Parking recommendation method and apparatus, computer device, and storage medium

Abstract

A parking recommendation method and apparatus, a computer device, and a storage medium. The method comprises: obtaining a field end impact factor and a vehicle end impact factor (step S110); by using a position relationship between a vehicle and the field end impact factor in a field end space region, constructing a neural network model of field end distribution, and calculating a trajectory impact factor of a travel trajectory (step S120); inputting the vehicle end impact factor and the trajectory impact factor to a preset parking evaluation model, so as to calculate a parking space evaluation value by using a transverse space configuration and a charging configuration needed by the vehicle, and further on the basis of the parking space evaluation value, obtaining candidate parking spaces (step S30); and recommending an optimal parking space from among the candidate parking spaces by using a recommendation strategy (step S140). According to the method, from the field end impact factor, the trajectory impact factor of the travel trajectory can be calculated, the candidate parking spaces are calculated, and the optimal parking space is recommended; and on the basis of a regular impact factor in consideration of impact factors such as transverse spaces needed by a charging pile device and door opening of the vehicle end, the confidence level of the optimal parking space is ensured, the user experience is improved, and the vehicle charging needs are met.

Description

Parking recommendation method, device, computer equipment and storage medium

[0001] This application claims priority to Chinese patent application No. 202311149168.4 filed on September 6, 2023, entitled “A parking recommendation method, device, computer equipment and storage medium,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present application relates to the field of intelligent driving technology, and in particular to a parking recommendation method, apparatus, computer equipment, and storage medium. Background Art

[0003] With the development of intelligent driving technology, the number of vehicles is increasing, and the parking technology related to vehicles is also improving. Currently, parking space recommendation technologies in parking technology include: parking lot recommendation and vehicle recommendation.

[0004] Parking lot recommendations utilize parking lot monitoring equipment at the parking lot (e.g., in a basement) to identify available parking spaces. These recommendations are then based on factors such as the vehicle's current location, path distance, and the location of the path's endpoints (e.g., elevator exits). However, implementation costs are high, and currently only a few parking lots support parking lot recommendations. Furthermore, these recommendations are based on static parking lot information and fail to consider the surrounding conditions of the recommended spaces (e.g., obstacles, road width, and the distance between adjacent vehicles and the recommended space), making parking difficult.

[0005] Vehicle-side parking space recommendation methods use real-time data to detect available parking spaces and obstacles, then perform path planning. Evaluation metrics are then established based on factors such as path length, maximum curvature, and the number of vehicle gear shifts. Parking spaces are then recommended based on these evaluations. However, current mainstream parking planning methods (such as geometric, optimization, and search) rely on path planning efficiency, and they fail to meet real-time requirements. Geometric methods cannot produce a complete, collision-free trajectory from the start point to the end point. The latter two methods also lack the real-time performance required (e.g., requiring over 2 seconds in extreme cases), which is insufficient for user interaction. Furthermore, multi-threaded systems place extremely high demands on computing power and memory.

[0006] Moreover, neither the parking space recommendation method at the terminal nor the vehicle-side parking space recommendation method takes into account the subjective needs of passengers (for example, the driver has lateral safety distance requirements due to weight, physical defects, etc.) and the current objective needs of the vehicle (for example, vehicle charging), resulting in a poor user experience after parking is completed. Summary of the Invention

[0007] Based on this, it is necessary to provide a parking recommendation method, device, computer equipment and storage medium to address the above technical problems.

[0008] In a first aspect, the present application provides a parking recommendation method, the method comprising:

[0009] In response to a parking request, obtaining a parking-side impact factor and a vehicle-side impact factor in the parking request;

[0010] By using the positional relationship between the current vehicle and the influencing factors in the parking space, a neural network model based on the parking distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space.

[0011] The vehicle-side influence factor and trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then candidate parking spaces are obtained based on the parking space evaluation value;

[0012] Use the preset recommendation strategy to recommend the best parking space from the candidate parking spaces.

[0013] In one embodiment, in response to a parking request, the step of obtaining a parking-side impact factor and a vehicle-side impact factor includes:

[0014] Use vehicle-side matching rules to filter the factor classification table;

[0015] According to the factor classification table, receive real-time field-side impact factors and vehicle-side impact factors.

