Parking lot vehicle entrance control method, device and equipment and storage medium

Abstract

The invention provides a parking lot vehicle entrance control method and device, equipment and a storage medium, and the method comprises the steps: receiving an entrance request of a parking lot of a temporary parking vehicle; acquiring current state data and parking reservation data of the parking lot; based on the current state data and the reserved parking data, predicting parking demands and parking space supply changes of different types of vehicles in current and future preset time periods; and the parking demand, the prediction result of the parking space supply change and the current parking data of the parking lot are utilized to determine whether to respond to the entrance request of the temporarily parked vehicle. By adopting the method, the scientificity and the high efficiency of parking lot entrance control are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of smart parking, and more specifically, to a method, device, equipment, and storage medium for controlling vehicle entry into a parking lot. Background Technology

[0002] With the continuous advancement of urbanization and the rapid growth of car ownership, parking lots, as an important component of urban transportation infrastructure, directly impact traffic flow and user parking experience through their operational management efficiency. Vehicle entry control, as the primary aspect of parking lot management, aims to ensure orderly vehicle entry while achieving the rational utilization of parking space resources, thus becoming a key technological area urgently needing optimization within the industry.

[0003] In existing technologies, mainstream parking lot vehicle entry control methods rely on parking space availability counting devices (such as inductive loops and ultrasonic detectors) to obtain the number of remaining parking spaces in real time, and adopt a single decision logic of "first come, first served": when the counting device shows that there are remaining parking spaces, temporary parking vehicles are allowed to enter; when there are 0 remaining parking spaces, the entry request of temporary parking vehicles is directly rejected. In some scenarios, manual judgment by on-site security guards is used as an auxiliary factor.

[0004] This existing technology has a significant flaw: it makes decisions solely based on static, current parking space availability data, completely disregarding the needs of users who have already reserved parking spaces for future periods through online platforms, resulting in a lack of foresight in decision-making. For example, when there are ample parking spaces available, the system may accept a large number of temporary parking vehicles, but subsequent reservation users may arrive and find no spaces available, leading to service breaches. This not only reduces the satisfaction of high-value reservation users but may also generate complaints and disputes, making it difficult to balance the revenue from temporary parking with the service guarantee needs of reservation users. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a parking lot vehicle entry control method, device, equipment and storage medium, which significantly improves the scientific nature and efficiency of parking lot entry control.

[0006] In a first aspect, embodiments of this application provide a parking lot vehicle entry control method, the method comprising: Receive parking lot entry requests from vehicles that are temporarily parking; Obtain the current status data and reserved parking data of the parking lot; Based on the current status data and the reserved parking data, predict the changes in parking demand and parking space supply for different types of vehicles in the current and future preset time periods. The prediction results of changes in parking demand and parking space supply, along with the current parking data of the parking lot, are used to determine whether the parking lot is responding to the entry request of the temporary parking vehicle.

[0007] Optionally, the current status data includes historical parking records, real-time parking records, and historical departure data; The method of predicting changes in parking demand and parking space supply for different types of vehicles within the current and future preset time periods based on the current status data and the reserved parking data includes: Based on the reserved parking data, the arrival demand of reserved vehicles in the current and future time periods is predicted; Based on the historical parking records and the real-time parking records, predict the arrival demand of temporary parking vehicles in the current and future time periods; Based on the historical departure data, the departure trend of vehicles in the parking lot is predicted for the current and future time periods to determine the changes in parking space supply.

[0008] Optionally, the prediction is achieved in the following way: Prediction is performed using a prediction model, wherein the prediction model is a long short-term memory network model, a gated recurrent unit model, or an XGBoost model; Specifically, the reservation parking data is input into the prediction model to obtain the arrival demand of the reserved vehicles; The historical parking records and the real-time parking records are input into the prediction model to obtain the arrival demand of temporary parking vehicles; The historical departure data is input into the prediction model to obtain the change in parking space supply.

