle's movement time is 32 seconds, if the vehicle successfully parks in the target parking area, the natural language description information corresponding to the parking result may be "Successfully parked to the left of the red car, taking 32 seconds." If the parking requirement information is "parking space to the right of the black SUV" and the vehicle's movement time is 15 seconds, if the vehicle collides or stops at any point along the movement path, the natural language description information corresponding to the parking result may be "Parking failed, failed to park to the right of the black SUV, taking 15 seconds."

[0068] In one embodiment of the present application, to ensure that the driver or passengers in the vehicle can promptly access the parking result, natural language description information corresponding to the parking result may be output via the vehicle's output device. The vehicle's output device may be an audio output device or a visual display device (e.g., a display screen), which is not limited in this application.

[0069] In one embodiment of the present application, to improve the accuracy of vehicle parking control, the vehicle's historical parking record can also be updated based on the natural language description information corresponding to the parking result. Specifically, the parking intention of the current parking attempt, the environmental information indicated by the surrounding image information, and the natural language description information corresponding to the parking result can be added to the vehicle's historical parking record to obtain an updated historical parking record, thereby improving the accuracy of subsequent determination of the target parking area and control of vehicle movement to the target parking area.

[0070] As can be seen from the above technical solutions, the embodiments of the present application analyze the semantics of parking request information to determine the parking intention that controls the vehicle to execute a parking maneuver. The system then characterizes the visual characteristics of the vehicle's surrounding environment based on the vehicle's surrounding image information. Based on the parking intention and visual characteristics, the system determines multiple candidate parking areas around the vehicle, thereby narrowing the scope of the selected parking areas and improving parking efficiency. Furthermore, the system also quantitatively scores the candidate parking areas based on the vehicle's surrounding environmental information and historical parking records, and selects target parking areas based on the scores. This multi-layered screening mechanism determines the parking area for the vehicle, and controls the vehicle's movement to the target parking area based on the relative positional relationship between the target parking area and the vehicle. This improves the accuracy of the parking area determination and enhances the safety of the parking process.

[0071] like Figure 3The figure shows a flowchart of a method for determining multiple candidate parking areas according to an embodiment of the present application. The order of the steps in the flowchart may be changed, and some steps may be omitted, depending on different needs. The flowchart of the method for determining multiple candidate parking areas according to an embodiment of the present application includes the following steps.

[0072] S30 , performing image segmentation on the surrounding image information based on the parking intention, and determining a reference target in the surrounding image information.

[0073] In one embodiment of the present application, since the parking intention can indicate the relative positional relationship between the parking area that meets the parking requirements and the vehicle, the surrounding image information can be segmented based on the parking intention to determine the pixel points in the surrounding image information that belong to the reference target, and then determine the reference target in the surrounding image information.

[0074] In one embodiment of the present application, image segmentation can be performed on the surrounding image information based on a model capable of image segmentation to obtain an image region in the surrounding image information that indicates a reference target. Exemplarily, the model capable of image segmentation can be a recurrent convolutional network model or a U-Net model, which is not limited in this application.

[0075] like Figure 10 , which is a schematic diagram of vehicle surrounding image information 1200. After performing image segmentation on surrounding image information 1200 using a model with image segmentation capabilities, it is determined that the regions belonging to the reference target include region 1201 and region 1202. Furthermore, the image information in regions 1201 and 1202 can be used to indicate the reference target surrounding the vehicle. Region 1203 can be image information corresponding to a candidate parking area in surrounding image information 1200.

[0076] S31 : Constructing first prompt data based on a preset first prompt word and image information corresponding to the reference target in the surrounding image information.

[0077] In one embodiment of the present application, to improve the accuracy of identifying candidate parking areas from surrounding image information, first prompt data may be constructed based on a first prompt word and image information corresponding to a reference target in the surrounding image information. The first prompt word is used to interpret and constrain the image information representing the reference target in the surrounding image information.

[0078] Exemplarily, the first prompt word may be in the form of: "Background enhancement: Please play the role of an expert who is proficient in the field of image recognition; Task prompt: Please predict the parking area adjacent to the reference target based on the image information of the reference target; Information prompt:; Output prompt: Please output the attribute information of the reference target and the location information of the corresponding candidate parking area."