[0016] In one embodiment, the steps of screening the factor classification table based on the parking instruction using the vehicle-side matching rules to determine the vehicle-side impact shadow and the field-side impact factor include:

[0017] According to the parking instruction, obtain the vehicle type, determine whether the vehicle is a tram, and output a first determination result;

[0018] If the first judgment result indicates that the vehicle is an electric vehicle, the factor classification table includes at least vehicle-side influencing factors including the vehicle's remaining power, and further determines whether the vehicle's remaining power is lower than a preset charging threshold, and outputs a second judgment result;

[0019] If, in the second judgment result, the remaining power of the vehicle is lower than the preset charging threshold, the factor classification table includes at least field-side influencing factors including obstacles and charging pile equipment; otherwise, the factor classification table includes at least field-side influencing factors including obstacles.

[0020] In one embodiment, the positional relationship between the current vehicle and the terminal influencing factor in the terminal spatial area, and the steps of constructing a neural network model based on the terminal distribution, include:

[0021] Perform ROI grid division on the field-side spatial area to abstractly represent the field-side influencing factors;

[0022] According to the distance between the vehicle in the driving trajectory and the terminal influencing factor, different grid resolutions are selected to obtain the correlation relationship representing the occupied grids, thereby constructing a neural network model of the terminal distribution.

[0023] In one embodiment, the step of calculating the trajectory influence factor of the vehicle's driving trajectory to each parking space includes:

[0024] Using the influencing factors of the parking lot and the correlation relationship between the occupied grids, the driving trajectory of the vehicle to the parking space is obtained, and the parking lot feature map is obtained based on this;

[0025] Input the field-side feature map into the neural network model, calculate and output the feature true value;

[0026] According to the true value of the feature and the preset cost function, the trajectory influencing factor of the vehicle's driving trajectory to each parking space is calculated.

[0027] In one embodiment, the step of calculating the trajectory influencing factor of the vehicle's driving trajectory to each parking space based on the feature true value and a preset cost function includes:

[0028] The cost function equation (1) is used to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space;

[0029] (1)

[0030] in, represents the cross entropy function, represents the weight function, Indicates the difference between the feature estimate and the true value when i=1; It means that when i=2,...,5, the power of the difference between each feature estimate and the corresponding feature true value is summed.

[0031] In one embodiment, the vehicle-side influence factor and the trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the required lateral space configuration and charging configuration of the vehicle, and then the step of obtaining a candidate parking space based on the parking space evaluation value includes:

[0032] Input the required lateral space configuration and charging configuration of the vehicle into the parking space evaluation function to calculate the parking space evaluation value;

[0033] The parking space evaluation value is calculated using the parking space evaluation function equation (2);

[0034] (2)

[0035] Among them, through Indicates the parking space evaluation value; hour, , , Indicates the dimensionless calibration value; Indicates the lateral distance between the four wheels at the end position and the obstacle. Indicates the distance between the destination location and the nearest charging pile device; Indicates the lateral space required for the vehicle to open the door. Indicates whether charging is needed by (0,1). Represents the weight coefficient.

[0036] Filter candidate parking spaces according to parking space evaluation values ​​and obtain a set of candidate parking spaces .

[0037] In one embodiment, the step of selecting and recommending the optimal parking space from the candidate parking spaces using a preset recommendation strategy includes:

[0038] The recommendation strategy expression (3) is used to select the optimal parking space from the candidate parking spaces;

[0039] The recommended strategy expression (3) is: .

[0040] In a second aspect, the present application provides a parking recommendation method, which is applied to the vehicle side and includes:

[0041] In response to a parking request, obtaining a field-side impact factor in the parking request;

[0042] By utilizing the positional relationship of the parking lot influencing factors in the parking lot spatial area, a neural network model based on the parking lot distribution is constructed to calculate the trajectory influencing factors of the vehicle arriving at each parking space.

