Parking recommendation method and device, computer equipment and storage medium

Abstract

The invention provides a parking recommendation method and device, computer equipment and a storage medium. The method comprises the steps of obtaining real-time parking space data and historical parking data of at least one parking lot; determining the current busy level of the parking lot based on the real-time parking space data and the historical parking data; determining a pricing result of each type of parking service based on the current busy level and the current demand intensity corresponding to each type of parking service; and generating parking recommendation information based on the pricing results of the various parking services, and providing the parking recommendation information to the user waiting for parking. The parking lot busy level is determined by integrating the real-time parking space data and the historical parking data, and the operation state of the parking lot can be effectively reflected through the busy level, so that pricing is more flexibly performed on each type of parking service according to the busy level and the current demand intensity, the recommendation dimension of the recommendation information is enriched according to the pricing result, and the user experience is improved. And the recommendation accuracy of parking recommendation is effectively improved.

Description

Technical Field

[0001] This application relates to the field of parking space management, specifically to a parking recommendation method, device, computer equipment, and storage medium. Background Technology

[0002] With the continuous increase in the number of motor vehicles in cities, parking demand is rising rapidly, and users generally face the problem of parking difficulties during their travels. Existing parking recommendation methods mainly rely on basic information provided by parking management systems or third-party parking applications, such as parking lot location, number of remaining parking spaces, and uniform hourly pricing standards. When using these systems, users can usually only view a single piece of information, such as the number of available parking spaces or a fixed price, and then decide whether to park there on their own.

[0003] However, existing technologies still have significant shortcomings: on the one hand, relying solely on the real-time number of remaining parking spaces cannot reflect the overall operational status of the parking lot, making it difficult for users to predict the convenience of parking; on the other hand, existing charging methods are generally fixed and lack flexibility; furthermore, recommended information is relatively limited, only focusing on the number of parking spaces and their location, failing to provide more targeted service options. These problems make it difficult for users to choose parking lots in a timely and reasonable manner, and also limit the utilization efficiency of parking lot resources.

[0004] Therefore, the accuracy of existing parking recommendation technologies is relatively low, and there is an urgent need to provide a parking recommendation method that can effectively improve the accuracy of parking recommendations. Summary of the Invention

[0005] This application provides a parking recommendation method, apparatus, computer device, and storage medium to solve the problem of low recommendation accuracy in traditional parking recommendation methods.

[0006] According to a first aspect of this application, one embodiment provides a parking recommendation method, comprising: Obtain real-time parking space data and historical parking data for at least one parking lot. Each parking lot has at least one type of parking service, and each type of parking service has a corresponding current demand intensity. Based on real-time parking space data and historical parking data, the current busy level of the parking lot is determined; The pricing for each type of parking service is determined based on the current level of busyness and the current demand intensity corresponding to each type of parking service. Based on the pricing results of various parking services, parking recommendation information is generated and provided to users waiting for parking.

[0007] Optionally, based on real-time parking space data and historical parking data, the current busy level of the parking lot can be determined, including: Based on real-time parking space data, calculate the current number of available parking spaces in the parking lot; Calculate the current vacancy rate of the parking lot based on the current number of available parking spaces; Based on historical parking data, the current period traffic flow and historical period traffic flow of the parking lot are determined. The historical period traffic flow corresponds to the historical vacancy rate within the historical period. Based on the current vacancy rate, current period traffic flow, historical period traffic flow, and historical vacancy rate, determine the current busy level of the parking lot.

[0008] Optionally, based on the current busy level of the parking lot and the current demand intensity corresponding to each type of parking service, the pricing result for each type of parking service is determined, including: If the parking service has a preset promotional strategy factor, the pricing result of the parking service is calculated based on the promotional strategy factor, the current busy level, and the current demand intensity. If the parking service does not have a pre-set promotional strategy factor, the pricing result of the parking service will be calculated based on the current busy level and the current demand intensity.

[0009] Optionally, based on the pricing results of various parking services, parking recommendation information is generated, including: Input real-time parking space data and historical parking data into a preset parking space demand prediction model so that the parking space demand prediction model can output parking space demand trends. Based on parking demand trends and busy levels, determine the number of available parking spaces for each type of parking service; Based on the number of available parking spaces and pricing results, parking recommendations are generated.

[0010] Optionally, based on the number of available parking spaces and pricing results, parking recommendation information is generated, including: When there are multiple parking lots, obtain the parking requests of users waiting to park. The parking requests include the current location, destination location, and estimated parking duration of the users waiting to park. Based on the current busy level and parking requests, select at least one recommended parking lot from multiple parking lots; Based on the number of available parking spaces and pricing results of at least one recommendable parking lot, parking recommendation information is generated.

[0011] Optionally, based on the number of available parking spaces in at least one recommendable parking lot and the pricing results, parking recommendation information is generated, including: Extract the parking preferences of users waiting to park from the pre-set user profile database; Based on parking preferences, available parking spaces, and pricing results, calculate a recommendation score for each type of parking service in each recommendable parking lot; Based on the recommendation score, each type of parking service in each recommendable parking lot is ranked to obtain a recommendation ranking table, which is then used as parking recommendation information.

[0012] Optionally, after providing parking recommendation information to users waiting for parking, the method further includes: Obtain the target parking lot and the target parking service type selected by the user waiting to park based on the parking recommendation information; Based on the target parking lot and the target parking service type, a corresponding parking order is generated and provided to the user waiting to park, so that the user can complete the payment for the parking order; After a user waiting to park completes payment for their parking order, an entry pass for the target parking lot is generated for that user.

[0013] According to a second aspect of this application, one embodiment provides a parking recommendation device, comprising: The first acquisition module is used to acquire real-time parking space data and historical parking data of at least one parking lot. Each parking lot has at least one type of parking service, and each type of parking service has a corresponding current demand intensity. The first determination module is used to determine the current busy level of the parking lot based on real-time parking space data and historical parking data. The second determining module is used to determine the pricing result for each type of parking service based on the current busy level and current demand intensity. The first recommendation module is used to generate parking recommendation information based on the pricing results of various parking services and provide the parking recommendation information to users waiting for parking.

[0014] According to a third aspect of this application, one embodiment provides a computer device including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer-readable instructions to implement the above-described parking recommendation method.