[0079] For example, when the vertices of the image information corresponding to the reference target in the surrounding image information include coordinates (10,10), (10,50), (50,10), and (50,50), the form of the first prompt data can be determined as: "Background enhancement: Please play the role of an expert who is proficient in the field of image recognition; Task prompt: Please predict the parking area adjacent to the reference target based on the image information of the reference target; Information prompt: The image information indicating the reference target in the image is within the range of pixel points with coordinates (10,10), (10,50), (50,10), and (50,50); Output prompt: Please output the attribute information of the reference target and the location information of the corresponding candidate parking area."

[0080] S32: Determine a plurality of candidate parking areas based on the first prompt data using a pre-trained generative model.

[0081] In one embodiment of the present application, the pre-trained generative model can be any one of a transfer learning model, a GPT model, and an adversarial generative model, and this application does not limit this. The first prompt data can be input into the pre-trained generative model to obtain an output result of the generative model. The output result of the generative model includes location information of multiple candidate parking areas. Exemplarily, the output result of the generative model can be in the form of [Vehicle ID: 001, Reference Target Attribute: Red, Vehicle Coordinates: (X1, Y1), Adjacent Space: (X2, Y2)]. The vehicle ID is used to indicate the vehicle's identification, the reference target attribute is used to indicate information describing the attributes of the reference target determined based on surrounding image information (e.g., the reference target's color, reference target type, reference target brand, etc.), the vehicle coordinates are used to indicate the vehicle's position in the corresponding vehicle coordinate system, and the adjacent space is used to indicate the position of the candidate parking area corresponding to the reference target in the vehicle coordinate system.

[0082] like Figure 4 1 is a flowchart of a method for determining a target parking area according to an embodiment of the present application. The order of the steps in the flowchart may be changed, and some steps may be omitted, depending on different needs. The flowchart of the method for determining a target parking area according to an embodiment of the present application includes the following steps.

[0083] S40 : Determine a score of any one of the plurality of candidate parking areas based on the surrounding image information, the parking intention information, the historical parking record of the vehicle, and a preset weight parameter.

[0084] In one embodiment of the present application, to improve the accuracy of determining a target parking area, a target parking area may be determined from multiple candidate parking areas based on a hierarchical matching strategy. The hierarchical matching strategy includes first determining the similarity between semantic features of the parking intention and image information of candidate parking areas in surrounding image information. Once the similarity reaches a first threshold, the target parking area is directly determined from the candidate parking areas based on the surrounding image information and the parking intention, and the vehicle is controlled to move from its current location to the target parking area. If the degree of match between the candidate parking area and the image features in the surrounding image information is low, or if the parking intention does not include descriptive information corresponding to the reference target, the specific location of the candidate parking area corresponding to the reference target cannot be determined based on the parking intention. The degree to which any candidate parking area meets the parking requirement may be determined based on the vehicle's historical parking history. The influence of surrounding image information, parking intention, and historical parking history on the determination of the target parking area is adjusted based on preset weighting parameters to determine a score for each parking area, thereby improving the accuracy of vehicle parking control based on the quantified data. In this way, when the vehicle is in a complex scene (for example, the target parking area is partially obscured, the driver or vehicle user has a deviation in the description of the attributes of the reference target, etc.), the success rate of parking position matching can be improved while ensuring the accuracy of controlling vehicle parking.

[0085] In one embodiment of the present application, when determining a score for any candidate parking area, the similarity between the image features of the candidate parking area in the surrounding image information and the semantics of the parking intention is evaluated to obtain a corresponding first score. The first score represents the degree of consistency between the image features of the candidate parking area in the surrounding image information and the parking intention. A second score is determined based on the semantic features of the parking intention and the environmental information of the vehicle's area. The second score evaluates the degree of consistency between the parking intention and the vehicle's current environment. A third score is also determined based on the vehicle's historical parking records and environmental information. The third score evaluates the degree of consistency between the environmental information during the vehicle's current parking maneuver and the relevant information recorded in the historical parking records. The vehicle's historical parking records record information related to the driver's control of the vehicle during parking maneuvers. For example, the historical parking records include information such as frequently selected parking space types (e.g., preference for spacious spaces or spaces near exits), parking time distribution, and special requirement tags (e.g., charging requirements or sun protection requirements). User behavior patterns can be analyzed using a collaborative filtering algorithm to generate a personalized preference vector, which is then mapped into a user preference score.