[0043] The vehicle-side influence factor and trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then candidate parking spaces are obtained based on the parking space evaluation value;

[0044] Use the preset recommendation strategy to recommend the best parking space from the candidate parking spaces.

[0045] In a third aspect, the present application provides a parking recommendation method, which is applied to a parking lot, wherein the parking lot is in communication with the vehicle. The method includes:

[0046] In response to a parking request, obtaining a vehicle-side impact factor in the parking request;

[0047] Based on the neural network model of the parking lot distribution, the positional relationship between the vehicle and the current parking lot influencing factor in the parking lot space is determined, and the trajectory influencing factor of the vehicle's driving trajectory to each parking space is calculated;

[0048] The vehicle-side influence factor and trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then candidate parking spaces are obtained based on the parking space evaluation value;

[0049] Use the preset recommendation strategy to recommend the best parking space from the candidate parking spaces.

[0050] In a fourth aspect, the present application provides a parking recommendation device, which includes: a factor detection module, a factor calculation module, a parking space acquisition module, and a parking space recommendation module, wherein:

[0051] The factor detection module is used to obtain the parking side impact factor and the vehicle side impact factor in response to the parking request;

[0052] The factor calculation module is used to use the positional relationship between the current vehicle and the influencing factors in the parking space to build a neural network model based on the parking distribution and calculate the trajectory influencing factors of the vehicle arriving at each parking space;

[0053] The parking space acquisition module is used to input the vehicle-side influence factor and trajectory influence factor into a preset parking evaluation model to calculate the parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then obtain candidate parking spaces based on the parking space evaluation value;

[0054] The parking space recommendation module is used to recommend the best parking space from the candidate parking spaces using a preset recommendation strategy.

[0055] In a fifth aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0056] In response to a parking request, obtaining a parking-side impact factor and a vehicle-side impact factor in the parking request;

[0057] By using the positional relationship between the current vehicle and the influencing factors in the parking space, a neural network model based on the parking distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space.

[0058] The vehicle-side influence factor and trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then candidate parking spaces are obtained based on the parking space evaluation value;

[0059] Use the preset recommendation strategy to recommend the best parking space from the candidate parking spaces.

[0060] In a sixth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0061] In response to a parking request, obtaining a parking-side impact factor and a vehicle-side impact factor in the parking request;

[0062] By using the positional relationship between the current vehicle and the influencing factors in the parking space, a neural network model based on the parking distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space.

[0063] The vehicle-side influence factor and trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the vehicle's required lateral space configuration and charging configuration, and then candidate parking spaces are obtained based on the parking space evaluation value;

[0064] Use the preset recommendation strategy to recommend the best parking space from the candidate parking spaces.

[0065] The above-mentioned parking recommendation method, device, computer equipment and storage medium have at least the following technical effects: due to the use of the field-side influence factor to calculate the trajectory influence factor of the vehicle's driving trajectory to each parking space, the parking space evaluation value is calculated based on the vehicle-side influence factor and the trajectory influence factor, and the lateral space configuration and charging configuration required by the vehicle are used to obtain the candidate parking space according to the parking space evaluation value. The optimal parking space is recommended based on the recommendation strategy. On the basis of conventional influencing factors such as driving trajectory distance and number of gear shifts, the lateral space configuration required by the vehicle is considered. For example, the lateral space required for opening the door on the vehicle side also needs to consider the location of the charging pile equipment, further ensuring the credibility of the optimal parking space, improving user experience, and meeting the parking and charging needs of new energy vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0067] FIG1 is a structural block diagram of a parking recommendation system in one embodiment.

[0068] FIG2 is a flow chart of a parking recommendation method according to an embodiment.

[0069] FIG3 is a schematic diagram of a neural network model in one embodiment.

[0070] FIG4 is a schematic diagram of ROI gridding in one embodiment.

[0071] FIG5 is a schematic diagram of a field-side characteristic diagram in one embodiment.

[0072] FIG6 is a flow chart of a parking recommendation method in another embodiment.

[0073] FIG7 is a flow chart of a parking recommendation method in another embodiment.

[0074] FIG8 is a structural block diagram of a parking recommendation device in one embodiment.