[0015] According to a fourth aspect of this application, one embodiment provides a readable storage medium having computer-readable instructions stored thereon, which, when executed by a processor, implement a parking recommendation method.

[0016] The parking recommendation method according to the above embodiments determines the parking lot busy level by comprehensively considering real-time parking space data and historical parking data. The busy level can effectively reflect the parking lot's operational status. Based on the busy level and the current demand intensity of each parking service, pricing for each type of parking service can be set more flexibly. Furthermore, the recommendation dimensions of the recommendation information are enriched based on the pricing results, effectively improving the accuracy of parking recommendations. This enables users to choose parking lots in a timely and reasonable manner, and also further increases the utilization rate of parking lot resources. Attached Figure Description

[0017] Figure 1This is one of the flowcharts of a parking recommendation method in one embodiment of the present invention; Figure 2 This is a second schematic flowchart of a parking recommendation method in one embodiment of the present invention; Figure 3 This is the third flowchart of a parking recommendation method in one embodiment of the present invention; Figure 4 This is the fourth flowchart of a parking recommendation method in one embodiment of the present invention; Figure 5 This is the fifth flowchart of a parking recommendation method in one embodiment of the present invention; Figure 6 This is a sixth flowchart illustrating the parking recommendation method in one embodiment of the present invention; Figure 7 This is the seventh flowchart of a parking recommendation method in one embodiment of the present invention; Figure 8 This is a schematic diagram of a parking recommendation system according to an embodiment of the present invention; Figure 9 This is a schematic diagram of a parking recommendation device in one embodiment of the present invention; Figure 10 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0018] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0019] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0020] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0021] In one embodiment, such as Figure 1 As shown, a parking recommendation method is provided, including the following steps: 101. Obtain real-time parking space data and historical parking data for at least one parking lot.

[0022] In this embodiment of the invention, the above-mentioned parking lot corresponds to at least one type of parking service, and each type of parking service corresponds to the current demand intensity.

[0023] Real-time parking data indicates the current parking space occupancy status and its timestamp, including at least the parking space ID, occupancy / vacancy information, total number of parking spaces, and available parking spaces. Historical parking data can be time-series statistics generated within a preset historical period, including at least historical vacancy rate, historical traffic flow distribution, and occupancy duration. Current demand intensity is a quantitative indicator of the relative demand level for a specific parking service during the current time period, and can be set based on factors such as current user query popularity for the area or type of service, and surrounding real-time events.

[0024] The aforementioned parking services can be parking right categories defined based on time boundaries or periodic rules. These parking services, categorized as such, may include, but are not limited to: For example, daily parking service allows users to park at any time within 24 hours after purchasing the service; Daytime parking service, such as weekday parking rights from 08:00 to 18:00 daily; Overnight parking services, such as overnight parking rights from 7:00 PM to 7:00 AM the following day; Monthly parking services, such as parking rights for 30 consecutive days from the date of purchase; Monthly daytime parking service, such as the right to park during the daytime hours every day for a consecutive month; Monthly nighttime parking service, such as the right to park every night for a consecutive month; Other customized time-limited services, such as parking rights available only on weekends, short-term packages for a certain number of consecutive hours, can be set according to actual needs.

[0025] For example, parking operators or system administrators can define and configure different parking service types. Each service type includes attributes such as: name, applicable time period rules (e.g., weekdays 08:00-18:00), base price (as a reference for dynamic pricing), applicable vehicle types (optional), and usage rules (e.g., overtime surcharge rules). These configurations are then stored in a database.

[0026] Specifically, the data acquisition and processing subsystem can obtain the aforementioned parking space data, historical parking data, and current demand intensity data by pulling or pushing data through the parking space status sensing device interface, vehicle access management device interface, and parking lot management system interface, at a preset sampling granularity, and completing basic deduplication, alignment, and integrity verification.

[0027] 102. Based on real-time parking space data and historical parking data, determine the current busy level of the parking lot.

[0028] In this embodiment of the invention, the current busy level can be a classification identifier for the occupancy status of the parking lot during the current time period, which can represent at least one of the following states: idle, moderate, busy, near saturation, or saturation.

[0029] Specifically, real-time parking data can be calculated to obtain the current number of available parking spaces and the current vacancy rate; secondly, comparative data corresponding to the current time period can be extracted from historical parking data (such as the average vacancy rate and traffic flow on the same day and time window last week) to form benchmark indicators. Then, based on preset grading rules or models, a comprehensive judgment is made between the current indicator and the benchmark indicator to output the busy level. Grading rules can employ threshold methods or learning model methods. In the threshold method, the vacancy rate is the primary factor, and the traffic flow is the secondary factor when setting segmented thresholds. For example, a vacancy rate ≥ 40% is considered idle, 20% ≤ vacancy rate < 40% is considered moderate, 10% ≤ vacancy rate < 20% is considered busy, 5% ≤ vacancy rate < 10% is considered close to saturation, and a vacancy rate < 5% is considered saturation. The method can also be combined with the deviation between the current inbound and outbound traffic flow and the historical same period (e.g., if the current vacancy rate is lower than the percentile or average by a certain margin, the threshold can be raised by one level). In the learning model method, vacancy rate, recent traffic flow change rate, and historical statistics are used as inputs, and a trained classification model is used to output level labels. In implementation, the state assessment unit only needs to align, calculate, and compare real-time indicators with historical benchmarks at a preset time granularity, and provide a first-level result or a smoothed stable level result within a single sampling period. For example, when the current vacancy rate is 15%, and the number of vehicles entering the area in the past 10 minutes is significantly higher than the historical average and shows a clear downward trend, the busy level can be determined as "busy" or "near saturation" based on thresholds and deviation correction rules; when the current vacancy rate is 45% and the traffic flow is lower than the historical average, it is determined as "idle".

[0030] 103. Based on the current busy level and the current demand intensity corresponding to each type of parking service, determine the pricing result for each type of parking service.

[0031] In this embodiment of the invention, the pricing result may refer to the real-time sales price calculated for each type of parking service in the current time period, which can be used to replace the single fixed price model.