[0009] Optionally, the step of using the predicted results of changes in parking demand and parking space supply, and the current parking data of the parking lot to determine whether to respond to the entry request of the temporary parking vehicle includes: A constrained optimization model is constructed and solved using the predicted results of the parking demand and the changes in parking space supply to determine the number of vehicles permitted to park temporarily in the current and future preset time periods. Based on the number of vehicles permitted to park temporarily and the current parking data of the parking lot, determine whether the parking lot is saturated. If it is saturated, do not respond to the entry requests of the vehicles to park temporarily.

[0010] Optionally, the step of constructing and solving a constrained optimization model using the predicted results of changes in parking demand and parking space supply to determine the number of vehicles permitted for temporary parking includes: Based on the predicted changes in parking demand and parking space supply, a constrained optimization model is constructed, wherein the constrained optimization model takes maximizing operating revenue as the objective function and satisfies the parking demand of all reserved vehicles as the constraint condition. Solving the constrained optimization model yields the optimal number of temporary parking vehicles allowed for each time period within the future timeframe. The optimal number of vehicles allowed to temporarily park during the current decision-making period is determined as the number of vehicles allowed to temporarily park.

[0011] Optionally, solving the constrained optimization model to obtain the optimal number of temporary parking vehicles allowed for each time period within the future time period includes: Multiple simulation conditions are constructed, each of which corresponds to a set of predicted parking demand and parking space supply change data. A training dataset is constructed based on the simulation conditions and the aforementioned constraint optimization model; Based on the training dataset, a statistical decision model is trained. The predicted results of the parking demand and the changes in parking space supply are input into the trained statistical decision model to obtain the number of vehicles permitted to park temporarily.

[0012] Optionally, constructing the training dataset based on each simulation condition and the constrained optimization model includes: For each simulation condition, the optimal admission decision for that simulation condition is obtained by solving the constraint optimization model. The simulation conditions and their corresponding optimal admission decisions are used as a set of training data. The training dataset is obtained by integrating the various training data.

[0013] Secondly, embodiments of this application provide a parking lot vehicle access control device, the device comprising: The access request receiving module is used to receive access requests from parking lots for vehicles that are temporarily parking. The parking lot data acquisition module is used to acquire the current status data and reserved parking data of the parking lot; The data prediction module is used to predict changes in parking demand and parking space supply for different types of vehicles in the current and future preset time periods based on the current status data and the reserved parking data. The access request response module is used to determine whether to respond to the entry request of the temporary parking vehicle based on the predicted results of the parking demand and the changes in the parking space supply, as well as the current parking data of the parking lot.

[0014] Optionally, the current status data includes historical parking records, real-time parking records, and historical departure data; The method of predicting changes in parking demand and parking space supply for different types of vehicles within the current and future preset time periods based on the current status data and the reserved parking data includes: Based on the reserved parking data, the arrival demand of reserved vehicles in the current and future time periods is predicted; Based on the historical parking records and the real-time parking records, predict the arrival demand of temporary parking vehicles in the current and future time periods; Based on the historical departure data, the departure trend of vehicles in the parking lot is predicted for the current and future time periods to determine the changes in parking space supply.

[0015] Optionally, the prediction is achieved in the following way: Prediction is performed using a prediction model, wherein the prediction model is a long short-term memory network model, a gated recurrent unit model, or an XGBoost model; Specifically, the reservation parking data is input into the prediction model to obtain the arrival demand of the reserved vehicles; The historical parking records and the real-time parking records are input into the prediction model to obtain the arrival demand of temporary parking vehicles; The historical departure data is input into the prediction model to obtain the change in parking space supply.