[0075] FIG9 is a diagram showing the internal structure of a computer device in one embodiment. Modes for Carrying Out the Invention

[0076] Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the described embodiments are merely some, and not all, of the embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the description of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0077] In the description of the present invention, unless otherwise specified or limited, the terms "disposed," "installed," and "connected" should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; and direct or indirect connections through an intermediary. Those skilled in the art will understand the specific meanings of these terms based on the specific circumstances.

[0078] The directions or positional relationships indicated by terms such as "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inside" and "outside" are based on the directions or positional relationships shown in the accompanying drawings, or are the directions or positional relationships in which the inventive product is usually placed when in use. They are only for the convenience and simplification of description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0079] The terms "first," "second," "third," etc. are merely used to distinguish between elements of similar nature and do not indicate or imply relative importance or a particular order.

[0080] The terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than the listed elements and may also include additional elements not specifically listed.

[0081] The parking recommendation method provided in this application can be applied to the parking recommendation system shown in Figure 1. The parking recommendation system includes a main control device 10, a field acquisition device 20, and a vehicle acquisition device 30. The main control device 10 is respectively connected to the field acquisition device 20 and the vehicle acquisition device 30. The field acquisition device 20 can be a vehicle-side sensing device applied to the vehicle, such as a vehicle-side camera, a vehicle-side recognition device, etc. The vehicle-side sensing device collects the distribution details of the field-side spatial area, such as the location information of parking spaces, obstacles, and charging pile equipment. After the vehicle-side sensing device identifies the field-side impact factor, the field-side impact factor is sent to the main control device. The field-side acquisition device 20 can also be a field-side host. For example, the field-side host stores the distribution details of the field-side spatial area, and the field-side impact factor is sent to the main control device through the field-side acquisition device 20. The field-side impact factors including charging pile equipment, obstacles, and parking spaces are obtained through the field-side acquisition device 20. The vehicle-side acquisition device 30 can connect to the vehicle-side cabin control device to obtain vehicle-side influencing factors, including the vehicle's remaining battery life and the required lateral clearance for door opening. The cabin control device uses in-vehicle cameras and other devices to capture driver and passenger information. Based on parameters such as weight, it calculates the required lateral clearance for passenger entry and exit at each parking space. It also outputs a signal indicating whether charging is required based on the vehicle's remaining battery life. After receiving the field-side and vehicle-side influencing factors, the main control device 10 executes the steps of the parking recommendation method described in this embodiment.

[0082] In one embodiment, as shown in FIG2 , a parking recommendation method is provided, comprising the following steps:

[0083] Step S110 : In response to a parking request, obtaining a site-side impact factor and a vehicle-side impact factor in the parking request.

[0084] Step S120 , using the positional relationship between the current vehicle and the parking lot influencing factors in the parking lot spatial area, a neural network model based on the parking lot distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space.

[0085] Step S130 , inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model, calculating a parking space evaluation value using the required lateral space configuration and charging configuration of the vehicle, and then obtaining a candidate parking space based on the parking space evaluation value;

[0086] Step S140: recommending the best parking space from the candidate parking spaces using a preset recommendation strategy.

[0087] In one embodiment, step S110, in response to a parking request, obtaining a parking-side impact factor and a vehicle-side impact factor, includes:

[0088] Step S111: Use the vehicle-side matching rules to select a factor classification table.

[0089] Step S112: Receive real-time field-side impact factors and vehicle-side impact factors according to the factor classification table.

[0090] The vehicle-side influencing factors in this embodiment may include the remaining battery power of the vehicle and driver / passenger demand information. For example, if the current vehicle is an electric vehicle, whether it needs to be charged is considered based on the remaining battery power of the vehicle. Based on the premise that the vehicle needs to be charged, the field-side influencing factors need to consider whether it is close to the charging pile equipment; for example, if the driver is a novice, the driving skills may be taken into consideration, and the vehicle-side influencing factors need to consider the number of gear changes, the distance to obstacles, and the curvature of the driving trajectory.