[0032] Specifically, based on a preset pricing model or pricing rules, the above input parameters are substituted into the calculation to obtain the pricing result for each type of parking service. The pricing model can be a product model based on factor adjustments, for example, using the base price of this type of parking service as a benchmark, and then combining it with busy level adjustment factors and demand intensity adjustment factors for weighted correction to form a dynamic floating price; The busy level adjustment factor can increase the price as the parking lot becomes busier. For example, the price factor is 0.8 at the idle level, 1.0 at the moderate level, 1.2 at the busy level, 1.5 at the near-saturation level, and 2.0 at the saturation level. The demand intensity adjustment factor can be set based on the user's query popularity for this type of service or the recent purchase conversion rate. For example, when the demand intensity is high, the price factor is greater than 1, and when the demand intensity is low, the price factor is less than 1.

[0033] If a certain type of parking service is configured with a preset promotional strategy factor, then the promotional strategy factor is further introduced into the above calculation to apply discounts or bonuses in order to achieve preferential promotion or inventory clearance; if there is no promotional strategy factor, then only adjustments can be made based on the busy level and demand intensity.

[0034] For example, if a parking lot is currently at a moderate busy level, the current demand intensity for a certain type of parking service is higher than the benchmark, and the service has a promotion factor of 0.9, then the base price can be multiplied by a busy level adjustment factor of 1.0, a demand intensity adjustment factor of 1.2, and a promotion factor of 0.9, resulting in a final price that is approximately the base price multiplied by 1.08. This method allows parking service prices to dynamically reflect the real-time status of the parking lot and user demand, improving the flexibility and rationality of resource allocation.

[0035] 104. Based on the pricing results of various parking services, generate parking recommendation information and provide it to users waiting for parking.

[0036] In this embodiment of the invention, parking recommendation information may refer to the display data after the structured aggregation of parking lots and their available parking services, including at least the parking lot identifier, service type name and applicable time period rules, corresponding pricing results (real-time sales price), effective time period and validity period, number of available parking spaces, and necessary usage restrictions.

[0037] Specifically, real-time parking data and historical parking data can be input into a preset parking demand prediction model to obtain parking demand trends, and the number of available parking spaces for each type of service can be calculated based on the demand trends and busy levels.

[0038] Then, factors such as "price, number of available parking spaces, busy level, user distance or preference (if already acquired)" are scored or sorted to form a user-oriented recommendation order and display items.

[0039] Finally, the recommendation list containing the above fields is returned or pushed to the user through the user terminal interface for display. The above generation can refer to the aggregation, sorting, and structured encapsulation of elements such as price, rules, and quantity; the above provision can refer to presenting or notifying the user on the terminal page through an application interface or message channel. For example, the recommendation information may include: "Parking Lot A - Daytime Service (08:00–18:00), ¥X, currently available 50 parking spaces, valid today; Parking Lot B - Nighttime Service (19:00–07:00 the next day), ¥Y, currently available 30 parking spaces," and after the user clicks on an item, the detailed rules for the corresponding service and the next order placement entry are returned.

[0040] In this embodiment of the invention, real-time parking space data and historical parking data of at least one parking lot are acquired. Each parking lot corresponds to at least one type of parking service, and each type of parking service corresponds to a current demand intensity. Based on the real-time parking space data and historical parking data, the current busy level of the parking lot is determined. Based on the current busy level and the current demand intensity, the pricing result for each type of parking service is determined. Based on the pricing results of each type of parking service, parking recommendation information is generated and provided to users waiting for parking. By comprehensively determining the parking lot busy level through real-time parking space data and historical parking data, the busy level can effectively reflect the operational status of the parking lot. Therefore, pricing for each type of parking service can be more flexible based on the busy level and current demand intensity. Furthermore, the recommendation dimensions of the recommendation information are enriched based on the pricing results, effectively improving the accuracy of parking recommendations. This allows users to choose parking lots in a timely and reasonable manner, while also further increasing the utilization rate of parking resources.

[0041] It is understood that in the specific implementation of this application, data related to real-time parking space data, historical parking data, parking requests, parking preferences, etc. are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required. Furthermore, the collection, use and processing of related data, as well as the construction, use and processing of user profile databases and parking space demand prediction models, must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0042] Reference Figure 2 , Figure 2 This is a second flowchart illustrating a parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, the current busy level of the parking lot is determined based on real-time parking space data and historical parking data, including: Step 201: Calculate the current number of available parking spaces in the parking lot based on real-time parking space data; Step 202: Calculate the current vacancy rate of the parking lot based on the current number of available parking spaces; Step 203: Based on historical parking data, determine the current period traffic flow and historical period traffic flow of the parking lot. The historical period traffic flow corresponds to the historical vacancy rate within the historical period. Step 204: Determine the current busy level of the parking lot based on the current vacancy rate, current period traffic flow, historical period traffic flow, and historical vacancy rate.

[0043] In this embodiment of the invention, the calculation process of the current vacancy rate can be as follows: current available parking spaces = total number of parking spaces - current occupied parking spaces; current vacancy rate = real-time available parking spaces / total number of parking spaces.

[0044] The current busy level can be determined by comprehensively comparing real-time vacancy rate (i.e., current vacancy rate), vehicle flow rate per unit time (i.e., current period traffic flow), and historical data from the same period. Real-time vacancy rate is calculated from real-time parking space data collected by parking space sensors, directly reflecting the parking lot's current vacancy level. Vehicle flow rate per unit time is calculated based on entry and exit records collected by license plate recognition (LPR) cameras and the barrier gate control system, counting the number of entries and exits within a preset time window to obtain traffic flow trends. Historical data from the same period can be extracted from a historical parking database; for example, the average vacancy rate and average traffic flow of the same time period on the same day last week can be selected as a comparison benchmark to reflect the typical operating level of the parking lot under similar conditions.

[0045] Based on the above indicators, the level of parking lot occupancy can be divided into several levels, such as idle, moderate, busy, near saturation, or saturation.

[0046] In one implementation, a preset rule engine can be used to perform weighted calculations and threshold judgments on various indicators. For example, when the real-time vacancy rate is low and the number of vehicles entering per unit time is higher than the same period in history, it can be determined as busy or close to saturation; when the real-time vacancy rate is high and the number of vehicles entering and leaving is lower than the same period in history, it can be determined as idle or moderate.

[0047] In another implementation, a pre-trained classification model can be used for automatic judgment. For example, based on decision trees, support vector machines, or other machine learning models, the model can use vacancy rate, traffic flow, and historical differences as input features to directly output the corresponding busyness level label. Both rule-based and model-based methods can achieve dynamic assessment of parking lot busyness, providing a basis for subsequent pricing and recommendations.