[0016] Optionally, the step of using the predicted results of changes in parking demand and parking space supply, and the current parking data of the parking lot to determine whether to respond to the entry request of the temporary parking vehicle includes: A constrained optimization model is constructed and solved using the predicted results of the parking demand and the changes in parking space supply to determine the number of vehicles permitted to park temporarily in the current and future preset time periods. Based on the number of vehicles permitted to park temporarily and the current parking data of the parking lot, determine whether the parking lot is saturated. If it is saturated, do not respond to the entry requests of the vehicles to park temporarily.

[0017] Optionally, the step of constructing and solving a constrained optimization model using the predicted results of changes in parking demand and parking space supply to determine the number of vehicles permitted for temporary parking includes: Based on the predicted changes in parking demand and parking space supply, a constrained optimization model is constructed, wherein the constrained optimization model takes maximizing operating revenue as the objective function and satisfies the parking demand of all reserved vehicles as the constraint condition. Solving the constrained optimization model yields the optimal number of temporary parking vehicles allowed for each time period within the future timeframe. The optimal number of vehicles allowed to temporarily park during the current decision-making period is determined as the number of vehicles allowed to temporarily park.

[0018] Optionally, solving the constrained optimization model to obtain the optimal number of temporary parking vehicles allowed for each time period within the future time period includes: Multiple simulation conditions are constructed, each of which corresponds to a set of predicted parking demand and parking space supply change data. A training dataset is constructed based on the simulation conditions and the aforementioned constraint optimization model; Based on the training dataset, a statistical decision model is trained. The predicted results of the parking demand and the changes in parking space supply are input into the trained statistical decision model to obtain the number of vehicles permitted to park temporarily.

[0019] Optionally, constructing the training dataset based on each simulation condition and the constrained optimization model includes: For each simulation condition, the optimal admission decision for that simulation condition is obtained by solving the constraint optimization model. The simulation conditions and their corresponding optimal admission decisions are used as a set of training data. The training dataset is obtained by integrating the various training data.

[0020] Thirdly, embodiments of this application provide a computer device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the parking lot vehicle entry control method described in any of the optional embodiments of the first aspect are performed.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the parking lot vehicle entry control method described in any of the optional embodiments of the first aspect.

[0022] The technical solution provided in this application includes, but is not limited to, the following beneficial effects: Receiving entry requests from parking lots for temporarily parked vehicles allows for the rapid capture of these entry needs, providing precise trigger signals for subsequent entry control decisions. This ensures that the entry requests of temporarily parked vehicles are responded to in a timely manner, avoiding user waiting or dissatisfaction caused by missed or delayed processing of requests, and laying an efficient foundation for the entire entry control process.

[0023] By acquiring the current status data and reserved parking data of the parking lot, a comprehensive understanding of the parking lot's real-time operation and future reservation demand can be obtained. This breaks the limitations of a single data dimension, making the core information needed for decision-making more complete and comprehensive, avoiding judgment bias caused by missing data, and providing solid data support for subsequent prediction and decision-making processes.

[0024] Based on the current status data and the reserved parking data, the changes in parking demand and parking space supply for different types of vehicles can be predicted in the current and future preset time periods. This allows for advance prediction of the supply and demand trends of parking space resources, making entry control decisions more forward-looking and overcoming the short-sighted limitations of relying solely on current static data. It also provides a scientific basis for the rational allocation of parking space resources.

[0025] By using the predicted results of changes in parking demand and parking space supply, as well as the current parking data of the parking lot, to determine whether to respond to the entry request of the temporary parking vehicle, the entry decision can be made more accurate and reasonable. This not only protects the parking rights of reserved vehicles, but also makes full use of parking space resources to increase the revenue of temporary parking, and effectively avoids the problems of service breach or resource idleness.

[0026] In summary, the above steps are interconnected. By accurately receiving demand, comprehensively acquiring data, scientifically predicting trends, and making reasonable decisions and responses, the scientific nature and efficiency of parking lot entry control have been significantly improved. This has achieved a dual optimization of temporary parking revenue and service guarantee for reserved users, effectively improving the level of parking lot operation and management and the user parking experience.

