Method for automatically finding parking space, method for controlling vehicle, and electronic device

Abstract

The invention provides a method for automatically searching for a parking space, and the method comprises the steps: receiving a parking space searching instruction transmitted by a vehicle user, and the parking space searching instruction comprises at least two demand preferences of the vehicle user for an expected parking space in different dimensions; obtaining parking lot information and parking state information of a parking lot where the vehicle user is located; based on the parking space searching instruction, the parking lot information and the parking state information, determining a final recommendation level for each candidate parking space in the parking lot; determining a target recommended parking space from the candidate parking spaces based on the final recommendation level; and outputting the target recommended parking space to the vehicle user. The application also proposes a method for controlling a vehicle, an electronic device and a computer program product. According to the scheme, the multi-dimensional composite requirements of the user for the ideal parking space can be comprehensively considered, the recommendation level is intelligently allocated, and the problem that the user is difficult to find the parking space in an unfamiliar parking lot is solved by taking the user requirements as the guidance.

Description

Technical Field

[0001] This application relates to a method for automatically finding parking spaces, a method for controlling a vehicle, an electronic device, and a computer program product. Background Technology

[0002] With the continuous increase in car ownership, parking difficulties in public parking areas such as parking garages and underground parking lots in large shopping malls are becoming increasingly prominent. While some existing navigation or driver assistance functions can provide parking lot maps or parking space guidance, it still requires a long period of searching in densely populated parking lots to find an available space. At the same time, users' demands for diverse and personalized parking spaces are gradually increasing, but existing systems generally lack the ability to match parking space attributes with user needs, making it difficult to recommend ideal parking spaces that meet individual requirements. Furthermore, some public parking areas have complex environments, and users' driving skills vary, leading to frequent parking difficulties.

[0003] To address this, existing solutions propose various parking space recommendation methods. However, these methods typically only rank and score available parking spaces based on a single dimension (e.g., simply by their distance from the mall entrance) or rely solely on preset general optimization objectives (e.g., minimizing the average walking distance across the mall) rather than directly responding to explicit user instructions.

[0004] Therefore, existing parking space search solutions still have many shortcomings. Summary of the Invention

[0005] The purpose of this application is to provide a method for automatically finding parking spaces, a method for controlling vehicles, an electronic device, and a computer program product, so as to at least solve some of the problems in the prior art.

[0006] According to a first aspect of this application, a method for automatically finding parking spaces is provided, the method comprising the following steps: Step S1: Receive a parking space search instruction from a vehicle user, wherein the parking space search instruction includes at least two demand preferences of the vehicle user for different dimensions of the desired parking space. Step S2: Obtain parking information and parking status information of the parking lot where the vehicle user is located; Step S3: Based on the parking space search instruction, the parking lot information, and the parking status information, determine the final recommendation level for each candidate parking space in the parking lot; Step S4: Based on the final recommendation level, determine the target recommended parking space from the candidate parking spaces; and Step S5: Output the determined target recommended parking space to the vehicle user.

[0007] This application specifically includes the following technical concept: This solution can not only directly respond to users' parking space search commands and accurately analyze multi-dimensional needs and preferences, but more importantly, it can comprehensively consider users' complex needs for ideal parking spaces and intelligently allocate differentiated recommendation levels based on the overall fit between each parking space and the user's needs, thereby accurately determining the final recommended parking space. Thus, it truly achieves a user-demand-oriented approach, effectively helping users quickly locate parking spaces that highly match their personalized needs in large parking lots, completely solving the pain point of users having difficulty finding parking spaces in unfamiliar parking lots.

[0008] In one exemplary embodiment, the at least two demand preferences include at least two of the following: parking space type preference, parking space location preference, hot zone preference, parking safety preference, and cost preference; the parking lot information includes: parking space spatial distribution information, road topology information, traffic flow information of each area of ​​the parking lot, parking space type information, parking space shape information, parking lot entrance and exit location information, and / or POI point of interest distribution information; the parking status information includes: parking space occupancy status information, parking density information, and / or vehicle parking posture information in each parking space.

[0009] In an exemplary embodiment, step S3 includes: determining a first type of evaluation index based on the parking space search instruction, the first type of evaluation index being used to measure the comprehensive matching degree between the candidate parking space and the at least two demand preferences; determining a second type of evaluation index based on parking lot information and parking status information, the second type of evaluation index including: the distance between the parking space and the current location of the vehicle, the parking space occupancy status, parking density, the distance from the parking space to the parking lot entrance / exit location, and / or the distance from the parking space to the elevator; determining a recommendation priority for each candidate parking space under each evaluation index dimension included in the first and second types of evaluation indexes; and summarizing the recommendation priorities determined for each candidate parking space under all evaluation index dimensions to determine the final recommendation level of the candidate parking space.

[0010] In one exemplary embodiment, the final recommendation level of a candidate parking space is determined by the following method: For each candidate parking space, across all evaluation metrics: -If the number and / or proportion of evaluation indicators that result in a recommendation priority of "not recommended" are greater than or equal to the first threshold, then the final recommendation level of the candidate parking space is "not recommended". - If the number and / or proportion of evaluation indicators with a recommendation priority of "not recommended" is less than the first threshold, and the number and / or proportion of evaluation indicators with a recommendation priority of "preferred recommendation" is greater than or equal to the second threshold, then the final recommendation level for the candidate parking space is "preferred recommendation"; and / or - If the number and / or proportion of evaluation indicators with a recommendation priority of "not recommended" is less than the first threshold, and the number and / or proportion of evaluation indicators with a recommendation priority of "generally recommended" is greater than or equal to the third threshold, then the final recommendation level of the candidate parking space is "generally recommended".