[0086] In one embodiment of the present application, the influence of surrounding image information, parking intent, and historical parking records on the determination of the target parking area can be adjusted based on preset weighting parameters to determine a score for any parking area, thereby improving the accuracy of vehicle parking control based on quantified data. For example, a first weight corresponding to a first score can be determined to be 60%, indicating that the first score has a 60% influence on the determination of the target parking area; a second weight corresponding to a second score can be determined to be 30%, indicating that the second score has a 60% influence on the determination of the target parking area; and a third weight corresponding to a third score can be determined to be 10%, indicating that the third score has a 10% influence on the determination of the target parking area.

[0087] In one embodiment of the present application, the method for determining the score of any candidate parking area is as follows: Figure 5 For detailed description of the corresponding weight parameters, please refer to Figure 6 Corresponding detailed description.

[0088] S41 : Determine the target parking area from the plurality of candidate parking areas based on the scores.

[0089] In one embodiment of the present application, the candidate parking area corresponding to the highest score may be determined as the target parking area.

[0090] like Figure 5 The figure shows a flowchart of a method for determining the scoring of candidate parking areas according to an embodiment of the present application. The order of the steps in the flowchart may be changed, and some steps may be omitted, depending on different needs. The flowchart of the method for determining the scoring of candidate parking areas according to an embodiment of the present application includes the following steps.

[0091] S50 : Determine area information of any one candidate parking area in the surrounding image information and environment information of the vehicle based on the surrounding image information.

[0092] In one embodiment of the present application, to determine the degree of match between the image features corresponding to the candidate parking areas in the surrounding image information and the parking intention, the region information of any candidate parking area in the surrounding image information may be determined based on the surrounding image information. The region information may be a collection of pixels corresponding to the candidate parking area in the surrounding image information.

[0093] In one embodiment of the present application, in order to evaluate the degree of conformity between the vehicle's current environment and the semantic information of the parking intention, the vehicle's environmental information can be determined based on the surrounding image information. The environmental information is used to indicate the environmental characteristics of the area where the vehicle is located. For example, the environmental information can be used to indicate the number of other vehicles in the area where the vehicle is located, and can also be used to indicate the color, brand and other attribute information of other vehicles in the area where the vehicle is located, and can also be used to indicate the location information, morphological information, etc. of other physical objects in the area where the vehicle is located. For example, in the case where the area where the vehicle is located is a parking lot, the environmental information can be used to indicate the location of the entrance and exit of the parking lot, the environmental information can also be used to indicate the location of obstacles in the parking lot, and the environmental information can also be used to indicate the speed limit information indicated by the speed limit signs in the parking lot.

[0094] S51 : Determine a first score of any one of the candidate parking areas based on the area information, the semantic information of the parking intention, and a first parameter among the weight parameters.

[0095] In one embodiment of the present application, the region information may be used to indicate image features of a candidate parking area in surrounding image information. The parking intention may represent the relative position between the parking area indicated in the parking requirement and the vehicle based on reference targets around the vehicle. A first similarity between the region information and the semantic information of the parking intention may be determined, and the degree of conformity between the parking intention and the region information may be evaluated based on the first similarity.

[0096] In one embodiment of the present application, a first score for a candidate parking area can be determined based on the product of the first similarity and the first parameter. A higher first score indicates a higher degree of consistency between the parking intention and the area information; a higher first parameter indicates a greater degree of influence of the area information of the candidate parking area in the surrounding image information on the determination of the target parking area. Therefore, a higher first score for any candidate parking area indicates a higher degree of compliance with the parking requirement, and thus a higher probability of determining the candidate parking area as the target parking area.

[0097] S52 : Determine a second score of any one of the candidate parking areas based on the semantic information of the parking intention, the environmental information, and a second parameter in the weight parameters.

[0098] In one embodiment of the present application, environmental information is used to indicate environmental characteristics of the area in which the vehicle is located. For example, when the vehicle's environment is a parking lot, the environmental information includes geometric information indicating the coordinates of the parking space within the parking lot. The environmental information also includes attribute information indicating the parking space type and height limit within the parking lot. For example, the environmental information can be used to indicate the parking space type and height limit within the parking lot. The vehicle's environmental information can be graph-structured data. The graph-structured data can be used to indicate text annotations within the environmental information (e.g., charging station locations), behavioral rules within the environmental information (e.g., no-parking zones), and dynamic events within the environmental information (e.g., temporary construction). For example, nodes within the graph-structured data can be entities within the parking lot (e.g., parking spaces, facilities, vehicle types) and their corresponding attributes (e.g., parking space nodes include width, height, and type). Edges within the graph-structured data can be used to indicate relationships between entities within the parking lot (e.g., parking space A is adjacent to parking space B, and the charging station area is on the B1 floor).

