An intelligent parking space recommendation system for smart parks

Abstract

This invention relates to the field of parking management technology, specifically disclosing an intelligent parking space recommendation system for smart parks; comprising: a data acquisition module to acquire historical parking data of different enterprises within the park on different dates; a time period segmentation module to determine the number of parking spaces for different enterprises during peak hours based on the historical parking data of different enterprises on different dates, and to segment user parking spaces for different enterprises during peak hours based on the number of parking spaces for different enterprises, thereby determining the real-time parking space information of enterprises; a parking space assessment module to determine the number of parking spaces occupied based on the real-time parking space information of enterprises, to obtain the enterprise parking space fluctuation assessment quantity for different enterprises on different dates by statistically analyzing the historical parking space occupancy quantity, to determine the dynamics of parking space usage of enterprises based on the enterprise parking space fluctuation assessment quantity, and to generate the parking demand degree of different enterprises in the park; and a demand prediction module to predict parking demand based on the assessment results of enterprise parking space fluctuation assessment quantity and parking demand degree.

Description

Technical Field

[0001] This invention relates to the field of parking management technology, specifically to an intelligent parking space recommendation system for smart parks. Background Technology

[0002] With the rapid development of smart park construction, parking management within parks has become a key issue affecting traffic efficiency and user experience. Traditional parking management systems often rely on fixed parking space allocation or simple vacancy detection, lacking accurate responses to dynamic parking demand. Within parks, different companies have significantly different parking needs: for example, IT companies may experience peak parking times during weekdays, while catering and entertainment companies see high demand in the evenings or on weekends. Existing systems often cannot distinguish these patterns, leading to uneven distribution of parking resources, with some areas experiencing parking shortages while others remain vacant.

[0003] In addition, existing technologies typically only focus on real-time parking space status and do not make full use of historical data for trend analysis and prediction. For example, some systems use sensors to detect empty parking spaces and guide vehicles, but do not consider fluctuations in enterprise-level demand, making it difficult to allocate resources in advance. This leads to congestion during peak hours, longer vehicle detour times, and parking space conflicts between park users (such as company employees) and temporary users (such as visitors).

[0004] In addition, the existing system is relatively simple in terms of route planning and priority scheduling, and does not integrate real-time traffic factors and enterprise characteristics, resulting in low efficiency in parking recommendation. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent parking space recommendation system for smart parks, which solves the technical problem of optimizing parking resource utilization, reducing congestion, and improving the overall operational efficiency of the park by dynamically analyzing enterprise parking patterns based on historical data, predicting demand changes, and intelligently dividing the system into recommendation units. This objective can be achieved through the following technical solutions:

[0006] An intelligent parking space recommendation system for smart parks includes:

[0007] The data acquisition module is used to obtain historical parking data of different companies within the park on different dates;

[0008] The time period segmentation module is used to determine the number of parking spaces for different companies during peak hours based on historical parking data of different companies on different dates, and to segment user parking spaces for different companies during peak hours based on the number of parking spaces for different companies during peak hours, thereby determining the real-time parking space information of companies. User parking spaces include parking spaces for users in the park and parking spaces for users outside the park.

[0009] The parking space assessment module is used to determine the number of parking spaces occupied based on the real-time parking space information of enterprises. The number of parking spaces occupied includes the number of parking spaces occupied by park users and non-park users. By statistically analyzing the historical number of parking spaces occupied, the module obtains the parking space fluctuation assessment of different enterprises on different dates. Based on the enterprise parking space fluctuation assessment, the module judges the parking space usage dynamics of park users and non-park users of the enterprise and generates the parking demand of different enterprises in the park.

[0010] The demand forecasting module is used to predict parking demand based on the assessment results of enterprise parking space fluctuation and parking demand, and to divide parking space recommendation units and allocate matching parking resources to different parking space recommendation units.

[0011] Preferably, historical parking data includes the parking space occupancy rate, vehicle location information, vehicle dwell time, and vehicle type information of the enterprise at different times on different dates.

[0012] Preferably, peak-hour parking is determined based on the company's historical parking data for different dates within the most recent preset time period. This includes calculating the average parking space occupancy rate for different time periods, identifying time periods with parking space occupancy rates higher than a preset threshold as peak-hour parking, and dividing user parking space time periods based on the number of parking spaces during peak hours.