[0011] In one exemplary embodiment, if the final recommendation level of all candidate parking spaces is "not recommended", then it is determined that there is no target recommended parking space, and a prompt message indicating that there is no recommended parking space is output to the vehicle user; if there is no candidate parking space with a final recommendation level of "preferred", but there is at least one candidate parking space with a final recommendation level of "generally recommended", then a target recommended parking space is selected from all "generally recommended" candidate parking spaces according to a preset sorting rule; if there is at least one candidate parking space with a final recommendation level of "preferred", then a target recommended parking space is selected from all "preferred" candidate parking spaces according to a preset sorting rule.

[0012] In one exemplary embodiment, the method further includes: continuously monitoring changes in parking lot information and parking status information; and, in response to detecting the changes, re-determining the final recommendation level of each candidate parking space, and dynamically updating the current target recommended parking space based on the re-determined final recommendation level.

[0013] In one exemplary embodiment, the method further includes: after outputting the determined target recommended parking space to the vehicle user, receiving a confirmation instruction or a rejection instruction from the vehicle user regarding the target recommended parking space; if a confirmation instruction is received, navigating the vehicle with the target recommended parking space as the destination; and / or, if a rejection instruction is received, removing the current target recommended parking space from the candidate parking spaces and re-determining the target recommended parking space from the remaining candidate parking spaces.

[0014] According to a second aspect of this application, a method for controlling a vehicle is provided, the method comprising the steps of: obtaining a target recommended parking space determined by the method according to a first aspect of this application; and controlling the vehicle to automatically drive to the target recommended parking space and automatically park in the target recommended parking space.

[0015] Therefore, it can not only accurately match the optimal parking space, but also support a fully automated process from finding a parking space and navigation to automatic parking, completely solving the problems of low parking space finding efficiency and difficult parking operation in traditional solutions.

[0016] According to a third aspect of this application, an electronic device is provided, the electronic device including a memory and a processor, the memory storing computer program instructions, which, when executed by the processor, enable the processor to perform the method according to the first and / or second aspects of this application.

[0017] According to a fourth aspect of this application, a computer program product includes computer program instructions, wherein, when executed by a processor, the computer program instructions enable the processor to perform the method according to the first and / or second aspects of this application. Attached Figure Description

[0018] The principles, features, and advantages of this application will be better understood below with reference to the accompanying drawings. The drawings include: Figure 1 A schematic diagram of an electronic device according to an exemplary embodiment of this application is shown; Figure 2 A flowchart illustrating a method for automatically finding parking spaces according to an exemplary embodiment of this application is shown; Figure 3 A flowchart of a method for controlling a vehicle according to an exemplary embodiment of this application is shown; and Figures 4A to 4E A schematic diagram of a display interface for an automatic parking space finding function of a vehicle according to an exemplary embodiment of this application is shown. Detailed Implementation

[0019] To make the technical problems to be solved, the technical solutions, and the beneficial technical effects of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and several exemplary embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit the scope of protection of this application.

[0020] Figure 1 A schematic diagram of an electronic device according to an exemplary embodiment of this application is shown.

[0021] The electronic device 10 can be integrated into a vehicle 1 equipped with intelligent driving functions (such as autonomous driving and / or driver assistance functions), specifically within the intelligent driving domain controller of the vehicle 1. The electronic device 10 includes a processor and a memory (not shown for simplicity). The memory stores computer program instructions, which can be stored in computer-readable storage media such as hard disks, RAM, or flash memory cards. The processor can be a central processing unit (CPU), microcontroller unit (MCU), graphics processing unit (GPU), neural network processing unit (NPU), digital signal processor (DSP), or other general-purpose processor. When the processor executes the computer program instructions in the memory, it can implement methods for automatically finding parking spaces and methods for controlling the vehicle 1.

[0022] like Figure 1As shown, the electronic device 10 can be connected to multiple vehicle sensors and actuators, for example, via an in-vehicle communication network (such as CAN, FlexRay, MOST, or other wired or wireless communication methods).

[0023] Specifically, the electronic device 10 can connect to the in-vehicle communication unit 11 (e.g., a T-Box) and obtain necessary information from the outside through the in-vehicle communication unit 11. For example, the electronic device 10 can obtain digital map data of the target parking lot (such as lane topology, parking space coordinates, parking space type, parking lot entrances and exits, and points of interest (POI) locations) from the map provider cloud platform (such as Gaode, Baidu, etc.) and the parking management server through the in-vehicle communication unit 11. In some embodiments, the parking map data can also be format-converted by the navigation domain controller of vehicle 1 before being provided to the electronic device 10 in a way that adapts the interface protocol. In addition, through the in-vehicle communication unit 11, the electronic device 10 can also communicate with the parking management server and the in-vehicle terminals of other vehicles 1 to obtain real-time parking status information in the parking lot, such as the occupancy status of each parking space, parking density, and real-time traffic flow.