[0099] In one embodiment of the present application, a second similarity between the semantic information of the parking intention and the environmental information can be determined, and the degree of conformity between the parking intention and the environmental information can be evaluated based on the second similarity. A second score for the candidate parking area can be determined based on the product of the second similarity and a second parameter. A higher second score indicates a higher degree of conformity between the parking intention and the environmental information; a higher second parameter indicates a greater degree of influence of the environmental information in the surrounding image information and the semantic information of the parking intention on the determination of the target parking area. Therefore, a higher second score for any candidate parking area indicates a higher degree of conformity between the candidate parking area and the parking requirement, and a higher probability of determining the candidate parking area as the target parking area.

[0100] S53 : Determine a third score of the any one candidate parking area based on the historical parking record, the environmental information, and a third parameter among the weight parameters.

[0101] In one embodiment of the present application, historical parking records are used to indicate relevant information about a vehicle's parking operations at any point in time, and can indicate the user's behavior in selecting a parking area. For example, historical parking records include information such as frequently selected parking space types (e.g., preference for spacious spaces or spaces near exits), parking time distribution, and special requirements (e.g., charging requirements or sun protection requirements).

[0102] In one embodiment of the present application, a third similarity between historical parking records and environmental information can be determined, and the degree of conformity between the historical parking records and environmental information can be evaluated based on the third similarity. A third score for the candidate parking area can be determined based on the product of the third similarity and a third parameter. A higher third score indicates a higher degree of conformity between the historical parking records and environmental information; a higher third parameter indicates a greater degree of influence of the historical parking records and environmental information on the determination of the target parking area. Therefore, a higher third score for any candidate parking area indicates a higher degree of conformity between the candidate parking record and the environmental information, and a higher probability of determining the candidate parking area as the target parking area.

[0103] S54: Determine a score for any one of the candidate parking areas based on the first score, the second score, and the third score.

[0104] In one embodiment of the present application, the sum of the first, second, and third scores of any candidate parking area can be determined as the score of the candidate parking area. For example, if the cosine similarity between the area information corresponding to any candidate parking area and the semantic information of the parking intention is 90%, and the first parameter is 90%, then the first score of the candidate parking area is 0.9*0.9=0.81. If the cosine similarity between the environmental information corresponding to any candidate parking area and the semantic information of the parking intention is 80%, and the second parameter is 5%, then the second score of the candidate parking area is 0.8*0.05=0.4. If the cosine similarity between the environmental information corresponding to any candidate parking area and the historical parking record is 90%, and the third parameter is 5%, then the third score of the candidate parking area is 0.9*0.05=0.45. Therefore, the score of the candidate parking area is 0.81+0.4+0.45=1.66.

[0105] In one embodiment of the present application, the scores of all candidate parking areas may be traversed, and the candidate parking area corresponding to the highest score may be determined as the target parking area.

[0106] like Figure 6 , is a flowchart of a method for determining weight parameters provided by an embodiment of the present application. Depending on different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted. The flowchart of the method for determining weight parameters provided by an embodiment of the present application includes the following steps.

[0107] S60 : Determine a first parameter of the weight parameters based on the similarity between the semantic information of the parking intention and the area information of any one of the candidate parking areas.

[0108] In one embodiment of the present application, the first parameter in the weight parameters may be used to indicate the degree of influence of the area information corresponding to any candidate parking area in the surrounding image information on the determination of the target parking area. A higher first parameter indicates a greater degree of influence of the area information corresponding to any candidate parking area in the surrounding image information on the determination of the target parking area.

[0109] Specifically, the first parameter in the weighting parameters can be determined based on the similarity between the semantic information of the parking intention and the regional information of any candidate parking area. The higher the similarity between the semantic information of the parking intention and the regional information of any candidate parking area, the more consistent the regional information corresponding to the candidate parking area in the surrounding image information is with the parking intention. The similarity between the semantic information of the parking intention and the regional information of any candidate parking area can be cosine similarity, Pearson correlation coefficient, or log-likelihood similarity. This application does not limit the specific method for calculating the similarity.