[0048] Reference Figure 3 , Figure 3 This is the third flowchart of a parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, based on the current busy level of the parking lot and the current demand intensity corresponding to each type of parking service, the pricing result for each type of parking service is determined, including: Step 301: If the parking service has a preset promotional strategy factor, then calculate the pricing result of the parking service based on the promotional strategy factor, the current busy level, and the current demand intensity. Step 302: If the parking service does not have a preset promotional strategy factor, the pricing result of the parking service is calculated based on the current busy level and the current demand intensity.

[0049] In this embodiment of the invention, specifically, the base price of the parking service is used as a benchmark, and a busyness adjustment function, a demand intensity function, and an optional promotional strategy factor are introduced to dynamically adjust the price: when the parking service is configured with a preset promotional strategy factor, the pricing result is jointly affected by the base price, busyness level, and demand intensity, and then the promotional strategy factor is added for discounts or bonuses; when no promotional strategy factor is configured, the price is adjusted only based on the busyness level and demand intensity to obtain the real-time sales price for the current time period.

[0050] The above pricing relationship can be specifically expressed as: Pricing result = Base price × f_busy (busyness level) × f_demand (demand intensity) × f_promo (promotion strategy factor) Where f_busy is an adjustment function that monotonically changes with the business of the parking lot. The busier the parking lot, the larger the function value. It is preferred to take a value in the range of [0.6, 1.8]. f_demand is an adjustment function that varies with the strength of current demand for this type of parking service. It can be calculated by normalizing observations such as query popularity, recent conversion rate, or sell-out speed. The stronger the demand, the larger the function value. f_promo is a promotional strategy factor. When a promotional activity is active, it is less than or greater than 1 to achieve a discount or price increase, respectively. When no promotion is configured, it takes the value of 1. To avoid abnormal price fluctuations, the input indicators can be denoised and smoothed before calculation, and price upper and lower limits and minimum change step size constraints can be applied after calculation to ensure that the price always stays within a preset reasonable range.

[0051] For example, the base price for "daytime parking service" at a certain parking lot is 100 yuan. The current busy level is "busy," corresponding to f_busy = 1.2, and the current demand intensity is high, corresponding to f_demand = 1.1. (1) With the promotion factor f_promo = 0.9, the real-time price is approximately 100 × 1.2 × 1.1 × 0.9 = 118.8 yuan; (2) Without any promotions, the real-time price is approximately 100 × 1.2 × 1.1 = 132 yuan.

[0052] After processing with range constraints and rounding rules, the pricing result is obtained for external display and subsequent order placement.

[0053] Reference Figure 4 , Figure 4 This is the fourth flowchart of a parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, parking recommendation information is generated based on the pricing results of various parking services, including: Step 401: Input real-time parking space data and historical parking data into the preset parking space demand prediction model so that the parking space demand prediction model outputs the parking space demand trend. Step 402: Based on parking demand trends and busy levels, determine the number of available parking spaces for each type of parking service; Step 403: Generate parking recommendation information based on the number of available parking spaces and pricing results.

[0054] In this embodiment of the invention, the parking space demand trend can be a quantitative prediction of the changes in parking space occupancy / vacancy in the short term; the number of available parking spaces can refer to the number of parking spaces that can be sold to the public for a certain parking service in the current period, reflecting the upper limit of available inventory and risk control constraints.

[0055] When acquiring parking space demand trends, real-time parking space data is aligned with historical parking data and input into a pre-defined parking space demand prediction model. The output is a time series of occupancy rates or available parking spaces for the next 1–4 hours. The prediction model can be a time series-based ARIMA / Prophet model or a recurrent neural network (RNN / LSTM) model. Preferred input features include current vacancy rate, near-window traffic flow, historical mean / quantiles for the same period, and daily / weekly rhythm indicators. To ensure robustness, missing data completion and anomalous truncation can be performed on the original sequence, interval confidence band constraints can be applied to the model output, and model parameters can be periodically reassessed using a rolling window approach.

[0056] When determining the number of available parking spaces, the available sales quota for each service type in the parking lot is calculated based on the forecast results and the current busy level, combined with the time span and historical sales rate of each service type. This can be formally expressed as follows: Available parking spaces = g(current available parking spaces, predicted available parking spaces, service type duration, historical sales rate, target inventory level).

[0057] The function `g` can be implemented as a rule engine or a learning-based allocator: when the parking lot is currently idle and maintains a high level of availability during the forecast period, higher available quotas are allocated to long-term services (such as daily or monthly daytime / nighttime); when the parking lot is busy or near saturation, the quotas for long-term services are reduced, prioritizing short-term services or services with higher current turnover rates. To avoid over-commitment, it is preferable to set a total inventory limit (not exceeding "current availability + forecast safety margin"), a single service limit (maximum percentage by service type), and a safety buffer (minimum reserved parking spaces). For services spanning multiple time periods, the time window occupied by the service is matched hourly with the forecast curve to ensure that over-allocation does not occur after overlapping across time periods. If necessary, lag and minimum adjustment step size can be introduced to reduce display fluctuations caused by frequent adjustments.

[0058] When generating parking recommendation information, the pricing result and the number of available parking spaces are used as the core fields. The system aggregates parking lot identification, service type name and applicable time period rules, effective and validity period, usage restrictions and other information to form structured recommendation items. The recommendations are then sorted according to distance, price attractiveness, available quota, busy level and user preferences and returned to the user terminal as parking recommendation information.

[0059] To ensure consistency between price and inventory, prices and quotas can be synchronously refreshed and cached within a fixed calculation granularity (e.g., 1–5 minutes), and atomic control of price upper and lower limits, minimum change step size, and inventory deduction can be applied externally.

[0060] For example, when forecasts indicate ample availability for the next two hours and the current parking level is vacant, the "Daily Parking Service" can secure a higher number of available parking spaces and more attractive pricing. If forecasts indicate an impending surge and the current parking level is busy, the "3-Hour Short-Term Service" will be prioritized, with a slightly increased number of available parking spaces, while the long-term service will have its availability reduced or temporarily unavailable.

[0061] Through the above process, recommendations that fit the supply and demand situation can be dynamically output while ensuring that the products are marketable and executable.