[0024] The electronic device 10 can also be connected to the vehicle status sensor 12, which includes, for example, a GPS sensor, a speed sensor, an acceleration sensor and a yaw rate sensor, for acquiring signals characterizing the motion and posture of the vehicle 1 itself.

[0025] The electronic device 10 can also be connected to the environmental sensor 13, which includes one or more combinations of cameras, millimeter-wave radar and lidar, for acquiring visual images of the parking lot environment, radar point clouds and other target detection signals, thereby enabling obstacle detection, lane line recognition, parking space status detection and so on.

[0026] The electronic device 10 can also be connected to the human-machine interface unit 20, which may include, for example, an input unit and an output unit. The input unit may be, for example, a voice recognition module, a text input module, virtual buttons, an in-cabin camera, etc., to receive parking space search commands issued by the user, and to receive feedback operations such as confirmation, rejection, or selection of system-recommended parking spaces by the user. The output unit may include, for example, a visual output module (e.g., a central control display, instrument panel display, head-up display, etc.) and an acoustic output module (e.g., a speaker or buzzer), to transmit relevant information such as the generated target recommended parking space, navigation guidance route, and operation prompts to the user.

[0027] The electronic device 10 can also be connected to the driving actuator 30 of the vehicle 1, which includes, for example, a transmission unit, a steering unit, and a braking unit. The driving actuator can, for example, perform corresponding driving actions according to control commands generated by the electronic device 10 and / or the intelligent driving domain controller, thereby guiding and controlling the vehicle 1 to automatically drive to the target recommended parking space and complete the automatic parking operation.

[0028] It should be understood that Figure 1 The number and types of various sensors and actuators connected to the electronic device 10 shown are merely examples and are not intended to be limiting. In practical applications, other types or numbers of sensors and actuators may be used in the vehicle 1 to meet specific needs and conditions.

[0029] It should also be understood that Figure 1 The software architecture shown is merely exemplary, and the connections and functional divisions between modules can be adaptively adjusted according to the actual system architecture. For example, in practice, electronic device 10 can also be implemented as a separate control module independent of the intelligent driving domain controller, or implemented in a cross-domain manner. Furthermore, the deployment method of electronic device 10 is not limited to the local vehicle 1, but can also be remotely deployed, for example, by setting it up on the vehicle manufacturer's back-end server or cloud platform.

[0030] Figure 2 A flowchart of a method for automatically finding parking spaces according to an exemplary embodiment of this application is shown. The method includes steps S1 to S5, wherein step S3 is exemplarily shown as including sub-steps S31 to S34.

[0031] In step S1, a parking space search instruction issued by the vehicle user is received. The parking space search instruction includes at least two demand preferences of the vehicle user for different dimensions of the desired parking space.

[0032] In one embodiment, the parking space search instruction can be received in multiple ways. For example, the parking space search instruction can be a voice instruction, in which case the in-vehicle voice recognition module can receive and parse the user's voice input, and extract the at least two desired preferences. Alternatively, the instruction can be a text instruction input by the user via a touchscreen module, or a gesture instruction from the user can be recognized by an in-vehicle camera.

[0033] At least two different dimensions of demand preferences, such as independent evaluation criteria explicitly expressed in user commands and belonging to different attribute categories, can be separately parsed and quantified in parking space recommendation mechanisms. For example, when a user issues a voice command, "Find me a charging parking space near a movie theater," after recognizing it as a parking space search command, a natural language processing model can parse out the two independent dimensions of demand preferences it contains: namely, parking space location preference (near the movie theater) and parking space type preference (charging parking space). Similarly, the voice command, "Find me a parking space near the elevator that is unlikely to be hit by a car opening its door next to me," can be parsed to include parking space location preference (near the elevator) and parking safety preference (avoiding collisions with neighboring cars opening their doors).

[0034] For example, the at least two demand preferences include at least two of the following: - Parking space type preference, such as: charging parking space, regular parking space, disabled parking space, women-only parking space, specific brand or model-specific parking space, family travel / large vehicle parking space, rental / monthly card parking space, temporary parking (such as 15-minute quick pickup) parking space; - Parking space location preference, such as: parking spaces close to specific facilities or points of interest, such as elevators, exits, stairs, shopping mall / cinema entrances, restrooms, charging stations, car washes, etc. - Hotspot preferences, such as: high-frequency parking areas for owners of specific car brands, popular parking areas, etc.; - Parking safety preferences, such as: parking space spaciousness, surrounding visibility, the orderly parking of nearby vehicles, ease of parking / exit, etc.; and / or - Cost preferences, such as eligibility for consumption credits, limited-time free parking, member-exclusive discount parking spaces, etc.

[0035] In step S2, the parking information and parking status information of the parking lot where the vehicle user is located are obtained.

[0036] In one embodiment, the vehicle can interact with a map provider's cloud platform via an onboard communication unit to receive map data for the target parking lot. Alternatively, this information can be obtained from a parking lot management server, or it can be identified using the vehicle's own environmental sensors.