[0091] To further explain, the vehicle-side influencing factor can be obtained directly from the vehicle's entire vehicle data, or from the pre-input data of the occupants of the vehicle; the field-side influencing factor can be obtained from the pre-installed sensing equipment on the vehicle side, or from the field-side system.

[0092] In one embodiment, the remaining battery power of the vehicle is taken into consideration. Further explanation, step S111, the step of screening the factor classification table using the vehicle-side matching rules, includes:

[0093] Step S1111, obtaining the vehicle type, determining whether the vehicle is a tram, and outputting a first determination result.

[0094] Step S1112: If the vehicle is an electric vehicle in the first judgment result, the factor classification table includes at least vehicle-side influencing factors including the remaining power of the vehicle, and continues to judge whether the remaining power of the vehicle is lower than the preset charging threshold, and outputs the second judgment result.

[0095] Step S1113: If the remaining power of the vehicle is lower than the preset charging threshold in the second judgment result, the factor classification table includes at least the field-side influencing factors including obstacles and charging piles; otherwise, the factor classification table includes at least the field-side influencing factors including obstacles.

[0096] To further explain, this embodiment considers whether the parking space needs to be close to the charging pile based on the vehicle type and charging requirements. Based on this, this embodiment selects different factor classification tables to match the field-side impact factor and the vehicle-side impact factor. In addition to considering the charging of the vehicle, the field-side impact factor needs to include the charging pile equipment and also consider the situation where the parking space is occupied.

[0097] In one embodiment, as shown in FIG3 , step S120 , using the positional relationship between the current vehicle and the terminal influencing factor in the terminal spatial area, constructing a neural network model based on the terminal distribution includes:

[0098] As shown in Figure 4, step S121 involves gridding the field-side spatial area into ROIs to abstractly represent the field-side influencing factors. Further explanation: Because the distribution of field-side influencing factors, such as parking spaces, obstacles, and charging piles, is continuous and cannot be fully enumerated, the field-side spatial area is gridded into ROIs to abstractly represent the field-side influencing factors.

[0099] Step S122 , selecting different grid resolutions based on the distance between the vehicle in the driving trajectory and the terminal influencing factor, obtaining the correlation relationship representing the occupied grids, and constructing a neural network model of the terminal distribution.

[0100] The grid parameters in the neural network model of this embodiment can be obtained according to the training method of a conventional neural network model and deployed online.

[0101] In one embodiment, the relationship between the map and the grid is represented by: , where the id value represents the grid number, the is_occupied value represents the grid occupancy status, and the property value represents the point attribute. The id value and property value are processed by one hot encoding to obtain a {0,1} vector.

[0102] In one embodiment, in step S120, the step of calculating the trajectory influence factor of the vehicle's driving trajectory to each parking space includes:

[0103] As shown in Figure 5, step S123 utilizes the field-side impact factor and the relationship between the grids representing occupancy to obtain the vehicle's driving trajectory from the parking space, and obtains a field-side feature map along the driving trajectory. Further explanation: the field-side feature map is represented by {N, M, k}, where N represents the trajectory number, M represents the grid number, and k represents the point attribute number. In one embodiment, different point attributes can be represented by {1, ..., k}, respectively. Point attributes can include the vehicle starting point, parking space end point, parking space boundary, obstacle boundary, and road boundary. Therefore, the k value can represent the number of point attribute types.

[0104] Step S124: Input the field-side feature map into the neural network model, calculate and output the feature true value. , , , , }, Indicates the parking space occupancy status. represents the trajectory length, Indicates the number of gear shifts along the driving trajectory, Indicates the shortest distance between the vehicle and the obstacle in the driving trajectory. represents the average curvature per unit distance. If a parking space is occupied, it is unavailable for parking; if it is unoccupied, it is available for parking. Furthermore, an occupied parking space is not limited to having other vehicles parked in it; it can also be locked or occupied by other non-vehicle obstacles.

[0105] As shown in Table 1 below, a feature truth table is used to represent the feature truth value output by the neural network model.

[0106]

[0107] Table 1

[0108] Step S125 , calculating the trajectory influencing factors of the vehicle's driving trajectory to each parking space based on the feature true value and the preset cost function.