[0062] Reference Figure 5 , Figure 5 This is the fifth flowchart of a parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, parking recommendation information is generated based on the number of available parking spaces and pricing results, including: Step 501: When there are multiple parking lots, obtain the parking requests of users waiting to park. The parking requests include the current location, destination location, and estimated parking duration of the users waiting to park. Step 502: Based on the current busy level and parking requests, select at least one recommended parking lot from multiple parking lots; Step 503: Generate parking recommendation information based on the number of available parking spaces in at least one recommendable parking lot and the pricing results.

[0063] In this embodiment of the invention, the parking request may also include preference information such as expected arrival time, price preference, and whether to prioritize seamless passage, which is reported by the user terminal and forwarded to the recommendation engine for processing via the interface service.

[0064] Upon receiving a parking request, candidate parking lots are recalled and initially screened based on the current traffic level. On the spatial side, spatial indexing is preferred for proximity retrieval, such as indexing parking lot coordinates using GeoHash or R-Tree, and filtering parking lots within a reasonable walking distance or driving detour cost based on "destination radius / path offset cost". On the status side, parking lot status assessment results prioritize currently available or moderately available parking lots, and if necessary, directly filter or reduce the weight of parking lots that are close to or saturated. On the demand side, matching can be done based on the expected parking duration and service type time boundaries; for example, when the expected parking duration is 2–3 hours, parking lots with higher available quotas for short-term or daytime services are prioritized. If users have historical preferences or frequently visited parking lot records, these types of parking lots can be added to the candidate set as recommended parking lots.

[0065] Then, using the pricing results and available parking spaces of the recommended parking lots as core fields, the system aggregates parking lot identifiers, service type names and applicable time period rules, effective and validity periods, usage restrictions, etc., to form structured recommendation items. These items are then sorted according to distance, price attractiveness, available quota, busy level, and user preferences, and returned to the user's terminal as parking recommendation information.

[0066] Reference Figure 6 , Figure 6 This is a flowchart of the parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, parking recommendation information is generated based on the number of available parking spaces in at least one recommendable parking lot and the pricing result, including: Step 601: Extract the parking preferences of users waiting to park from the preset user profile database; Step 602: Based on parking preferences, available parking spaces, and pricing results, calculate the recommendation score for each type of parking service in each recommendable parking lot; Step 603: Based on the recommendation score, sort each type of parking service in each recommendable parking lot to obtain a recommendation ranking table, and use the recommendation ranking table as parking recommendation information.

[0067] In this embodiment of the invention, the user profile database stores structured information related to users' historical behavior; parking preferences may include the user's frequently selected service type (e.g., daytime / nighttime / daily), price sensitivity (e.g., more sensitive to discounts or more sensitive to timeliness), acceptable walking distance or detour cost, preference for license plate-based contactless access, and historically frequented parking lots, etc. When there is insufficient historical data, the system's default preferences or statistical preferences based on similar users can be used as initial values.

[0068] After obtaining parking preferences, a recommendation score can be calculated for each type of parking service in each recommendable parking lot based on parking preferences, available parking spaces, and pricing results. The recommendation score can be used to characterize the degree of matching between the "parking lot - service type" candidate items and user needs. Its calculation can adopt a weighted normalization method: mapping key elements related to user experience to feature values ​​in the range [0,1], and then summing them according to their weights.

[0069] Features that can be included include, but are not limited to: Price attractiveness (based on the discount of the pricing result relative to the base price or regional average, the larger the discount, the higher the score), availability of parking spaces (the higher the available space, the higher the score; a saturation mapping is preferred to avoid extreme values ​​dominating), time matching (the overlap between the user's expected arrival time and expected parking duration and the applicable service time period; the more overlap, the higher the score; if the overlap is insufficient, the weight is reduced or eliminated), distance / walking time (the closer the distance or the shorter the walking distance, the higher the score), busy level penalty (moderately reduce the weight when busy or close to saturation, and do not reduce the weight or slightly increase the weight when idle or moderate), preference matching (the degree of overlap between candidate items and the user's historical preferences, such as service type / frequented parking lots / seamless access support, etc.).

[0070] Weights can be preset by expert rules or adaptively updated based on user feedback; to ensure robustness, upper and lower limits and segmentation can be applied to sensitive features such as price, distance, and inventory to avoid abnormal scores caused by a single factor.

[0071] Several business thresholds and constraints can be set in the rating calculation. For example, when the number of available parking spaces for an item is lower than the minimum available threshold, it can be directly set as unrecommended or significantly downgraded; when the time matching degree is lower than a preset threshold, it can be set as unrecommended to avoid users being unable to use the service after placing an order due to time period incompatibility; when the pricing result exceeds the user's price limit (if configured), it can be set as unrecommended or downgraded. To reduce recommendation jitter, an exponential smoothing or minimum step size strategy can be used for ratings within adjacent calculation periods, triggering a significant change in ranking position only when the score change exceeds a threshold.

[0072] After calculating the recommendation score, each parking service type in each recommendable parking lot is ranked based on the score, resulting in a recommendation ranking table. This ranking table serves as the parking recommendation information. The recommendation ranking table can output the first few entries by the "parking lot - service type" dimension, along with necessary fields for display, including parking lot identifier, service type name and applicable time period rules, pricing result, current available parking spaces, estimated walking time or distance, validity period description, and entry method prompts.

[0073] To ensure a good user experience, ratings and sorting can be synchronously refreshed and briefly cached at fixed time intervals (e.g., 1–5 minutes); and instant updates can be triggered when there are significant changes in inventory or price. For items with the same or similar ratings, a tie-breaking rule can be adopted, prioritizing proximity, number of available parking spaces, and lower price; when a parking lot has insufficient parking spaces during the user's expected arrival time, it can be automatically downgraded in display or replaced with a similar service from an alternative parking lot.

[0074] Through the above processing, stable and actionable parking recommendation information can be generated, taking into account user preferences, inventory availability, and real-time prices.

[0075] Reference Figure 7 , Figure 7 This is the seventh flowchart of a parking recommendation method in one embodiment of the present invention; in one embodiment of the present invention, after providing parking recommendation information to users waiting for parking, the method further includes: Step 701: Obtain the target parking lot and the target parking service type selected by the user waiting to park based on the parking recommendation information; Step 702: Based on the target parking lot and the target parking service type, generate the corresponding parking order and provide the parking order to the user waiting to park so that the user can complete the payment for the parking order; Step 703: After the user waiting to park completes the payment for the parking order, generate an entry voucher for the target parking lot for the user waiting to park.