[0037] For example, parking information may include: - Parking space spatial distribution information, such as the distribution coordinates and unique numbers of each parking space; - Road topology information, including lane lines, driving directions, and connectivity; - Traffic flow information for each area of ​​the parking lot, such as vehicle density and average speed, congestion, etc. in each zone; - Parking space type information, such as charging parking space, regular parking space, etc.; - Parking space shape information, such as size and orientation; -Location information of parking lot entrances and exits; and / or - POI (Point of Interest) distribution information, such as elevator locations (vertical / escalator), cinemas, shopping malls, hotels, restrooms, car washes, supermarkets, etc.

[0038] In one embodiment, the vehicle-mounted communication unit interacts with the parking management server to obtain real-time parking space occupancy status monitored and reported by sensors deployed at the parking spaces. Alternatively, parking status information can also be detected through the vehicle's own environmental perception sensors. Or, for example, vehicle-to-everything (V2X) communication technology can be used to interact with other parked vehicles or vehicles moving within the parking lot to indirectly obtain parking status information of the surrounding area.

[0039] For example, the parking status information includes: - Parking space occupancy status information, such as indicating whether the parking space is currently occupied or vacant; - Parking density information, such as real-time vehicle distribution density reflecting different zones or floors within a parking lot, can be used to identify areas with concentrated vacancy spaces; and / or - Information on the vehicle's parking posture in each parking space, such as the vehicle model, whether it is centered, whether it is parked over the lines, the gap between the vehicle and the edge of the parking space, and the direction the vehicle is facing. This information can be used, for example, to assess the actual available space in adjacent vacant parking spaces and the difficulty of parking / exiting.

[0040] In step S3, based on the parking space search instruction, the parking lot information, and the parking status information, a final recommendation level is determined for each candidate parking space in the parking lot.

[0041] In this context, "candidate parking spaces" can refer to all parking spaces in the parking lot or on the parking floor where the vehicle is currently located, or it can be a set of vacant parking spaces after initial screening. If the candidate parking spaces are all parking spaces, then the occupancy status must be used as an evaluation indicator when determining the recommendation level. Usually, occupied parking spaces will be directly judged as not recommended. If the candidate parking spaces are only vacant parking spaces, then the occupancy status does not need to be used as an independent evaluation indicator when determining the recommendation level.

[0042] In one embodiment, the final recommendation level can be determined using a parking space recommendation model. This model can be built based on preset rules, such as generating levels through a comprehensive judgment logic of multi-dimensional evaluation indicators; specific implementation details will follow below. Alternatively, the parking space recommendation model can also be a machine learning model, such as a trained classification or regression model, which can directly output the final recommendation level for each candidate parking space based on the input information.

[0043] In one embodiment, the final recommendation level can be expressed in a quantitative form, such as by numerical values ​​or scores, to reflect the degree of recommendation. Alternatively, it can be expressed in a qualitative grading form, such as by dividing it into levels like "not recommended," "generally recommended," and "preferred recommended," so that users can understand it intuitively and the system can make subsequent decisions.

[0044] exist Figure 2 In the specific embodiment shown, step S3 is exemplarily illustrated as including sub-steps S31 to S34.

[0045] In sub-step S31, a first type of evaluation index is determined based on the parking space search instruction. This first type of evaluation index is used to measure the comprehensive matching degree between the candidate parking space and at least two demand preferences of the vehicle user.

[0046] For example, if a vehicle user's needs and preferences are "close to the elevator" and "charging parking space", then the corresponding first-class evaluation indicators can be set as "walking distance to the elevator" and "whether the parking space has charging function".

[0047] For example, if a vehicle user's needs and preferences are "a spacious parking space that makes it easy to take out / put in an infant" and "parking in an area where XX brand car owners frequently park", then the corresponding first-category evaluation indicators can be set as "parking space size and surrounding space openness" and "whether the parking space is located in the hot zone where XX brand car owners have historically parked frequently".

[0048] Furthermore, it's possible that vehicle users express multiple parallel needs within the same dimension. In such cases, these complex needs can be integrated into a refined evaluation metric for a single dimension. For example, if a user simultaneously requests both "charging parking space" and "women-only parking space," the corresponding first-category evaluation metric could be set as "parking space type match," which could be further divided into three levels: "fully matched," "partially matched," and "not matched," to quantify the degree to which parking spaces align with the vehicle user's multiple preferences within that dimension.

[0049] In sub-step S32, a second type of evaluation index is determined based on parking lot information and parking status information. The second type of evaluation index includes, for example, the distance between the parking space and the current location of the vehicle, the parking space occupancy status, parking density, the distance from the parking space to the parking lot entrance and / or the distance from the parking space to the elevator.

[0050] The second type of evaluation indicators are, for example, general indicators that do not directly stem from specific needs explicitly expressed in user instructions. However, based on driving experience and behavioral research, these indicators generally affect parking convenience, efficiency, and the overall experience. By introducing the second type of evaluation indicators, we can not only respond to users' proactive preferences but also comprehensively consider objective environmental factors and traffic efficiency during the recommendation process. This allows us to further optimize the actual availability and safety of parking spaces and the smoothness of the overall parking process while meeting personalized needs, achieving a more comprehensive and rational intelligent recommendation.

[0051] In one embodiment, the selection of the second type of evaluation indicators can be associated with the first type of evaluation indicators. That is, for different types of the first type of evaluation indicators, corresponding combinations of the second type of evaluation indicators can be predefined or dynamically matched. For example, when a user's demand preference is "parking space for disabled persons", "distance to accessible elevator / passage" can be automatically associated when selecting the second type of evaluation indicators, thereby further optimizing the rationality and convenience of the actual user experience while meeting explicit needs.