[0027] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart of a parking lot vehicle entry control method provided in Embodiment 1 of this application is shown; Figure 2 This document shows a flowchart of a method for predicting changes in parking demand and parking space supply provided in Embodiment 1 of this application. Figure 3 A flowchart of an entry request confirmation response method provided in Embodiment 1 of this application is shown; Figure 4 A flowchart of a method for determining the number of vehicles permitted to temporarily park, provided in Embodiment 1 of this application, is shown. Figure 5 This document shows a flowchart of a method for determining the optimal number of temporarily parked vehicles provided in Embodiment 1 of this application; Figure 6 A flowchart of a training dataset construction method provided in Embodiment 1 of this application is shown; Figure 7 The overall execution flowchart of a parking lot temporary parking vehicle entry control method provided in Embodiment 1 of this application is shown; Figure 8 This paper shows a schematic diagram of the structure of a parking lot vehicle entry control device provided in Embodiment 2 of this application; Figure 9 A schematic diagram of the structure of a computer device provided in Embodiment 3 of this application is shown. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0031] Example 1 To facilitate understanding of this application, the following is combined with... Figure 1 The flowchart illustrating a parking lot vehicle entry control method provided in Embodiment 1 of this application describes Embodiment 1 in detail.

[0032] See Figure 1 As shown, Figure 1 A flowchart of a parking lot vehicle entry control method provided in Embodiment 1 of this application is shown, wherein the method includes steps S101 to S104: S101: Receives entry requests from parking lots for temporarily parked vehicles.

[0033] Specifically, the entry request is initiated automatically by the license plate recognition (LPR) camera, vehicle detector and other equipment at the parking lot entrance after collecting the identity information of the temporarily parked vehicle. It is sent to the central decision-making system in real time without human intervention, as the starting trigger signal for access judgment.

[0034] S102: Obtain the current status data and reserved parking data of the parking lot.

[0035] Specifically, the current status data comes from Hive data warehouse tool tables and contains three core types of information: First, the physical parking space status obtained through inductive loops and ultrasonic detectors (there may be a data delay of 10 minutes to 1 hour); second, historical parking records (entry and exit times and parking durations of temporary parking vehicles, monthly pass vehicles, and fixed package vehicles over the past year); third, real-time parking records (instant data such as the total number of vehicles in the parking lot, the number of occupied parking spaces, and the number of temporary parking vehicles allowed in the current time period); and fourth, historical departure data (pattern data such as the distribution of departure times of past vehicles and peak departure times).

[0036] The parking reservation data comes from the central operation platform and includes order information for all future pre-order users, covering features such as pre-order time period, parking duration, vehicle identification, and historical fulfillment rate. This data is the core basis for predicting the arrival demand of reserved vehicles.

[0037] S103: Based on the current status data and the reserved parking data, predict the changes in parking demand and parking space supply for different types of vehicles in the current and future preset time periods.

[0038] Specifically, the prediction is achieved through the system's built-in predictive analysis module, which integrates regression models, time series models, or neural network models (such as LSTMLong Long Short-Term Memory Network, GRU Gated Recurrent Unit, and XGBoost Extreme Gradient Boosting). It adopts a "two-way prediction" logic, predicting both demand and supply.

[0039] Demand forecasting targets two types of users: priority users (including monthly pass users, pre-purchase users, and other groups with priority parking rights) and non-priority users (temporary parking users without priority rights), outputting the expected arrival numbers and peak arrival times for each future time period. Supply forecasting focuses on the exit patterns of vehicles within the parking lot, outputting the expected number of parking spaces to be released for each future time period, clarifying the dynamic supply of parking spaces.

[0040] The forecast time window is flexible and adjustable. In practical applications, it can cover the next moment, several hours, or a day, or it can be set according to the daily / weekly / monthly average dimension to adapt to the operation scenarios of different parking lots.