[0013] Preferably, the parking space assessment module obtains the parking space fluctuation assessment volume of different enterprises on different dates by statistically analyzing the historical parking space occupancy volume. Based on the enterprise parking space fluctuation assessment volume, it judges the parking space usage dynamics of enterprise park users and non-park users, and generates the parking demand level of different enterprises in the park. Specific methods include:

[0014] Based on the historical parking space occupancy data of different companies on different dates, the similarity of parking space occupancy among different companies under the same date type is determined, and companies with similar parking needs are screened based on the similarity of parking space occupancy.

[0015] By analyzing the fluctuations in parking space occupancy among businesses with similar parking needs over a recent preset period on different date types, the daily fluctuation assessment amount is determined for each date type. This daily fluctuation assessment amount is then used to calculate the parking demand level of businesses with similar parking needs.

[0016] Determine if the parking demand exceeds a preset demand threshold. If so, mark the enterprise as a high-demand enterprise and allocate parking resources to it first; otherwise, arrange them in order of parking demand.

[0017] Preferably, parking demand is predicted based on the enterprise's parking space fluctuation assessment and parking demand level. Specific methods include:

[0018] When the enterprise's parking space fluctuation assessment volume, i.e., the fluctuation assessment volume for several consecutive days, exceeds the preset fluctuation threshold and the parking demand is higher than the preset demand threshold, it is predicted that the enterprise will have high parking demand in the future preset period.

[0019] The division of parking space recommendation units will be dynamically adjusted based on the prediction results, and more temporary parking space resources will be allocated to enterprises with high demand.

[0020] Preferably, the method for dividing parking space recommendation units includes:

[0021] Using the parking space occupancy fluctuation threshold as a constraint, the goal is to ensure that the parking space fluctuation of enterprises in different parking space recommendation units does not exceed the fluctuation threshold, and to minimize the difference between the number of enterprises in different parking space recommendation units and the number of parking space recommendation units.

[0022] Based on the company's real-time parking information and parking demand, the company is assigned to different parking recommendation units, and matching parking resources are allocated to each unit, including adjustments to the ratio of park user parking spaces to temporary parking spaces.

[0023] Preferably, when an enterprise is assigned to a separate parking space recommendation unit, the enterprise is treated as a separate parking space recommendation entity for parking resource allocation, and the parking space recommendation strategy is adjusted in real time to adapt to the enterprise's individual needs.

[0024] Preferably, the parking space recommendation strategy includes:

[0025] The system collects real-time data on the current parking space status and vehicle trajectories of the enterprise through the 5G communication network. Based on the location information of the enterprise's recommended parking space units, the system uses a path planning algorithm to calculate the optimal parking route. The path planning algorithm includes a line resistance model, which is used to evaluate the route's turning angle, the number of shock absorbers, and real-time traffic conditions.

[0026] Based on the time the parking request is generated, the vehicle's historical violation records, and the enterprise's priority, a reservation value is calculated to generate a queue sequence. Parking spaces are then reserved and scheduled according to the reservation value, enabling collaborative parking management among different enterprises.

[0027] The beneficial effects of this invention are:

[0028] This invention analyzes historical data and forecasts demand, sequentially employing data acquisition, time-segmentation, parking space assessment, and demand forecasting modules to effectively divert and predict parking in smart parks. Specifically, by predicting future peak parking demand, it dynamically adjusts the ratio of existing user parking spaces to temporary parking spaces in advance, rationally allocating parking spaces and avoiding severe congestion caused by unreasonable parking space allocation. Furthermore, it reduces time wasted searching for parking spaces and decreases driving complexity through optimal route planning. Additionally, it establishes fair reservation rules based on reservation time, credit, and priority, making the parking process more orderly and equitable. Ultimately, it maximizes resource utilization, optimizes user experience, and refines park management.

[0029] Of course, any product implementing this invention does not necessarily need to achieve all the advantages described above at the same time. Attached Figure Description

[0030] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a schematic diagram of an intelligent parking space recommendation system module for smart parks according to the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Please see Figure 1 As shown, the present invention is an intelligent parking space recommendation system for smart parks. The system includes a data acquisition module, a time period segmentation module, a parking space evaluation module, and a demand prediction module. These modules are connected through a park network (such as 5G) and exchange data in real time.