[0052] In one embodiment, since the first type of evaluation index directly reflects the degree of fit between the candidate parking space and the user's explicitly expressed needs and preferences, the first type of evaluation index can be assigned a higher weight or priority so that it plays a leading role in the determination of the final recommendation level.

[0053] In sub-step S33, a recommendation priority is determined for each candidate parking space under each evaluation indicator dimension included in the first and second evaluation indicators. The recommendation priority can be labeled using discrete levels, including at least three levels: "not recommended," "generally recommended," and "preferred recommendation."

[0054] In one embodiment, the recommendation priority can be further characterized by priority labels P1, P2, and P3, corresponding to the three recommendation levels mentioned above, where P1 represents "preferred recommendation," P2 represents "general recommendation," and P3 represents "not recommended." Furthermore, the recommendation priority can be further subdivided according to actual needs, for example, expanded to five or more levels, including "strongly recommended," "preferred recommendation," "general recommendation," "barely recommended," and "not recommended," to provide a more refined evaluation distinction.

[0055] For each evaluation metric, such as predefined mappings between different values ​​or states of candidate parking spaces within that dimension and their recommendation priorities, a corresponding recommendation priority is assigned based on the actual parking lot information and parking status information, according to the specific value or state of the candidate parking space on that metric. Each candidate parking space is evaluated across multiple evaluation metric dimensions, and a corresponding recommendation priority is obtained for each dimension.

[0056] For example, for the evaluation indicator "walking distance to the elevator", the following can be preset: if the distance is ≤20 meters, the recommendation priority is P1 (preferred); if 20 meters < distance ≤50 meters, the recommendation priority is P2 (generally recommended); if the distance is >50 meters, the recommendation priority is P3 (not recommended).

[0057] For example, for the evaluation indicator "whether it is a charging parking space", the following can be set: if yes, the recommendation priority is P1 (preferred recommendation); if no, the recommendation priority is P2 (general recommendation).

[0058] In another embodiment, recommendation priority may not be represented by discrete levels, but by continuous numerical values. For example, for distance-based evaluation indicators, a continuous score between 0 and 1 can be calculated directly from the distance value using a function (such as an inverse proportional function or an exponential decay function). A higher score indicates higher priority for that indicator, thereby achieving a smoother and more refined quantitative evaluation.

[0059] In sub-step S34, for each candidate parking space, the recommendation priority determined under all evaluation index dimensions is summarized to determine the final recommendation level of the candidate parking space.

[0060] In one embodiment, the aggregation method can employ statistical methods, such as averaging the recommendation priorities obtained across all dimensions, calculating the median, or constructing a normal distribution. In another embodiment, the determination can also be based on preset rules related to the number or proportion of priorities. Furthermore, a trained machine learning model can be used, taking the set of recommendation priorities obtained across each dimension as input, and directly outputting the final recommendation level for the candidate parking space.

[0061] In one embodiment, different weights can be assigned to different evaluation indicators. For example, a higher weight can be assigned to the first type of evaluation indicator and a lower weight can be assigned to the second type of evaluation indicator. Weighted calculations are performed during the aggregation process to reflect the degree of influence of different dimensions on the final decision.

[0062] For example, the determination can be based on preset rules related to the number or proportion of recommended priorities. Specifically, for each candidate parking space, across all evaluation metrics: - If the number and / or proportion of evaluation indicators that result in "Not Recommended P3" are greater than or equal to the first threshold, then the final recommendation level for the candidate parking space is "Not Recommended P3". For example, the first threshold can be set to 1, meaning that if P3 appears in any evaluation indicator dimension, the final recommendation level is directly determined as "Not Recommended".

[0063] - If the number and / or proportion of evaluation indicators that result in "P3 not recommended" is less than the first threshold, and the number and / or proportion of evaluation indicators that result in "P1 recommended" is greater than or equal to the second threshold, then the final recommendation level for the candidate parking space is "P1 recommended". For example, the second threshold can be set to 1, meaning that if at least one P1 appears in any evaluation indicator dimension, the final recommendation level is "recommended".

[0064] - If the number and / or proportion of evaluation indicators that result in "P3 not recommended" is less than the first threshold, and the number and / or proportion of indicators that result in "P2 generally recommended" is greater than or equal to the third threshold, then the final recommendation level of the candidate parking space is "P2 generally recommended".

[0065] The above thresholds can be flexibly configured according to actual needs, and can be weighted and calculated in combination with the weights of each evaluation indicator to achieve a more flexible and reasonable level determination logic.

[0066] For example, for a candidate parking space: If all evaluation metrics result in P1, it is defined as "preferred recommendation"; If all evaluation metrics result in P1 or P2, it is defined as "normal recommendation"; When a dimension of the evaluation indicator appears as P3, and the proportion of P3 is less than a certain fixed value, it is defined as "general recommendation"; When a dimension of the evaluation indicator appears as P3, and the proportion of P3 is greater than a certain value, it is defined as "not recommended". When there are no available parking spaces, it is also defined as "not recommended".