[0041] S104: Using the predicted results of the parking demand and the changes in parking space supply, and the current parking data of the parking lot, determine whether the parking lot is responding to the entry request of the temporary parking vehicle.

[0042] Specifically, the core of the current parking data includes the number of occupied parking spaces and the number of vehicles that have been granted temporary parking access during the current period. Combined with the supply and demand forecast results, the optimal number of vehicles that can be granted temporary parking access during the current period is obtained through a constrained optimization model. Then, by comparing the number of vehicles that have been granted access with the optimal number, it is determined whether there is still room for access, and thus it is determined whether to respond to the entry request.

[0043] In one optional implementation, the current status data includes historical parking records, real-time parking records, and historical departure data.

[0044] Specifically, historical parking records are the total parking data of the parking lot over the past year, covering information such as entry and exit times, parking durations, and vehicle types for various types of users, including temporary parking, monthly passes, and fixed packages. This data is used to uncover long-term fluctuation patterns in parking demand.

[0045] Real-time parking records provide the immediate status data of vehicles in the parking lot, including the total number of parking spaces, the number of occupied parking spaces, the number of remaining parking spaces, and the number of vehicles that have been granted temporary parking access during this period, providing a current basis for real-time decision-making. Historical departure data is data related to the departure of past vehicles, including the distribution of parking duration, peak departure times (such as 6 pm on weekdays and 9 pm in shopping malls), and differences in departure patterns on different dates, which is a key basis for predicting changes in parking space supply.

[0046] See Figure 2 As shown, Figure 2 The flowchart illustrates a method for predicting changes in parking demand and parking space supply provided in Embodiment 1 of this application. The method, based on the current status data and the reserved parking data, predicts changes in parking demand and parking space supply for different types of vehicles within the current and future preset time periods, including steps S201-S203: S201: Based on the reserved parking data, predict the arrival demand of reserved vehicles in the current and future time periods.

[0047] Specifically, the reservation parking data includes core features such as the order time of the pre-purchase user, parking duration, vehicle identification, and historical fulfillment rate. After inputting this data into the prediction model, the model will output the expected number of reserved vehicles arriving in each future time period and the concentrated arrival time window, providing accurate reference for reserving parking space resources and ensuring that service commitments are fulfilled.

[0048] S202: Based on the historical parking records and the real-time parking records, predict the arrival demand of temporary parking vehicles in the current and future time periods.

[0049] Specifically, historical parking records are used to uncover long-term patterns in temporary parking demand, such as differences in traffic flow between weekdays and weekends, peak hour distribution, and traffic fluctuations during holidays; real-time parking records are used to capture current trends in temporary parking traffic, such as the cumulative number of temporary stops in the current period and the rate of traffic growth.

[0050] The forecasting process also incorporates factors such as weather and surrounding activities (e.g., shopping district promotions, stadium events) to further improve forecast accuracy. Ultimately, it outputs the total demand for temporary parking vehicles, peak demand at different times, and distribution, providing data support for revenue optimization.

[0051] S203: Based on the historical departure data, predict the departure trend of vehicles in the parking lot during the current and future time periods to determine the changes in parking space supply.

[0052] Specifically, by analyzing the distribution of vehicle parking duration in historical departure data (such as the proportion of short-term, medium-term, and long-term parking), the departure probability at different times, and the correlation between vehicle type and departure patterns, the number of parking spaces to be released in future time periods can be predicted.

[0053] The forecast results directly clarify the dynamic changes in parking space supply. For example, if 10 vehicles are expected to leave during a certain period, that corresponds to 10 newly available parking spaces, providing a core basis for judging whether more temporary parking vehicles can be accommodated.

[0054] In an optional implementation, the prediction is achieved in the following way: Prediction is performed using a prediction model, wherein the prediction model is a long short-term memory network model, a gated recurrent unit model, or an XGBoost model.