[0076] In this embodiment of the invention, the user's selection can be completed through various terminals, such as a smartphone app, an in-vehicle central control system, or a touch-screen query terminal deployed in public places. The user can browse the parking lot's geographical location, the real-time number of available parking spaces, and recommended parking service options (including service type, price, applicable time period, and usage rules) on the interface, and make a selection based on their personal travel needs. During this process, a map service interface can be called to dynamically mark the parking lot location on a map or navigation view, and service availability and price details can be displayed on the details page.

[0077] After the user selects the target parking lot and the target parking service type, a corresponding parking order is generated based on these criteria. Order information includes, but is not limited to: the target parking lot identifier, the selected service type and applicable time period, the user's vehicle information (such as license plate number), the estimated effective time, the order amount, and the payment method. The generated parking order is displayed on the user's terminal confirmation interface, where the user can verify the information and submit a payment request. The payment process is executed by an integrated payment processing module, for example, by calling third-party payment interfaces (such as Alipay SDK, WeChat Pay SDK, UnionPay payment interface), using an encrypted channel to ensure secure order payment. Upon successful payment, the payment result is fed back to the server in real time and simultaneously returned to the user's terminal.

[0078] After a user completes payment for a parking order, an entry pass for the target parking lot is generated. The "entry pass" can take several forms: one is a QR code with dynamic anti-counterfeiting features generated via the app, which the user can scan at the parking lot gate to enter; the other is to bind the paid parking service to the user's license plate, and use license plate recognition equipment to automatically confirm the validity of the order and control the gate to open when the vehicle enters.

[0079] To further enhance the user experience, a one-click navigation function can be provided, allowing users to generate route guidance to the target parking lot by calling the terminal or in-vehicle navigation module. Simultaneously, the backend order and voucher management module will store order status, validity period, and voucher information in the database for subsequent verification and reconciliation. After parking, users can be guided to rate their parking experience through the App's feedback and evaluation interface, and the evaluation data will be sent back to the database for subsequent service optimization and recommendation model updates.

[0080] Through the above steps, a complete closed loop between parking recommendation information and actual transaction execution is achieved, ensuring seamless connection for users in the recommendation, payment and entry stages.

[0081] In one possible embodiment, the present invention also provides another parking recommendation method (not shown), the method comprising: Step 1: User Initiates Request / System Triggers Recommendation: The user enters their destination and expected parking duration through their terminal device (such as a mobile app), or the system proactively pushes recommendations based on the user's frequently used travel patterns at specific times. The user's terminal device sends the request to the backend server via the network.

[0082] Step 2: Data Acquisition and Real-time Analysis: Sensors (geomagnetic, video, etc.) and entrance / exit equipment (LPR, barrier gates) deployed in each parking lot continuously upload data to the data acquisition and processing subsystem. The parking lot status and busyness assessment subsystem analyzes this data in real time, calculates the idleness and busyness level of each parking lot, and makes short-term predictions.

[0083] Step 3: Parking Service Status and Price Updates: The parking service dynamic management and pricing subsystem, based on the latest parking lot status, invokes its internal algorithm unit to refresh the number of available parking spaces and real-time prices for each service option in each parking lot. This information is cached or updated in high-speed data storage.

[0084] Step 4: Parking lot recall and initial screening: The recall unit of the recommendation engine subsystem filters candidate parking lots from the database or cache based on user requests (e.g., within 1 kilometer, expected to park for 3 hours) and parking lot status (e.g., vacancy rate > 20%).

[0085] Step 5: Parking service option matching and personalized ranking: For each candidate parking lot, the recommendation engine's matching and ranking unit obtains its currently available service options and dynamic prices, and combines them with user historical preferences (obtained from the user profile database) to perform comprehensive scoring and ranking.

[0086] Step 6: Display of Recommended Results: The user interaction and transaction terminal and interface subsystem will return the sorted list of parking lots and the most attractive parking service options for each (e.g., "Parking lot A, daytime service ¥X, 50+ available; Parking lot B, 3-hour temporary parking ¥Y, 30+ available") to the user terminal device via API for display.

[0087] Step 7: User Selection and Payment: After comparing options on the terminal device, the user selects "Parking Lot A - Daytime Service", confirms the information, and completes the payment through the integrated payment interface.

[0088] Step 8: Service Authorization and Recording: The system records orders, and users enter the venue using electronic vouchers (such as QR codes) generated after payment or their linked license plates. User behavior (browsing, clicking, purchasing, rating) is transmitted back and recorded through user terminal devices for model optimization.

[0089] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0090] In one embodiment, a parking recommendation system is provided, such as Figure 8 As shown, the parking recommendation system includes parking space status sensing equipment, vehicle access management equipment, data acquisition and processing subsystem, parking lot status assessment subsystem, recommendation engine subsystem, dynamic management and pricing subsystem, user interaction and transaction subsystem, and user terminal equipment.

[0091] The data acquisition and processing subsystem / module is configured to acquire and process multi-source data in real time and historically through various sensing and data interface devices.

[0092] Sensor interface unit: Connects to various sensors deployed in the parking lot, such as ground sensors, to obtain real-time parking space status data (i.e., real-time parking space data, parking space ID, occupied / vacant information, total number of parking spaces, and available parking spaces).

[0093] Vehicle access management equipment interface unit: Connects to the parking lot's gate control system, license plate recognition (LPR) camera, ETC recognition equipment, etc., to obtain vehicle entry and exit data (i.e. historical parking data, recording the precise entry and exit time of vehicles and license plate information, used for analyzing traffic flow and turnover rate).

[0094] User data interface unit: Acquires user historical behavior data (i.e., user profile, such as user's historical parking choices, records of purchased parking service types, preferred time periods, etc., usually from user terminal applications or backend databases).

[0095] Parking lot basic information management unit: Stores and manages basic parking lot information data (such as parking lot geographical coordinates, charging standards, operating hours, parking space distribution map, special parking space information, etc.).

[0096] Data processing unit: including data server and preprocessing software, which cleans (removes noise and errors), aggregates (by time or region), extracts features (such as calculating real-time occupancy rate and turnover rate) and stores in structure (such as storing in time series database or relational database) the collected raw data.