[0034] Firstly, the data acquisition module collects historical parking data from different companies on different dates through sensors (such as geomagnetic sensors and cameras) deployed in the park. The historical parking data includes the parking space occupancy rate, vehicle location information, vehicle stay time, and vehicle type (such as employee vehicles and visitor vehicles) of different companies on different dates. For example, the module records the weekday data of Company A from June 1 to June 30, 2024: the average parking space occupancy rate is 90% from 8:00 to 9:00 am, the vehicle location is mainly concentrated in the East Parking Lot, and the stay time is mostly more than 8 hours (employee vehicles), while the occupancy rate is only 30% during the same period on weekends, and the vehicle type is mainly temporary visitors.

[0035] Secondly, the time period segmentation module calculates the peak-hour parking information for each enterprise based on historical parking data, mainly the number of vehicles parked during peak hours;

[0036] Peak-hour parking is determined based on historical parking data from different dates within a recent preset timeframe. This includes calculating the average parking occupancy rate for different time periods, identifying periods with occupancy rates exceeding a preset threshold as peak-hour parking, and then dividing user parking time slots based on the number of vehicles parked during peak hours. The peak-hour determination method involves: calculating the average parking occupancy rate for different time periods, identifying periods exceeding a preset threshold (e.g., 80%) as peak hours; and then dividing user parking time slots based on the number of vehicles parked during peak hours. This is achieved by prioritizing the allocation of park-based user parking spaces (fixed parking spaces) to company employees during peak hours, while non-park-based user parking spaces (temporary parking spaces) are used for off-peak hours or for temporary users. For example, if Company B has two peak periods in the last 30 days (8:00-9:00 AM and 5:00-6:00 PM), the system will set its parking spaces as park-based user-only during peak hours and open them as temporary parking spaces during other times.

[0037] Third, the parking space assessment module determines the number of parking spaces occupied based on real-time parking space information of enterprises, and calculates the enterprise's parking space fluctuation assessment and parking demand based on historical data. Specifically, it obtains the enterprise's parking space fluctuation assessment on different dates by statistically analyzing the historical parking space occupancy. Based on the enterprise's parking space fluctuation assessment, it judges the parking space usage dynamics of the enterprise's park users and non-park users, and generates the parking demand of different enterprises in the park. Among them, the number of parking spaces occupied includes the number of parking spaces occupied by park users and the number of parking spaces occupied by non-park users.

[0038] The parking space assessment module obtains the parking space fluctuation assessment of different enterprises on different dates by statistically analyzing historical parking space occupancy. Based on the enterprise parking space fluctuation assessment, it judges the parking space usage dynamics of both park users and non-park users, and generates the parking demand level for different enterprises in the park, specifically including:

[0039] Based on the historical parking space occupancy data of different companies on different dates, the similarity of parking space occupancy among different companies under the same date type is determined, and companies with similar parking needs are screened based on the similarity of parking space occupancy.

[0040] By determining the daily fluctuation assessment of parking space occupancy for businesses with similar parking needs under different date types within a recent preset time period, and combining the daily fluctuation assessment to calculate the parking demand level of businesses with similar parking needs, parking demand is predicted based on the business's parking space fluctuation assessment and parking demand level. Specifically, when a business's parking space fluctuation assessment, i.e., several consecutive daily fluctuation assessments, exceeds a preset fluctuation threshold and the parking demand level is higher than a preset demand level threshold, it is predicted that the business will have high parking demand in the future preset time period. The division of parking space recommendation units is dynamically adjusted according to the prediction results, and more temporary parking space resources are allocated to businesses with high demand.

[0041] Determine if the parking demand exceeds a preset demand threshold. If so, mark the enterprise as a high-demand enterprise and allocate parking resources to it first; otherwise, arrange them in order of parking demand.

[0042] Examples of the above-mentioned technical implementation methods are as follows:

[0043] The number of parking spaces occupied is obtained through statistics: real-time monitoring of the number of parking spaces occupied by users in the park and the number of parking spaces occupied by users outside the park; for example, at 10:00 am on a certain weekday, the number of parking spaces occupied by users in the park for Company C is 50, and the number of parking spaces occupied by users outside the park is 10.