[0067] The following specific example illustrates how to determine the final recommendation level for candidate parking spaces. In this example, a vehicle user issues a parking space search command: "Find me a women-only parking space with a charging station near a movie theater." As shown in Table 1 below, the recommendation priority of one candidate parking space in the parking lot is presented in matrix form under multiple evaluation index dimensions. Each row in the matrix represents an evaluation index dimension. The first two rows, for example, represent the first type of evaluation index, directly related to the user's expressed needs and preferences. The subsequent rows represent the second type of evaluation index, covering general indicators related to parking convenience, efficiency, and environmental conditions. Each column corresponds to a recommendation priority level (from right to left, for example, "Preferred P1", "Generally Recommended P2", and "Not Recommended P3").

[0068] Table 1

[0069] Table 1 shows that, within the rows corresponding to each evaluation metric dimension, different states of candidate parking spaces under each evaluation metric dimension correspond to different recommendation priorities. Based on the actual situation, the recommendation priority of candidate parking spaces under each evaluation metric dimension can be determined.

[0070] In this embodiment, the first type of evaluation index includes: The distance between the parking space and the movie theater is as follows: if the distance is "near", then P1 is the preferred choice; if the distance is "medium" or "far", then P2 is the general recommendation. Parking space type hit rate: If none of the parking space types specified by the user are matched, the recommendation priority is not recommended (P3); if some are matched, the recommendation priority is generally recommended (P2); if all are matched, the recommendation priority is P1.

[0071] The second category of evaluation indicators includes, for example: The distance between the parking space and the vehicle's current position. If the distance is "near", it corresponds to P1; if it is "medium" or "far", it corresponds to P2. Parking space occupancy status: if occupied, it corresponds to P3; if unoccupied, it corresponds to P2. Parking density: low density corresponds to P1, medium or high density corresponds to P2. The distance from the parking space to the parking lot exit is P1 if the distance is "near" and P2 if the distance is "medium" or "far". The distance from the parking space to the elevator is P1 if the distance is "near", and P2 if the distance is "medium" or "far".

[0072] In this specific embodiment, for the exemplary candidate parking spaces, the recommendation priorities obtained under various evaluation index dimensions are as follows: Distance of parking space from the movie theater: P2 Parking space type hit rate: P2 Distance between parking space and vehicle's current position: P1 Parking space occupancy status: P2 Parking density: P2 Distance from parking space to parking lot exit: P2 Distance from parking space to elevator: P2 Based on the preset aggregation rules, the final recommendation level of the candidate parking space is determined to be, for example, P2 (general recommendation).

[0073] In step S4, a target recommended parking space is determined from the candidate parking spaces based on the final recommendation level.

[0074] In one embodiment, if all candidate parking spaces have a final recommendation level of "not recommended," then it is determined that there is no target recommended parking space, and a prompt message indicating that there is no recommended parking space is displayed to the vehicle user. If there are no candidate parking spaces with a final recommendation level of "preferred," but at least one candidate parking space with a final recommendation level of "generally recommended," then a target recommended parking space is selected from all "generally recommended" candidate parking spaces according to a preset sorting rule. If there are at least one candidate parking space with a final recommendation level of "preferred," then a target recommended parking space is selected from all "preferred" candidate parking spaces according to a preset sorting rule.

[0075] In one embodiment, the preset sorting rule is to sort the candidate parking spaces according to the driving distance or estimated arrival time between the candidate parking spaces and the vehicle's current location, and select the closest or shortest arrival time from multiple "preferred" candidate parking spaces. Alternatively, the preset sorting rule can be to score the candidate parking spaces according to their relevance to user needs and preferences, and select the one with the highest score.

[0076] In one embodiment, due to frequent changes in parking lot status (such as parking spaces being occupied or vehicles leaving), the determination of the target recommended parking space is not a one-time process but can be dynamically updated in real time. For example, changes in parking lot information and parking status information can be continuously monitored. In response to the monitored changes (or changes that meet predetermined conditions), the final recommendation level of each candidate parking space is re-determined, and the current target recommended parking space is dynamically updated based on the re-determined final recommendation level. For example, if the re-determination result indicates that there is a candidate parking space with a higher final recommendation level than the current target recommended parking space, then a change to the target recommended parking space can be performed.

[0077] In step S5, the determined target recommended parking space is output to the vehicle user.

[0078] In one embodiment, the vehicle's current location can be dynamically marked on the parking lot floor plan displayed on the in-vehicle display, and the target recommended parking space, the expected driving route to the parking space, and the parking space number, floor, and walking distance to key points of interest can be displayed. Furthermore, a recommendation reason can be generated and displayed, such as "Recommendation reason: This parking space is a charging parking space and is approximately 15 meters from the elevator," and the locations of elevators, exits, and other relevant points of interest can be simultaneously marked on the floor plan. Additionally, parking space information and recommendation reasons can be announced via the in-vehicle voice system, such as "We have found charging parking space number 205 in area B, approximately 15 meters from the elevator."

[0079] In one embodiment, when a change in the parking lot status is detected and it is determined that the current recommendation results need to be dynamically updated, a change notification can be sent to the user through the in-vehicle human-machine interface, and interactive options for confirmation or cancellation can be provided.