[0097] Parking lot status and busyness assessment subsystem / module: Connects to the data acquisition and processing subsystem, has built-in analysis models and algorithms, and is configured for: Based on real-time parking space data, historical data, and preset models (such as time series analysis models and machine learning models), the real-time number and proportion of available parking spaces in each parking lot are dynamically evaluated on a dedicated processing unit (such as an application server or embedded processor).

[0098] Calculate and output the current busy level (i.e., the current busy level, such as: idle, moderate, busy, near saturation, saturation).

[0099] Predict the parking space demand trend and busy status in the short term (e.g., the next 1-4 hours).

[0100] The assessment results will serve as the core basis for parking lot recommendations and dynamic pricing and quantity allocation of parking service options.

[0101] Parking Service Dynamic Management and Pricing Subsystem / Module: Connects to the parking lot status and busyness assessment subsystem, configured for: Parking Service Type Definition and Management Unit: Allows operators to define multiple parking service types / options (i.e., various types of parking services), for example: Daily parking service (e.g., 24-hour parking rights).

[0102] Daytime parking service (e.g., parking rights during specific working hours, 08:00-18:00).

[0103] Nighttime parking service (e.g., parking rights during specific nighttime hours, from 19:00 to 07:00 the next day).

[0104] Monthly parking service (parking rights for a calendar month or 30 consecutive days).

[0105] Parking service is available during the daytime hours of each month.

[0106] Parking service during nighttime hours on a monthly basis.

[0107] Other customized time-based parking services (such as specific times or durations on weekends).

[0108] A dynamic parking space allocation unit can be provided: Based on the real-time number of available parking spaces, the assessment results of the busyness of the parking lot, the predicted trend, and the historical service sales data, the number of parking spaces currently available for various parking service options in different parking lots can be dynamically determined (i.e., the number of parking spaces provided in batches, or understood as the number of parking spaces mentioned above).

[0109] The dynamic pricing strategy execution unit dynamically calculates prices for various parking service options based on real-time parking lot activity levels, available parking spaces, parking service types, demand forecasts, and available operational strategies (such as promotions and inventory clearance). For example, when activity levels are low and there are many available parking spaces, prices can be appropriately discounted; when activity levels are high and there are few available parking spaces, prices can be appropriately increased or discounts reduced. The pricing model can be deployed on a central server or edge computing nodes.

[0110] Parking Lot and Parking Service Recommendation Engine Subsystem / Module: Connects to the parking lot status and busyness assessment subsystem, the parking service dynamic management and pricing subsystem, and the data acquisition and processing subsystem (for obtaining user preferences), configured for: Parking lot recall and sorting unit: Based on the user's current location (obtained via GPS on the user's terminal device), destination, historical preferences and other information (i.e. parking preferences), combined with the real-time availability and busyness of each parking lot, the unit initially filters and sorts candidate parking lots.

[0111] Parking service option matching and recommendation unit: For each recommended parking lot, it displays the currently available parking service options, detailed information about each option (such as applicable time periods and total price), and dynamically calculated real-time prices. Personalized service option recommendations can be made based on users' historical spending habits and preferred time periods.

[0112] Recommendation logic processing unit: Taking into account factors such as the distance to parking lots, availability, price attractiveness of parking service options, and user preferences, it generates the final recommendation list (i.e., parking recommendation information).

[0113] User interaction and transaction terminal and interface subsystem / module: Connects to the parking lot and parking service recommendation engine subsystem, configured for: User terminal application / interface unit: Displays a list of recommended parking lots on user devices (such as smartphone applications, in-vehicle navigation system displays, and dedicated query terminal devices), with each parking lot accompanied by an overview of its available parking spaces, the types of parking service options available, detailed information, and real-time prices.

[0114] User Selection and Transaction Processing Unit: Allows users to browse and compare different parking lots and parking service options. Supports users to select and purchase (book) specific parking services at specific parking lots, completing online payment through an integrated payment gateway (such as a payment interface module connecting to a bank or third-party payment platform).

[0115] Order Management and Voucher Generation Unit: Provides order management functions and generates entry vouchers (such as dynamic QR codes displayed on user terminals, or seamless access through license plate association).

[0116] Navigation interface unit (optional): Provides navigation services to the selected parking lot and can call the navigation application on the user terminal.

[0117] In one embodiment, a parking recommendation device is provided, which corresponds one-to-one with the parking recommendation method in the above embodiments. For example... Figure 9 As shown, the parking recommendation device includes a first acquisition module 901, a first determination module 902, a second determination module 903, and a first recommendation module 904. Detailed descriptions of each functional module are as follows: The first acquisition module 901 is used to acquire real-time parking space data and historical parking data of at least one parking lot. The parking lot corresponds to at least one type of parking service, and each type of parking service corresponds to the current demand intensity. The first determining module 902 is used to determine the current busy level of the parking lot based on the real-time parking space data and historical parking data; The second determining module 903 is used to determine the pricing result for each type of parking service based on the current busy level and the current demand intensity. The first recommendation module 904 is used to generate parking recommendation information based on the pricing results of various parking services, and to provide the parking recommendation information to users waiting for parking.

[0118] Optionally, the first determining module 902 is further configured to: Based on the real-time parking space data, calculate the current number of available parking spaces in the parking lot; Based on the current number of available parking spaces, calculate the current vacancy rate of the parking lot; Based on the historical parking data, the current period traffic flow and historical period traffic flow of the parking lot are determined, and the historical period traffic flow corresponds to the historical vacancy rate within the historical period. The current busy level is determined based on the current vacancy rate, the current period traffic flow, the historical period traffic flow, and the historical vacancy rate.

[0119] Optionally, the second determining module 903 is further configured to: If the parking service corresponds to a preset promotional strategy factor, the pricing result of the parking service is calculated based on the promotional strategy factor, the current busy level, and the current demand intensity. If the parking service does not have a pre-set promotional strategy factor, the pricing result of the parking service is calculated based on the current busy level and the current demand intensity.

[0120] Optionally, the first recommendation module 904 is further configured to: The real-time parking space data and historical parking data are input into a preset parking space demand prediction model so that the parking space demand prediction model outputs the parking space demand trend. Based on the parking demand trend and the busy level, determine the number of available parking spaces for each type of parking service; Based on the number of available parking spaces and the pricing results, the parking recommendation information is generated.