[0110] S61 : Determine a second parameter in the weight parameters based on the similarity between the semantic information of the parking intention and the environmental information.

[0111] In one embodiment of the present application, the second parameter in the weight parameter may be used to indicate the degree to which the environmental information corresponding to the vehicle's location in the surrounding image information influences the determination of the target parking area. A higher second parameter indicates a higher degree of influence of the environmental information corresponding to the vehicle's location in the surrounding image information on the determination of the target parking area.

[0112] Specifically, the second parameter in the weighting parameters can be determined based on the similarity between the semantic information of the parking intention and the environmental information corresponding to the vehicle's area in the surrounding image information. The higher the similarity between the semantic information of the parking intention and the environmental information corresponding to the vehicle's area in the surrounding image information, the more consistent the environmental information corresponding to the vehicle's area in the surrounding image information is with the parking intention. The method for determining the similarity between the semantic information of the parking intention and the environmental information is the same as the method for determining the similarity between the semantic information of the parking intention and the area information of any candidate parking area, and will not be further described here.

[0113] S62: Determine a third parameter in the weight parameters based on the similarity between the historical parking record and the environmental information.

[0114] In one embodiment of the present application, the third parameter in the weight parameter may be used to indicate the degree of influence of the vehicle's historical parking record on the determination of the target parking area. A higher third parameter indicates a higher degree of influence of the vehicle's historical parking record on the determination of the target parking area.

[0115] Specifically, the third parameter in the weighting parameters can be determined based on the similarity between historical parking records and environmental information. A higher similarity between historical parking records and environmental information indicates a higher degree of similarity between the vehicle's historical parking records and the environmental information of the vehicle's current location, and thus a greater degree of influence of the vehicle's historical parking records on the determination of the target parking area. The method for determining the similarity between historical parking records and environmental information is similar to the method for determining the similarity between the semantic information of the parking intention and the area information of any candidate parking area, and will not be further elaborated here.

[0116] like Figure 8 , which is a flow chart of a method for updating a vehicle's movement path provided in one embodiment of the present application. Depending on different requirements, the order of the steps in the flow chart may be changed, and some steps may be omitted. The flow chart of the method for updating a vehicle's movement path provided in an embodiment of the present application includes the following steps.

[0117] S80, when controlling the vehicle to move according to the moving path, if the vehicle satisfies a preset second condition, determining a positional relationship between the vehicle and the target parking area based on real-time image information of the vehicle acquired in real time.

[0118] In one embodiment of the present application, when a vehicle is controlled to move to a target parking area according to a movement path, the relative position between the vehicle and the target parking area changes in real time due to the real-time changes in the vehicle's position. Since other objects (e.g., pedestrians, other vehicles, etc.) surrounding the vehicle may be located in the movement path, to avoid collisions with other objects and potential accidents, the positional relationship between the vehicle and the target parking area can be determined based on real-time image information of the vehicle. Based on this positional relationship, the movement path of the vehicle to the target parking area is dynamically updated to enhance parking safety. This dynamic decision-making mechanism allows the vehicle's movement path to be dynamically adjusted based on environmental changes, ensuring a high success rate for parking maneuvers.

[0119] Specifically, when the vehicle meets the preset conditions, indicating that the vehicle's position has changed significantly, the moving path can be dynamically updated based on the vehicle's real-time position. The preset conditions include any one of the following: the vehicle's speed is greater than or equal to a preset speed threshold; the distance between the vehicle and any target in the real-time image information is less than or equal to a preset distance threshold. When the vehicle's speed is greater than or equal to the preset speed threshold, it indicates that the vehicle is moving at a higher speed, and the vehicle's position may change significantly in a short period of time. Therefore, the moving path can be updated based on the vehicle's real-time position; when the distance between the vehicle and any target in the real-time image information is less than or equal to the preset distance threshold, it indicates that the vehicle has a high probability of collision during movement. In order to improve driving safety, the moving path can be updated based on the vehicle's real-time position, and the vehicle can be controlled to move to the target parking area based on the updated moving path to ensure that the vehicle can avoid obstacles during movement.

[0120] In one embodiment of the present application, the third coordinate of the vehicle's current position in the vehicle coordinate system can be determined based on the vehicle's movement time, speed, and movement path, and the positional relationship between the vehicle and the target parking area can be determined based on the coordinate difference between the first coordinate and the third coordinate of the target parking area.