[0080] In one embodiment, after outputting the identified target recommended parking space to the vehicle user, a confirmation query interface can also be displayed to the vehicle user, requesting confirmation from the vehicle user whether they agree to use the target recommended parking space as the final destination. The vehicle user's feedback instruction is then detected. This feedback instruction includes, for example, a confirmation instruction and a rejection instruction, and optionally, an instruction for the user to manually select a parking space from the list of recommended parking spaces provided by the system.

[0081] If the system receives confirmation from the vehicle user regarding the target recommended parking space, it will navigate the vehicle to that location. After user confirmation, the interface can further display the navigation path from the current location to the target parking space, or directly initiate the automatic parking guidance process, thus providing the user with an intuitive and seamless interactive experience.

[0082] If a rejection instruction is received from a vehicle user, the current target recommended parking space can be removed from the candidate parking spaces, and a new target recommended parking space can be determined from the remaining candidate parking spaces.

[0083] Figure 3 A flowchart of a method for controlling a vehicle according to an exemplary embodiment of this application is shown. The method includes steps 310 and 320.

[0084] In step 310, obtain using Figure 2 The method shown identifies the target recommended parking space.

[0085] In one embodiment, the system can proactively receive parking space search instructions from vehicle users. Alternatively, when the vehicle arrives within the parking lot area and the automatic parking function is activated, the system can proactively inquire about the user's parking space preferences via a human-machine interface, then receive at least two user-inputted preferences, and execute a parking space recommendation process based on these preferences.

[0086] In one embodiment, after a vehicle user issues a parking space search command, the vehicle immediately enters an automatic cruise mode to search for a parking space. In this mode, the vehicle begins to autonomously control its speed and steering, driving along feasible paths within the parking lot, while the system continuously evaluates candidate parking spaces along the way until a target recommended parking space that matches the user's preferences is determined.

[0087] In step 320, the vehicle is controlled to automatically drive to the target recommended parking space and automatically park in the target recommended parking space.

[0088] In one embodiment, the parking process can be automatically initiated upon receiving confirmation from the user regarding the target recommended parking space. The vehicle first automatically drives to the vicinity of the target recommended parking space based on a planned route, and then automatically performs the parking operation to complete the entire parking process.

[0089] In another embodiment, if the vehicle is already in an automatic cruise state searching for a parking space before the target recommended parking space is determined, then after the target recommended parking space is determined, it can directly start from the current position, adjust or continue to execute driving control, so as to guide the vehicle to automatically drive to the target recommended parking space and complete automatic parking.

[0090] In another embodiment, during automatic parking, if the vehicle has not yet begun parking and a better parking space than the currently recommended target parking space is detected, the vehicle user can be prompted to update the recommended target parking space. If the vehicle has already begun the automatic parking process, the dynamic updating of the recommended target parking space can be paused to avoid operational interruption and safety hazards.

[0091] Figures 4A to 4E A schematic diagram of a display interface for an automatic parking space finding function of a vehicle according to an exemplary embodiment of this application is shown.

[0092] Figure 4A The display screen 50 of the vehicle's central control display shows the vehicle having just entered a public parking lot. The screen displays a parking lot floor plan, showing the distribution of parking spaces within a certain range around the vehicle. Finding parking spaces scarce, the vehicle user issues a voice command, "Find me a charging parking space near the movie theater," triggering the automatic parking space search function. The microphone icon in the upper left corner of the screen indicates that the system is receiving voice input; the converted text of the command can be displayed next to it.

[0093] Figure 4B In the process, the interface content has been updated according to user commands: for example, the upper left corner of the interface can display the text "Looking for charging parking spaces near movie theaters," while continuously and dynamically updating the vehicle's real-time location icon and the distribution and real-time occupancy status of surrounding parking spaces. At this time, the vehicle has automatically entered the "Find a Parking Space" automatic cruise mode, autonomously driving slowly along the passable lanes in the parking lot, and continuously scanning the surrounding environment using onboard sensors, waiting for the evaluation results of the target recommended parking space.

[0094] exist Figure 4CThe system has identified a suitable recommended parking space. At this point, a text prompt box in the upper left corner of the display interface may ask, "Parking space found, park now?", providing two touch options, such as "Y" and "N" (or "Yes" and "No"). Simultaneously, the location of the recommended parking space 510 is marked on the displayed parking lot floor plan, and the expected automatic parking path from the vehicle's current location to the recommended parking space 510 is drawn. Furthermore, the display interface also simultaneously marks the locations of key facilities related to the recommended parking space 510, such as the charging station location 511 and the elevator location 512 leading directly to the cinema, to help vehicle users intuitively understand the reasons for the recommendation.

[0095] exist Figure 4D In the process, if the vehicle user clicks the "N" (or "No") option, they can refuse to park in the currently recommended parking space. In response to this action, the display will show "Continue searching for other parking spaces" and the vehicle will re-enter parking space search mode. Simultaneously, the recommended parking space and its corresponding automatic parking path marked on the parking lot map will disappear, and the vehicle will continue searching for other suitable recommended parking spaces while in automatic cruise control mode.

[0096] exist Figure 4E In the process, vehicle users can click the "Y" (or "Yes") option to confirm acceptance of parking in the recommended parking space. The display will then show "Parking for you" and show the vehicle's parking process along the planned path in real time. Once the vehicle is successfully parked, the interface will update to show that the vehicle is stably parked in the space, and may display text such as "Parking space found and successfully parked," thus intuitively indicating to the user that the entire parking space search and parking process has been successfully completed.