[0044] The assessment of parking space fluctuations for enterprises is achieved by: statistically analyzing the historical number of parking spaces occupied, calculating the daily fluctuation of parking space occupancy rate, and calculating the standard deviation of parking space occupancy rate. The smaller the standard deviation, the lower the fluctuation, and vice versa. For example, the standard deviation of parking space occupancy rate for enterprise D in the most recent 30 working days is 0.15 (low fluctuation), while the standard deviation of parking space occupancy rate for enterprise E is 0.45 (high fluctuation).

[0045] Parking demand is assessed by calculating parking space occupancy similarity and daytime fluctuation. Parking space occupancy similarity is calculated by comparing the parking space occupancy patterns of different companies on the same date type. For example, using cosine similarity to calculate the parking space occupancy curves of companies F and G on weekdays, a similarity of 0.85 (high similarity) is obtained, indicating that they have similar demand. Daytime fluctuation is calculated by calculating the fluctuation of parking space occupancy (e.g., weekday fluctuation range) of companies in the park on different date types within the most recent preset time (e.g., 30 days). For example, the daytime fluctuation assessment of company H on weekdays is 0.6 (based on the difference between the maximum and minimum values).

[0046] Finally, the parking demand score is calculated: combining similarity and daytime fluctuation assessment, using the weighted formula: Parking demand score = 0.7 × average similarity + 0.3 × daytime fluctuation assessment; assuming that the average similarity of enterprise F is 0.8 and the daytime fluctuation assessment is 0.5, then the demand score = 0.7 × 0.8 + 0.3 × 0.5 = 0.71; the preset demand threshold is 0.7, therefore enterprise F is marked as a high-demand enterprise and is given priority in resource allocation.

[0047] Fourth, the demand forecasting module is used to predict parking demand based on the assessment results of enterprise parking space fluctuation and parking demand level, and to divide parking space recommendation units and allocate matching parking resources to different parking space recommendation units. For example, if enterprise I's fluctuation assessment exceeds the preset fluctuation threshold of 0.5 and its parking demand level is higher than 0.7, it is predicted that it will have high parking demand on the next working day; subsequently, the module divides the parking space recommendation units:

[0048] The methods for dividing parking space recommendation units include:

[0049] Using the parking space occupancy fluctuation threshold as a constraint, the goal is to ensure that the parking space fluctuation of enterprises in different parking space recommendation units does not exceed the fluctuation threshold, and to minimize the difference between the number of enterprises in different parking space recommendation units and the number of parking space recommendation units.

[0050] Based on a company's real-time parking information and parking demand, companies are assigned to different parking recommendation units, and each unit is allocated matching parking resources, including adjustments to the ratio of parking spaces for park users and temporary parking spaces. When a company is assigned to a separate parking recommendation unit, the company is treated as a separate entity for parking resource allocation, and the parking recommendation strategy is adjusted in real time to suit the company's individual needs.

[0051] Specifically, the unit division method uses a parking space occupancy fluctuation threshold (e.g., 0.5) as a constraint to ensure that the difference in fluctuation among enterprises within the same unit does not exceed the threshold, and to minimize the difference in the number of units and enterprises. For example, the park has enterprises J (fluctuation 0.3), K (fluctuation 0.4), and L (fluctuation 0.6). Since the difference in fluctuation between J and K is 0.1 (≤0.5), they can be grouped together; L's fluctuation of 0.6 is more than 0.5 compared to J / K, so it needs to be grouped separately. Ultimately, two units are divided: Unit 1 (J and K) and Unit 2 (L), with the minimum number of units and a balanced number of enterprises. By allocating parking resources to each unit, the ratio of user parking spaces to temporary parking spaces in the park is adjusted; for example, more temporary parking spaces are allocated to the high-demand Unit 2 (e.g., increasing the proportion of temporary parking spaces to 40%).

[0052] Parking space recommendation strategies include:

[0053] The system collects real-time data on the current parking space status and vehicle trajectories of enterprises through a 5G communication network. Based on the location information of the enterprise's recommended parking space units, a path planning algorithm is used to calculate the optimal parking route. The path planning algorithm includes a traffic resistance model to evaluate the route's turning angle, the number of shock absorbers, and real-time traffic conditions. Based on the parking request generation time, vehicle historical violation records, and enterprise priority, a reservation value is calculated to generate a queue sequence. Parking space reservations are then scheduled according to the reservation value, enabling collaborative parking management among different enterprises.