[0121] S81: Update the moving path based on the position relationship to obtain an updated moving path.

[0122] In one embodiment of the present application, real-time image information can be segmented to identify any target in the real-time image information. Based on the coordinates of the pixel points corresponding to the target in the real-time image information, the fourth coordinate of the target in the vehicle coordinate system can be determined. When updating the movement path based on the positional relationship, the updated movement path can be adjusted in real time based on the fourth coordinate to ensure that the updated movement path can avoid any target in the real-time image information, thereby avoiding traffic accidents.

[0123] S82: Control the vehicle to travel to the target parking area according to the updated movement path.

[0124] In one embodiment of the present application, after determining the updated movement path, a control instruction can be generated based on the updated movement path, and the operation of vehicle components (e.g., power source, steering components, etc.) can be controlled based on the control instruction, thereby controlling the vehicle to travel to the target parking area.

[0125] In one embodiment of the present application, the method further comprises the step of determining parking intention. Figure 111 is a flowchart of a method for determining parking intention according to an embodiment of the present application. The order of the steps in the flowchart may be changed, and some steps may be omitted, depending on different needs. The flowchart of the method for determining parking intention according to an embodiment of the present application includes the following steps.

[0126] S110 , parsing the semantics of the pre-acquired parking requirement to obtain corresponding structured semantic information.

[0127] In one embodiment of the present application, to improve the efficiency of the vehicle parking process, a parking intention can be determined based on acquired parking request information, and the vehicle can be controlled to park in a parking area based on the parking intention. The parking request information can be a user's request expressed via natural language commands received by the vehicle's electronic equipment. For example, the parking request information can be "park in the space next to the red car on the left" or "park in the space directly in front."

[0128] To mitigate the negative impact of ambiguous semantics in parking request information on the parking control process, the semantics of the parking request can be parsed to obtain corresponding structured semantic information. For example, the structured semantic information can be [reference object: red car, position relationship: adjacent to the left, action: parking].

[0129] S111 , determining, based on sensor data of the vehicle, position parameters corresponding to the orientation information in the structured semantic information; the position parameters are used to characterize a relative positional relationship between the parking area indicated by the parking requirement and the vehicle.

[0130] In one embodiment of the present application, due to the presence of semantically ambiguous information in the parking requirement information, in order to ensure that the parking requirement information controlling the vehicle to perform the parking operation can be understood, a parking intention can also be determined based on the parking requirement, wherein the parking intention is used to indicate the control of the parking operation on the vehicle. Based on the sensor data of the vehicle, the position parameters corresponding to the orientation information in the structured semantic information are determined; the position parameters are used to characterize the relative positional relationship between the parking area indicated by the parking requirement and the vehicle. For example, if the parking requirement information is "parking in the parking space to the left of the red car," the corresponding structured semantic information can be [reference object: red car, position relationship: adjacent to the left, action: parking]. If the parking requirement information is "parking in the parking space to the right of the black SUV," the corresponding structured semantic information can be [reference object: black SUV, position relationship: adjacent to the right, action: parking].

[0131] S112: Determine the parking intention according to the structured semantic information and the position parameter.

[0132] In one embodiment of the present application, parking demand information can be parsed based on a model equipped with natural language processing capabilities to obtain parking intentions. Exemplarily, the model equipped with natural language processing capabilities can be a pre-trained generative model or a transfer learning model, which is not limited in this application. Specifically, based on the parking demand information, the semantics of the parking demand information are output by the model equipped with natural language processing capabilities to obtain corresponding structured semantic information, and the parking intention is determined based on the structured semantics. To mitigate the negative impact of ambiguous semantics in the parking demand information on the parking control process, the fuzzy semantics can be converted into quantified information based on the vehicle's position information, and an executable parking intention can be generated. Exemplarily, if the structured semantic information is [reference object: red car, position relationship: left adjacent, action: park], the information with ambiguous semantics can be "left adjacent." Based on the spatial relationship mapping between the vehicle and the parking space, "left adjacent" can be converted into specific position parameters (e.g., lateral distance within 1.5 meters, longitudinal error within 0.3 meters) to obtain the corresponding parking intention. For example, a parking intention could be "parking within a 1.5-meter horizontal and 0.3-meter vertical area on the left side of the red car." The above embodiment can convert natural language information into clear quantitative features, which can then be matched with visual information captured by the vehicle's camera to determine the parking location.