[0121] Optionally, the first recommendation module 904 is further configured to: When there are multiple parking lots, the parking request of the user waiting to park is obtained. The parking request includes the user's current location, destination location, and estimated parking duration. Based on the current busy level and the parking request, at least one recommended parking lot is selected from the multiple vehicles. The parking recommendation information is generated based on the number of available parking spaces in at least one of the recommended parking lots and the pricing results.

[0122] Optionally, the first recommendation module 904 is further configured to: Extract the parking preferences of the users waiting to park from the preset user profile database; Based on the parking preferences, the number of available parking spaces, and the pricing results, calculate a recommendation score for each type of parking service in each of the recommended parking lots; Based on the recommendation score, each type of parking service in each of the recommended parking lots is sorted to obtain a recommendation ranking table, and the recommendation ranking table is used as the parking recommendation information.

[0123] Optionally, the device further includes: The second acquisition module is used to acquire the target parking lot and the target parking service type selected by the user waiting to park based on the parking recommendation information. The second generation module is used to generate a corresponding parking order based on the target parking lot and the target parking service type, and provide the parking order to the user waiting to park so that the user waiting to park can complete the payment for the parking order; The third generation module is used to generate an entry voucher for the target parking lot for the user waiting to park after the user has completed the payment for the parking order.

[0124] Each module in the aforementioned parking recommendation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0125] In one embodiment, a computer device is provided, which may be a terminal device, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a readable storage medium storing computer-readable instructions. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer-readable instructions implement a parking recommendation method. The readable storage medium provided in this embodiment includes both non-volatile and volatile readable storage media.

[0126] In this application embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, it implements the steps of the parking recommendation method described above.

[0127] In one embodiment of the application, a readable storage medium is provided, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, they implement the steps of the parking recommendation method described above.

[0128] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0129] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A parking recommendation method characterized by comprising: The method comprises: acquiring real-time parking space data and historical parking data of at least one parking lot, the parking lot corresponding to at least one type of parking service, each type of the parking service corresponding to a current demand intensity; determining a current busy level of the parking lot based on the real-time parking space data and the historical parking data; determining a pricing result of each type of the parking service based on the current busy level and the current demand intensity corresponding to each type of the parking service; generating parking recommendation information based on the pricing result of each type of the parking service, and providing the parking recommendation information to a user to be parked.

2. The parking recommendation method according to claim 1, wherein, The determination of the current busy level of the parking lot based on the real-time parking space data and the historical parking data comprises: calculating a current number of empty parking spaces of the parking lot based on the real-time parking space data; calculating a current vacancy rate of the parking lot based on the current number of empty parking spaces; determining a current period traffic flow and a historical period traffic flow of the parking lot based on the historical parking data, the historical period traffic flow corresponding to a historical vacancy rate in a historical period; determining the current busy level of the parking lot based on the current vacancy rate, the current period traffic flow, the historical period traffic flow, and the historical vacancy rate.

3. The parking recommendation method according to claim 1, wherein The determination of the pricing result of each type of the parking service based on the current busy level and the current demand intensity corresponding to each type of the parking service comprises: if the parking service corresponds to a preset promotion strategy factor, calculating the pricing result of the parking service according to the promotion strategy factor, the current busy level, and the current demand intensity; if the parking service does not correspond to a preset promotion strategy factor, calculating the pricing result of the parking service according to the current busy level and the current demand intensity.

4. The parking recommendation method according to claim 1, wherein The generation of the parking recommendation information based on the pricing result of each type of the parking service comprises: inputting the real-time parking space data and the historical parking data into a preset parking space demand prediction model, so that the parking space demand prediction model outputs a parking space demand trend; determining a number of available parking spaces of each type of the parking service based on the parking space demand trend and the busy level; generating the parking recommendation information based on the number of available parking spaces and the pricing result.

5. The parking recommendation method according to claim 4, characterized by, The generation of the parking recommendation information based on the number of available parking spaces and the pricing result comprises: when the parking lot is multiple, acquiring a parking request of the user to be parked, the parking request comprising a current location, a destination location, and a predicted parking duration of the user to be parked; based on the current busy level and the parking request, screening at least one recommendable parking lot from the multiple parking lots; generating the parking recommendation information based on the number of available parking spaces and the pricing result of at least one of the recommendable parking lots.

6. The parking recommendation method according to claim 5, wherein The generation of the parking recommendation information based on the number of available parking spaces and the pricing result of at least one of the recommendable parking lots comprises: extracting a parking preference of the user to be parked from a preset user portrait library; Based on the parking preference, the number of available parking spaces, and the pricing result, a recommendation score of each type of parking service in each of the recommendable parking lots is calculated; Based on the recommendation score, each type of parking service in each of the recommendable parking lots is ranked to obtain a recommendation ranking table, and the recommendation ranking table is taken as the parking recommendation information.

7. The parking recommendation method according to claim 1, wherein, After the parking recommendation information is provided to the user to be parked, the method further comprises: obtaining a target parking lot and a target parking service type selected by the user to be parked according to the parking recommendation information; Based on the target parking lot and the target parking service type, a corresponding parking order is generated, and the parking order is provided to the user to be parked to enable the user to be parked to complete payment of the parking order; After the user to be parked completes the payment of the parking order, an entry voucher of the target parking lot is generated for the user to be parked.

8. A parking recommendation device characterized by comprising: Comprise: The first acquisition module is configured to acquire real-time parking space data and historical parking data of at least one parking lot, wherein the parking lot corresponds to at least one type of parking service, and each type of parking service corresponds to a current demand intensity; The first determination module is configured to determine a current busy level of the parking lot based on the real-time parking space data and the historical parking data; The second determination module is configured to determine a pricing result of each type of parking service based on the current busy level and the current demand intensity; The first recommendation module is configured to generate parking recommendation information based on the pricing result of each type of parking service, and provide the parking recommendation information to a user to be parked. 9.A computer device, comprising a memory, a processor, and computer readable instructions stored on the memory and running on the processor, wherein, The processor executes the computer readable instructions to implement the parking recommendation method of any one of claims 1 to 7.

10. A readable storage medium, having stored thereon computer readable instructions, characterized in that, The computer readable instructions are executed by the processor to implement the parking recommendation method of any one of claims 1 to 7.

Images