[0109] In one embodiment, step S125, the step of calculating the trajectory influencing factor of the vehicle's driving trajectory to each parking space based on the feature true value and a preset cost function, includes:

[0110] The cost function equation (1) is used to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space;

[0111] (1)

[0112] in, represents the cross entropy function, represents the weight function, Indicates the difference between the feature estimate and the true value when i=1; It means that when i=2,...,5, the power of the difference between each feature estimate and the corresponding feature true value is summed.

[0113] In one embodiment, step S130, inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the required lateral space configuration and charging configuration of the vehicle, and then obtaining a candidate parking space based on the parking space evaluation value, includes:

[0114] Step S131, inputting the required lateral space configuration and charging configuration of the vehicle into a parking space evaluation function to calculate a parking space evaluation value;

[0115] The parking space evaluation value is calculated using the parking space evaluation function equation (2);

[0116] (2)

[0117] Among them, through Indicates the parking space evaluation value; hour, , , Indicates the dimensionless calibration value; Indicates the lateral distance between the four wheels at the end position and the obstacle. Indicates the distance between the destination location and the nearest charging pile device; Indicates the lateral space required for the vehicle to open the door. Indicates whether charging is needed by (0,1). represents the weight coefficient;

[0118] Step S132: Filter candidate parking spaces based on parking space evaluation values ​​to obtain a set of candidate parking spaces. .

[0119] In one embodiment, step S140, the step of selecting the optimal parking space from the candidate parking spaces using a preset recommendation strategy, includes: selecting the optimal parking space from the candidate parking spaces using a recommendation strategy expression (3); the recommendation strategy expression (3) is: .

[0120] In one embodiment, as shown in FIG6 , a parking recommendation method is provided, which is applied to the vehicle side. The method includes the following steps:

[0121] Step S210 : In response to a parking request, obtaining a site-side impact factor in the parking request.

[0122] Step S220 , using the positional relationship of the field-side impact factors in the field-side spatial area, a neural network model based on the field-side distribution is constructed to calculate the trajectory impact factors of the vehicle arriving at each parking space.

[0123] In step S230 , the vehicle-side influence factor and the trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtain a candidate parking space based on the parking space evaluation value.

[0124] Step S240: recommending the best parking space from the candidate parking spaces using a preset recommendation strategy.

[0125] In one embodiment, as shown in FIG7 , a parking recommendation method is provided, which is applied to a parking terminal, where the parking terminal is in communication with the vehicle terminal. The method includes the following steps:

[0126] Step S310: In response to a parking request, obtaining a vehicle-side impact factor in the parking request.

[0127] Step S330 : Based on the field-side distributed neural network model, the positional relationship between the vehicle and the current field-side impact factor in the field-side spatial area is determined, and the trajectory impact factor of the vehicle's driving trajectory to each parking space is calculated.

[0128] In step S330 , the vehicle-side influence factor and the trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtain a candidate parking space based on the parking space evaluation value.

[0129] Step S340: recommending the best parking space from the candidate parking spaces using a preset recommendation strategy.

[0130] In at least one of the parking recommendation methods provided above, the trajectory influence factor of the vehicle's driving trajectory to each parking space is calculated from the field-side influence factor, and the vehicle-side influence factor is further used to calculate the candidate parking spaces. Based on the recommendation strategy, the optimal parking space is recommended. On the basis of conventional influencing factors such as driving trajectory distance and number of gear shifts, the lateral space configuration required by the vehicle is considered, such as the lateral space required for opening the door on the vehicle side, the lateral distance between the wheels and obstacles, etc. The location of the charging pile equipment also needs to be considered to further ensure the credibility of the optimal parking space, improve the user experience, and meet the parking and charging needs of new energy vehicles.

[0131] It should be understood that although the various steps in the flow charts of Figures 1-7 are shown in sequence according to the instructions of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in Figures 1-7 can include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0132] In one embodiment, as shown in FIG8 , a parking recommendation device is provided, comprising: a factor detection module 110 , a factor calculation module 120 , a parking space acquisition module 130 , and a parking space recommendation module 140 , wherein:

[0133] The factor detection module 110 is configured to obtain, in response to a parking request, a parking-side impact factor and a vehicle-side impact factor in the parking request.