[0133] See Figure 12 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Electronic device 100 includes memory 12 and processor 13. Memory 12 is used to store computer-readable instructions, and processor 13 is used to execute the computer-readable instructions stored in the memory to implement a parking control method described in any of the above embodiments.

[0134] In an embodiment of the present application, the electronic device 100 further includes a bus, and a computer program stored in the memory 12 and executable on the processor 13 , such as a parking control program.

[0135] Figure 12 Only the electronic device 100 having the memory 12 and the processor 13 is shown. It can be understood by those skilled in the art that Figure 12 The structure shown does not constitute a limitation on the electronic device 100 , and the electronic device 100 may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0136] Combine Figure 2The memory 12 in the electronic device 100 stores a plurality of computer-readable instructions for implementing the parking control method, and the processor 13 can execute the plurality of instructions to implement the following: obtaining a parking intention and surrounding image information of the vehicle, wherein the parking intention is used to indicate a need to control the vehicle to park via structured information; performing dynamic parking matching based on the parking intention and the surrounding image information of the vehicle to obtain a plurality of candidate parking areas; if a confidence level of any candidate parking area meets a preset confidence threshold, determining the candidate parking area as a target parking area; if the confidence levels corresponding to the plurality of candidate parking areas do not meet the preset confidence threshold, determining a target parking area from the plurality of candidate parking areas based on the surrounding image information, the parking intention, the vehicle's historical parking record, and a preset weight parameter; performing parking planning based on the positional relationship of the target parking area, and controlling the vehicle to travel to the target parking area according to the corresponding planning result.

[0137] Specifically, the specific implementation method of the processor 13 for the above instructions can refer to Figure 2 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0138] Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 100 and does not constitute a limitation on the electronic device 100. The electronic device 100 may have a bus structure or a star structure. The electronic device 100 may also include more or less other hardware or software than shown in the figure, or a different arrangement of components. For example, the electronic device 100 may also include input and output devices, network access devices, etc.

[0139] It should be noted that the electronic device 100 is only an example. Other existing or future electronic products that are suitable for this application should also be included in the scope of protection of this application and incorporated herein by reference.

[0140] The memory 12 includes at least one type of readable storage medium, which can be either non-volatile or volatile. The readable storage medium includes flash memory, a removable hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 12 can be an internal storage unit of the electronic device 100, such as a removable hard disk of the electronic device 100. In other embodiments, the memory 12 can also be an external storage device of the electronic device 100, such as a plug-in removable hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. The memory 12 can be used not only to store application software installed in the electronic device 100 and various types of data, such as the code of a parking control program, but can also be used to temporarily store data that has been output or is about to be output.

[0141] In some embodiments, the processor 13 may be comprised of an integrated circuit, such as a single packaged integrated circuit or multiple packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor 13 is the control core (control unit) of the electronic device 100, connecting the various components of the electronic device 100 via various interfaces and circuits. It executes programs or modules stored in the memory 12 (e.g., a parking control program) and accesses data stored in the memory 12 to perform various functions and process data.

[0142] The processor 13 executes the operating system of the electronic device 100 and various installed applications. The processor 13 executes the applications to implement the steps in the above-mentioned parking control method embodiments, such as Figure 2 Steps shown.

[0143] For example, the computer program may be divided into one or more modules / units, which are stored in the memory 12 and executed by the processor 13 to complete the present application. The one or more modules / units may be a series of computer-readable instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device 100.

[0144] The above-mentioned integrated unit implemented in the form of a software functional module can be stored in a computer-readable storage medium. The above-mentioned software functional module stored in a storage medium includes a number of instructions for causing a computer device (which can be a personal computer, computer equipment, or network device, etc.) or a processor to execute a portion of a parking control method described in various embodiments of this application.

[0145] If the modules / units integrated in the electronic device 100 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment methods, and can also instruct the relevant hardware devices to complete them through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the steps of each of the above-mentioned method embodiments.

[0146] The computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory, and other memory.

[0147] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.

[0148] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 12 The diagram is represented by only one arrow, but it does not mean that there is only one bus or one type of bus. The bus is configured to implement connection and communication between the memory 12 and at least one processor 13, etc.