[0055] Specifically, the Long Short-Term Memory (LSTM) network model excels at capturing long-term dependencies in time-series data and can effectively handle cross-time period regularities in parking data, making it suitable for scenarios with long prediction periods and strong data correlations. The Gated Recurrent Unit (GRU) model simplifies the network structure while retaining the core advantages of LSTM, resulting in higher computational efficiency and making it suitable for parking lots with high traffic volume and the need for rapid prediction. The XGBoost model is suitable for handling nonlinear features and can effectively integrate multi-dimensional correlation factors such as weather and surrounding activities, improving prediction accuracy in complex scenarios.

[0056] Specifically, the reserved parking data is input into the prediction model to obtain the arrival demand of the reserved vehicles.

[0057] Specifically, the input parking reservation data includes features such as the time period, quantity, parking duration, historical fulfillment rate, and vehicle type of the pre-order. The model learns the correlation between these features and arrival behavior, and outputs the expected number and time distribution of reserved vehicles for each future time period, ensuring that the reserved parking spaces accurately match the demand.

[0058] The historical parking records and the real-time parking records are input into the prediction model to obtain the arrival demand of temporary parking vehicles.

[0059] Specifically, the input data includes the number of vehicles temporarily parked in different historical time periods, the real-time cumulative number of temporary parking, weekday / weekend identifiers, time period indexes, weather parameters, and surrounding activity information. By mining the patterns in this data, the model outputs the total demand for temporary parking vehicles in the future, the peak demand in each time period, and the distribution, providing support for profit maximization decisions.

[0060] The historical departure data is input into the prediction model to obtain the change in parking space supply.

[0061] Specifically, the input data includes information such as the number of vehicles leaving the parking lot in different historical time periods, parking duration, vehicle type, and date type (weekday / weekend / holiday). By learning the departure patterns in this data, the model outputs the number of parking spaces expected to be released in different future time periods, i.e. the change in parking space supply, providing a basis for dynamically adjusting the number of temporary parking spaces.

[0062] In an optional implementation, see Figure 3 As shown, Figure 3 The flowchart illustrates a method for confirming an entry request response according to Embodiment 1 of this application. The method involves determining whether to respond to the entry request of a temporary parking vehicle using the predicted results of changes in parking demand and parking space supply, as well as the current parking data of the parking lot. This includes steps S301-S302: S301: Construct and solve a constrained optimization model using the predicted results of the parking demand and the changes in parking space supply to determine the number of vehicles permitted to park temporarily in the current and future preset time periods.

[0063] Specifically, the core objective of the constrained optimization model is to "maximize the revenue from temporary parking". The objective function includes items such as revenue from temporary parking fees, operating cost penalties, service breach penalties, and opportunity cost penalties.

[0064] The rigid constraint of the model is that "priority users (monthly pass, pre-purchase) who arrive at any time in the future can obtain a parking space", while limiting the total number of parking spaces occupied by temporary parking and priority users to no more than the total capacity of the parking lot, that is, "monthly pass parking spaces occupied ≤ total number of parking spaces" and "temporary parking spaces occupied ≤ total number of parking spaces - monthly pass parking spaces".

[0065] The solution adopts an engineering approach of "offline large-scale optimization simulation + online statistical model" to avoid the computational delay caused by real-time solution of complex mixed integer linear programming (MILP) models, ensuring decision-making efficiency. The solution result is the optimal number of temporary parking spaces for each time period that dynamically adapts to supply and demand, rather than the number of static reserved parking spaces.

[0066] S302: Based on the number of vehicles permitted to enter for temporary parking and the current parking data of the parking lot, determine whether the parking lot is saturated. If it is saturated, do not respond to the entry request of the temporary parking vehicles.

[0067] Specifically, the saturation judgment criterion is "the number of currently occupied parking spaces + the number of currently applied temporary parking vehicles > the total number of parking spaces - the number of reserved priority user parking spaces", where the number of reserved priority user parking spaces is determined based on the prediction results of S301.