[0134] The factor calculation module 120 is used to use the positional relationship between the current vehicle and the yard-side influencing factor in the yard-side spatial area to build a neural network model based on the yard-side distribution and calculate the trajectory influencing factor of the vehicle's driving trajectory to each parking space.

[0135] The parking space acquisition module 130 is used to input the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtain a candidate parking space based on the parking space evaluation value.

[0136] The parking space recommendation module 140 is configured to recommend an optimal parking space from candidate parking spaces using a preset recommendation strategy.

[0137] The specific definition of the parking recommendation device can be found in the definition of the parking recommendation method above and will not be further elaborated here. Each module in the aforementioned parking recommendation device may be implemented in whole or in part via software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in hardware form, or may be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0138] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in Figure 9. The computer device includes a processor, memory, a network interface, and a database connected via a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is configured to store parking recommendation data. The network interface of the computer device is configured to communicate with an external terminal via a network connection. When executed by the processor, the computer program implements a parking recommendation method.

[0139] Those skilled in the art will understand that the structure shown in Figure 9 is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0140] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0141] In response to a parking request, a yard-side influence factor and a vehicle-side influence factor in the parking request are obtained; a neural network model based on the yard-side distribution is constructed using the positional relationship between the current vehicle and the yard-side influence factor in the yard-side spatial area to calculate the trajectory influence factor of the vehicle's driving trajectory to each parking space; the vehicle-side influence factor and the trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then candidate parking spaces are obtained based on the parking space evaluation value; and a preset recommendation strategy is used to recommend the optimal parking space from the candidate parking spaces.

[0142] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0143] In response to a parking request, a yard-side influence factor and a vehicle-side influence factor in the parking request are obtained; a neural network model based on the yard-side distribution is constructed using the positional relationship between the current vehicle and the yard-side influence factor in the yard-side spatial area to calculate the trajectory influence factor of the vehicle's driving trajectory to each parking space; the vehicle-side influence factor and the trajectory influence factor are input into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then candidate parking spaces are obtained based on the parking space evaluation value; and a preset recommendation strategy is used to recommend the optimal parking space from the candidate parking spaces.

[0144] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0145] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A parking recommendation method, characterized in that: The method comprises: In response to a parking request, obtaining a parking-side impact factor and a vehicle-side impact factor in the parking request; By using the positional relationship between the current vehicle and the influencing factors of the terminal in the terminal space area, a neural network model based on the terminal distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory to each parking space; Inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtaining a candidate parking space according to the parking space evaluation value; The optimal parking space is recommended from the candidate parking spaces using a preset recommendation strategy.

2. The parking recommendation method according to claim 1, characterized in that: The step of obtaining the parking-side impact factor and the vehicle-side impact factor in response to the parking request includes: Use vehicle-side matching rules to filter the factor classification table; According to the factor classification table, real-time field-side influencing factors and vehicle-side influencing factors are received.

3. The parking recommendation method according to claim 2, characterized in that: The step of using the vehicle-side matching rules to filter the factor classification table includes: Obtain the vehicle type, determine whether the vehicle is a tram, and output a first determination result; If the vehicle is an electric vehicle in the first judgment result, the factor classification table includes at least vehicle-side influencing factors including the remaining power of the vehicle, and continues to judge whether the remaining power of the vehicle is lower than a preset charging threshold, and outputs a second judgment result; If in the second judgment result, the remaining power of the vehicle is lower than the preset charging threshold, the factor classification table includes at least field-side influencing factors including obstacles and charging pile equipment; otherwise, the factor classification table includes at least field-side influencing factors including obstacles.

4. The parking recommendation method according to claim 1, characterized in that: The step of constructing a neural network model based on field-side distribution by using the positional relationship between the current vehicle and the field-side influencing factor in the field-side spatial area includes: Performing ROI grid division on the field end space area to abstractly represent the field end impact factor; According to the distance between the vehicle in the driving trajectory and the field-end influencing factor, different grid resolutions are selected to obtain the correlation relationship representing the occupied grids to construct a neural network model of field-end distribution.