[0054] The above parking space recommendation strategy is implemented by the system collecting parking space status and vehicle trajectory in real time through the 5G network, and using a path planning algorithm to calculate the optimal parking route; the driving resistance model evaluates the route's turning degree, the number of shock absorbers, and real-time traffic conditions (such as congestion index); for example, if a route has many turns and many shock absorbers, the driving resistance value is high, and the system will prioritize recommending routes with low driving resistance values.

[0055] Simultaneously, a reservation value is calculated based on the parking request generation time, the vehicle's historical violation records (such as the number of illegal parking incidents), and the company's priority. The reservation value is calculated by weighting the importance ratio of the parking request generation time, the vehicle's historical violation records (such as the number of illegal parking incidents), and the company's priority: Reservation Value = 0.5 × (1 - Generation Time Offset) + 0.3 × (1 - Violation Record Ratio) + 0.2 × Company Priority Score. The generation time offset refers to the distance between the request time and the peak period, the violation record ratio refers to the proportion of historical violations to the total number of parking incidents, and the company priority is determined by the demand level. A queue sequence is generated based on the reservation value for parking space reservation scheduling. For example, if company M has a high reservation value, its employee vehicles will be allocated parking spaces with priority.

[0056] The operation process of the above embodiment is as follows:

[0057] Assume the park has companies P, Q, and R:

[0058] First, data was collected; during June weekdays, Company P had an average parking space occupancy rate of 85% from 8:00 to 9:00 AM, Company Q had an occupancy rate of 80% from 6:00 to 7:00 PM, and Company R had an occupancy rate of 60% from 12:00 to 1:00 PM.

[0059] Second, time periods are divided; a preset threshold of 80% is set, with peak periods for enterprises P and Q, but not for enterprise R; parking spaces for enterprise P's park users are reserved for the morning peak, and those for enterprise Q are reserved for the evening peak.

[0060] (1) Parking space assessment; the fluctuation assessment value of enterprise P is 0.55 (high), the fluctuation assessment value of enterprise Q is 0.35 (medium), and the fluctuation assessment value of enterprise R is 0.20 (low); calculate the parking demand degree: the similarity between enterprise P and enterprise Q is 0.75, the demand degree of enterprise P = 0.7×0.75+0.3×0.55=0.69 (slightly lower than the threshold of 0.7), but after adjustment based on real-time data, it is marked as medium demand;

[0061] (2) Demand forecast: The fluctuation of enterprise P is 0.55, which is greater than the fluctuation threshold of 0.5, indicating that it has high demand; the forecasts for enterprise Q and R are medium demand.

[0062] (3) Unit division: with a fluctuation threshold of 0.5, enterprise P is divided into unit 1, and enterprises Q and R are divided into unit 2; unit 1 is allocated more temporary parking spaces (e.g., 30%).

[0063] (4) Recommendation strategy: When an employee of company P requests parking, the system calculates the reservation value: if the generation time is before the morning rush hour, there is no violation record, and the company has a high priority, the reservation value is 0.9, and parking spaces are allocated first and routes with fewer turns are recommended.

[0064] Through the above technical processes, this system coordinates multiple modules to dynamically optimize parking resources by analyzing, predicting, and dynamically allocating parking demand based on historical data, thereby improving parking space utilization and user satisfaction.