[0149] An embodiment of the present application further provides a computer-readable storage medium (not shown), in which computer-readable instructions are stored. The computer-readable instructions are executed by a processor in an electronic device to implement the parking control method described in any of the above embodiments.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0151] In addition, the functional modules in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0152] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices listed in the specification may also be implemented by a single unit or device through software or hardware. Terms such as first and second are used to indicate names and do not imply any particular order.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A parking control method, characterized in that: The method comprises: Acquiring a parking intention and surrounding image information of the vehicle, wherein the parking intention is used to indicate a need to control the vehicle to park through structured information; Performing dynamic parking matching based on the parking intention and surrounding image information of the vehicle to obtain multiple candidate parking areas; If the confidence level of any candidate parking area meets a preset confidence threshold, the candidate parking area is determined as the target parking area; if the confidence levels corresponding to the multiple candidate parking areas do not meet the preset confidence threshold, the target parking area is determined from the multiple candidate parking areas based on the surrounding image information, the parking intention, the vehicle's historical parking record, and a preset weight parameter; Parking planning is performed based on the target parking area, and the vehicle is controlled to travel to the target parking area according to the corresponding planning result.

2. The parking control method according to claim 1, characterized in that: The determining of a target parking area from a plurality of candidate parking areas based on the surrounding image information, the parking intention information, the historical parking record of the vehicle, and a preset weight parameter includes: determining a score for any one of the plurality of candidate parking areas based on the surrounding image information, the parking intention information, the historical parking record of the vehicle, and a preset weight parameter; The target parking area is determined from the plurality of candidate parking areas based on the scores.

3. The parking control method according to claim 2, characterized in that: The method for determining the score of any candidate parking area includes: determining, based on the surrounding image information, area information of the any one candidate parking area in the surrounding image information and environment information of the vehicle; determining a first score for any one of the candidate parking areas based on the area information, the semantic information of the parking intention, and a first parameter among the weight parameters; determining a second score for the any one candidate parking area based on the semantic information of the parking intention, the environmental information, and a second parameter among the weight parameters; determining a third score for the any one candidate parking area based on the historical parking record, the environmental information, and a third parameter among the weight parameters; A score of the any one candidate parking area is determined based on the first score, the second score, and the third score.

4. The parking control method according to any one of claims 2 or 3, characterized in that: The method further includes: determining the weight parameter, including: determining a first parameter of the weight parameters based on a similarity between the semantic information of the parking intention and the area information of any one of the candidate parking areas; determining a second parameter in the weight parameters based on a similarity between the semantic information of the parking intention and the environmental information; A third parameter in the weight parameters is determined based on similarity between the historical parking record and the environmental information.

5. The parking control method according to claim 1, characterized in that: The dynamic parking matching is performed based on the parking intention and the surrounding image information of the vehicle to obtain multiple candidate parking areas including: performing image segmentation on the surrounding image information based on the parking intention, and determining a reference target in the surrounding image information; constructing first prompt data based on a preset first prompt word and image information corresponding to the reference target in the surrounding image information; Based on the first prompt data, a pre-trained generative model is used to determine a plurality of candidate parking areas and a confidence level corresponding to each candidate parking area.

6. The parking control method according to claim 1, characterized in that: The method further comprises: When controlling the vehicle to move according to the movement path indicated by the planning result, if the vehicle meets a preset condition, determining a positional relationship between the vehicle and the target parking area based on real-time image information of the vehicle acquired in real time; updating the movement path based on the positional relationship to obtain an updated movement path; The vehicle is controlled to travel to the target parking area according to the updated movement path.

7. The parking control method according to claim 6, characterized in that: The preset condition includes any one of the following: The speed of the vehicle is greater than or equal to a preset speed threshold; The distance between the vehicle and any target in the real-time image information is less than or equal to a preset distance threshold.

8. The parking control method according to claim 1, characterized in that: The method of determining the parking intention includes: Parse the semantics of the pre-acquired parking requirements to obtain the corresponding structured semantic information; determining, based on sensor data of the vehicle, position parameters corresponding to the orientation information in the structured semantic information; the position parameters being used to characterize a relative positional relationship between the parking area indicated by the parking requirement and the vehicle; The parking intention is determined according to the structured semantic information and the position parameter.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, and the processor implements the parking control method according to any one of claims 1 to 8 when executing a computer program stored in the memory.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the parking control method according to any one of claims 1 to 8 is implemented.

Images