[0097] Although specific embodiments of this application are described in detail herein, they are given for illustrative purposes only and should not be construed as limiting the scope of this application. Various substitutions, modifications, and alterations can be conceived without departing from the spirit and scope of this application.

Claims

1. A method for automatically finding a parking space, the method comprising the steps of: Step S1, receiving a parking space finding instruction issued by a vehicle user, the parking space finding instruction comprising at least two demand preferences of different dimensions for a desired parking space by the vehicle user; Step S2, obtaining parking lot information and parking state information of a parking lot where the vehicle user is located; Step S3, determining a final recommendation level for each candidate parking space in the parking lot based on the parking space finding instruction, the parking lot information and the parking state information; Step S4, determining a target recommended parking space from the candidate parking spaces based on the final recommendation levels; and Step S5, outputting the determined target recommended parking space to the vehicle user.

2. The method according to claim 1, wherein: the at least two demand preferences comprise at least two of the following: a parking space type preference, a parking space location preference, a hot zone preference, a parking safety preference and a cost preference; the parking lot information comprises: parking space spatial distribution information, road topology information, traffic flow information of each area of the parking lot, parking space type information, parking space shape information, parking lot entrance location information and / or POI (Point of Interest) distribution information; the parking state information comprises: parking space occupancy state information, parking density information and / or vehicle parking attitude information on each parking space. Step S3 comprises:

3. The method of claim 1 or 2, wherein, determining a first type of evaluation index based on the parking space finding instruction, the first type of evaluation index being used to measure the comprehensive matching degree of a candidate parking space with the at least two demand preferences; determining a second type of evaluation index based on the parking lot information and the parking state information, the second type of evaluation index comprising: the distance of a parking space from the current location of the vehicle, the occupancy state of the parking space, the parking density, the distance of the parking space from the parking lot entrance location and / or the distance of the parking space from an elevator; determining a recommendation priority for each candidate parking space in each evaluation index dimension contained in the first type of evaluation index and the second type of evaluation index, respectively; and for each candidate parking space, aggregating the recommendation priorities determined for it in all evaluation index dimensions to determine the final recommendation level of the candidate parking space. The final recommendation level of a candidate parking space is determined in the following manner:

4. The method of claim 3, wherein, for each candidate parking space, in all evaluation index dimensions: if the number and / or proportion of evaluation indexes resulting in a recommendation priority of "not recommended" is greater than or equal to a first threshold, the final recommendation level of the candidate parking space is "not recommended"; if the number and / or proportion of evaluation indexes resulting in a recommendation priority of "not recommended" is less than the first threshold, and the number and / or proportion of evaluation indexes resulting in a recommendation priority of "highly recommended" is greater than or equal to a second threshold, the final recommendation level of the candidate parking space is "highly recommended"; and / or if the number and / or proportion of evaluation indexes resulting in a recommendation priority of "not recommended" is less than the first threshold, and the number and / or proportion of evaluation indexes resulting in a recommendation priority of "general recommendation" is greater than or equal to a third threshold, the final recommendation level of the candidate parking space is "general recommendation".

5. The method according to claim 4 or 5, wherein, ​ if all the final recommendation levels of the candidate parking spaces are "not recommended", it is determined that there is no target recommended parking space, and a prompt information of no recommended parking space is output to the vehicle user; if there is no candidate parking space with the final recommendation level of "priority recommended", but there is at least one candidate parking space with the final recommendation level of "general recommended", a target recommended parking space is selected from all the "general recommended" candidate parking spaces according to a preset sorting rule; and / or if there is at least one candidate parking space with the final recommendation level of "priority recommended", a target recommended parking space is selected from all the "priority recommended" candidate parking spaces according to a preset sorting rule.

6. The method of any one of claims 1 to 5, wherein, The method further comprises: continuously monitoring changes in parking lot information and parking state information; and in response to monitoring the changes, re-determining the final recommendation level of each candidate parking space, and dynamically updating the current target recommended parking space based on the re-determined final recommendation level.

7. The method of any one of claims 1 to 6, wherein, The method further comprises: after outputting the determined target recommended parking space to the vehicle user, receiving a confirmation instruction or a rejection instruction of the target recommended parking space from the vehicle user; if the confirmation instruction is received, the vehicle is guided to the target recommended parking space as the destination; and / or if the rejection instruction is received, the current target recommended parking space is removed from the candidate parking spaces, and a target recommended parking space is re-determined in the remaining candidate parking spaces.

8. A method for controlling a vehicle, the method comprising the steps of: obtaining a target recommended parking space determined according to the method of any one of claims 1 to 7; controlling the vehicle to automatically drive to the target recommended parking space and automatically park in the target recommended parking space.

9. An electronic device comprising a memory and a processor, the memory storing computer program instructions, when the computer program instructions are executed by the processor, the processor can execute the method according to any one of claims 1 to 8.

10. A computer program product comprising computer program instructions, wherein, The computer program instructions, when executed by the processor, enable the processor to execute the method according to any one of claims 1 to 8. The computer program instructions, when executed by the processor, enable the processor to execute the method according to any one of claims 1 to 8.

Images