[0065] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0066] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this application, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent parking space recommendation system for a smart park, characterized in that, The method comprises the following steps: a data collection module is used to collect historical parking data of different enterprises in a park on different dates; a time period division module is used to determine the number of peak period parking of different enterprises according to the historical parking data of different enterprises on different dates, and to divide the time periods of user parking spaces of different enterprises according to the number of peak period parking of different enterprises, so as to determine real-time parking space information of the enterprises, wherein the user parking spaces include park user parking spaces and non-park user parking spaces; a parking space evaluation module is used to determine the number of parking space occupation according to the real-time parking space information of the enterprises, wherein the number of parking space occupation includes the number of park user parking space occupation and the number of non-park user parking space occupation; an enterprise parking space fluctuation evaluation quantity of different enterprises on different dates is obtained by counting the historical parking space occupation quantity, and the parking space use dynamics of park users and non-park users of the enterprises are determined according to the enterprise parking space fluctuation evaluation quantity, so as to generate the parking demand degree of different enterprises in the park; the parking space evaluation module obtains the enterprise parking space fluctuation evaluation quantity of different enterprises on different dates by counting the historical parking space occupation quantity, determines the parking space use dynamics of park users and non-park users of the enterprises according to the enterprise parking space fluctuation evaluation quantity, and generates the parking demand degree of different enterprises in the park, and the specific method comprises the following steps: the similarity of parking space occupation of different enterprises on the same date type is determined according to the historical parking space occupation quantity of different enterprises on different dates, and similar parking demand enterprises are screened based on the similarity of parking space occupation; the inter-day fluctuation evaluation quantity on different date types is determined by counting the fluctuation of the number of parking space occupation of the similar parking demand enterprises on different date types within a preset time in the recent past, and the parking demand degree of the similar parking demand enterprises is calculated in combination with the inter-day fluctuation evaluation quantity; it is determined whether the parking demand degree is higher than a preset demand threshold, if yes, the enterprise is marked as a high demand enterprise, and parking resources are preferentially allocated; otherwise, the enterprises are sequentially arranged according to the size of the parking demand degree; a demand prediction module is used to predict parking demand based on the evaluation results of the enterprise parking space fluctuation evaluation quantity and the parking demand degree, and to divide parking space recommendation units, and to allocate matching parking resources to different parking space recommendation units. 2.The intelligent parking space recommendation system for smart park according to claim 1, wherein, The historical parking data includes the parking space occupation rate, vehicle position information, vehicle stay time and vehicle type information of the enterprises on different dates and at different time periods. 3.The intelligent parking space recommendation system for smart park according to claim 1, wherein, The peak period parking is determined according to the historical parking data of the enterprises on different dates within a preset time in the recent past, including calculating the average value of the parking space occupation rate at different time periods, determining the time period with a parking space occupation rate higher than a preset threshold as the peak period parking, and dividing the time periods of user parking spaces according to the number of peak period parking. 4.The intelligent parking space recommendation system for smart park according to claim 1, wherein, The specific method for predicting parking demand based on the enterprise parking space fluctuation evaluation quantity and the parking demand degree comprises the following steps: when the enterprise parking space fluctuation evaluation quantity, i.e. the inter-day fluctuation evaluation quantity for a plurality of consecutive days, exceeds a preset fluctuation threshold and the parking demand degree is higher than a preset demand degree threshold, it is predicted that the enterprise has high parking demand within a preset time period in the future; the division of the parking space recommendation units is dynamically adjusted according to the prediction results, and more temporary parking space resources are allocated to the high demand enterprises. 5.The intelligent parking space recommendation system for smart park according to claim 4, wherein, The division method of the parking space recommendation unit comprises: With the parking space occupation fluctuation threshold as a constraint condition, the fluctuation of the parking spaces of enterprises in different parking space recommendation units is not greater than the fluctuation threshold, and the difference between the number of enterprises in different parking space recommendation units is minimized and the number of parking space recommendation units is minimized as a target, the number of parking space recommendation units is determined; According to the real-time parking space information and parking demand degree of the enterprise, the enterprise is allocated to different parking space recommendation units, and the matching parking resources, including the proportion adjustment of the park user parking space and the temporary parking space, are allocated to each unit. 6.The intelligent parking space recommendation system for smart park according to claim 5, wherein, When the enterprise is divided into separate parking space recommendation units, the parking resource allocation of the enterprise is carried out as a separate parking space recommendation subject, and the parking space recommendation strategy is adjusted in real time to adapt to the individual needs of the enterprise. 7.The intelligent parking space recommendation system for smart park according to claim 6, wherein, The parking space recommendation strategy comprises: The current enterprise parking space state and vehicle trajectory are collected in real time through a 5G communication network, the optimal parking route is calculated according to the position information of the current enterprise parking space recommendation unit by using a path planning algorithm, the path planning algorithm comprises a row resistance value model for evaluating the route turning degree, the number of shock absorbing belts and the real-time traffic condition; The reservation value is calculated based on the generation time of the parking request, the vehicle historical violation record and the enterprise priority, the queuing sequence is generated, the parking space reservation scheduling is carried out according to the reservation value sorting, and the collaborative parking management of different enterprises is realized.

Images