5. The parking recommendation method according to claim 4, characterized in that: The step of calculating the trajectory influencing factor of the vehicle's driving trajectory to each parking space includes: Using the field-side influencing factor and the correlation relationship representing the occupied grid, a driving trajectory of the vehicle to the parking space is obtained, and a field-side feature map is obtained along the driving trajectory; Inputting the field-side feature map into the neural network model, calculating and outputting the feature true value; According to the feature true value and the preset cost function, the trajectory influencing factor of the vehicle's driving trajectory to each parking space is calculated.

6. The parking recommendation method according to claim 5, characterized in that: The step of calculating the trajectory influencing factor of the vehicle's driving trajectory to each parking space according to the feature true value and the preset cost function includes: The cost function equation (1) is used to calculate the trajectory influencing factor of the vehicle's driving trajectory to each parking space; （1） in, represents the cross entropy function, represents the weight function, Indicates the difference between the feature estimate and the true value of the feature when i=1; It means that when i=2,...,5, the power of the difference between each feature estimate and the corresponding feature true value is summed.

7. The parking recommendation method according to claim 1, characterized in that: The step of inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtaining a candidate parking space according to the parking space evaluation value includes: Input the lateral space configuration and charging configuration required by the vehicle into the parking space evaluation function to calculate the parking space evaluation value; Calculate the parking space evaluation value using the parking space evaluation function equation (2); （2） Among them, through Indicates the parking space evaluation value; hour, ， ， It indicates the dimensionless calibration value; Indicates the lateral distance between the four wheels and the obstacle at the end position. Indicates the distance between the end point and the nearest charging pile device; Indicates the lateral space required for the vehicle to open its doors. Indicates whether charging is needed by (0,1). represents the weight coefficient; Filter candidate parking spaces according to the parking space evaluation value and obtain the candidate parking space set 。 8. The parking recommendation method according to claim 1, characterized in that: The step of selecting and recommending the best parking space from the candidate parking spaces by using a preset recommendation strategy includes: Using the recommendation strategy expression (3) to select the optimal parking space from the candidate parking spaces; The recommended strategy expression (3) is: 。 9. A parking recommendation method, applied to a vehicle, characterized in that: The method comprises: In response to a parking request, obtaining a field-side impact factor in the parking request; By using the positional relationship of the field-side influencing factors in the field-side spatial area, a neural network model based on field-side distribution is constructed to calculate the trajectory influencing factors of the vehicle's driving trajectory when it arrives at each parking space; Inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtaining a candidate parking space according to the parking space evaluation value; The optimal parking space is recommended from the candidate parking spaces using a preset recommendation strategy.

10. A parking recommendation method, applied to a parking terminal, wherein the parking terminal is connected to a vehicle terminal by communication, characterized in that: The method comprises: In response to a parking request, obtaining a vehicle-side impact factor in the parking request; Based on the neural network model of the terminal distribution, determine the positional relationship between the vehicle and the current terminal influencing factor in the terminal space area, and calculate the trajectory influencing factor of the vehicle's driving trajectory to each parking space; Inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model to calculate a parking space evaluation value using the lateral space configuration and charging configuration required by the vehicle, and then obtaining a candidate parking space according to the parking space evaluation value; The optimal parking space is recommended from the candidate parking spaces using a preset recommendation strategy.

11. A parking recommendation device, characterized in that: The device comprises: A factor detection module, configured to obtain, in response to a parking request, a field-side influencing factor and a vehicle-side influencing factor in the parking request; A factor calculation module is used to construct a neural network model based on the field distribution by using the position relationship between the current vehicle and the field influence factor in the field space area, and calculate the trajectory influence factor of the vehicle arriving at each parking space; A parking space acquisition module, used for inputting the vehicle-side influence factor and the trajectory influence factor into a preset parking evaluation model, so as to calculate a parking space evaluation value by using the lateral space configuration and charging configuration required by the vehicle, and then acquiring a candidate parking space according to the parking space evaluation value; The parking space recommendation module is used to recommend the best parking space from the candidate parking spaces using a preset recommendation strategy.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images