[0068] If the system determines that the parking space is saturated, it sends a "reject" command to the barrier gate and provides clear feedback to the user through the parking lot entrance screen (such as "parking space reserved, please try again later") to reduce user uncertainty; if the parking space is not saturated, it sends a "allow entry" command and opens the barrier gate to allow entry.

[0069] In an optional implementation, see Figure 4 As shown, Figure 4 The flowchart illustrates a method for determining the number of vehicles permitted to park temporarily, as provided in Embodiment 1 of this application. The method involves constructing and solving a constrained optimization model using the predicted results of parking demand and parking space supply changes to determine the number of vehicles permitted to park temporarily. This includes steps S401-S403: S401: Based on the predicted changes in parking demand and parking space supply, construct a constrained optimization model, wherein the constrained optimization model takes maximizing operating revenue as the objective function and satisfies the parking demand of all reserved vehicles as the constraint condition.

[0070] Specifically, the mathematical expression for the objective function is:

[0071] The characters in the formula are defined as follows: : Temporary parking fee for each segment (actual parameters, which can be set according to the operating scenario); Time period Number of vehicles accepted for temporary parking (decision variable, non-negative integer); The starting index of the current decision-making period (e.g., dividing a 24-hour day into 48 half-hour periods). (The sequence number corresponding to the current time period). : The total number of time periods in the future forecast (i.e., the total number of time periods included in the forecast time window); Temporary parking fee rate coefficient (used to adjust the weight of temporary parking revenue, set according to the operation strategy); Total number of parking spaces in the parking lot (a fixed parameter determined based on the physical facilities of the parking lot); : Daily equivalent cost of monthly pass (actual parameters, calculated from the monthly pass price); Opportunity cost penalty coefficient (used to penalize situations where temporary parking resources are not fully utilized; set based on operating costs). Total temporary parking demand (the total scale of temporary parking demand within the predicted future time window); Whether to accept the first 0-1 decision variables for temporary parking demand ( =1 indicates acceptance. =0 indicates rejection); Service breach penalty coefficient (used to penalize situations where priority user needs are not met; set based on user complaint costs). Total demand from monthly pass users (the total number of monthly pass users expected to attend within a future time window). Whether to accept the first 0-1 decision variables for monthly card demand ( =1 indicates a reserved parking space. =0 indicates a service breach. Since monthly card users are priority users, the actual solution... (Constantly 1).

[0072] The constraints include three items, and their mathematical expressions and character definitions are as follows: 1. Constraint 1: ,in For time period Number of parking spaces occupied by monthly pass users This refers to the total number of parking spaces in the parking lot, meaning that "monthly pass users' parking space occupancy shall not exceed the total parking lot capacity"; 2. Constraint Two: ,in For time period The number of vehicles accepted for temporary parking For time period Number of parking spaces occupied by monthly pass users This refers to the total number of parking spaces in the parking lot, meaning that "priority will be given to the needs of monthly pass users, and the total number of parking spaces occupied by temporary parking and monthly pass users shall not exceed the total capacity." 3. Constraint Three: , ,in To accommodate decision-making variables related to temporary parking demand, The decision variable for ensuring the demand for monthly passes is defined as "the decision outcome is only a binary option of 'accept' or 'reject'".

[0073] S402: Solve the constrained optimization model to obtain the optimal number of temporary parking vehicles allowed for each time period in the future.

[0074] Specifically, the solution process is divided into two stages: offline and online. The offline stage involves creating a simulation environment to generate tens of thousands of calculations, including predicted temporary parking demand (MM), predicted monthly pass demand (QQ), current time period index (k), and current remaining parking spaces (…). In the randomized simulation scenarios, the MILP model is solved for each scenario to obtain the optimal admission decision; in the online stage, the offline-trained statistical model is used to quickly output the results, avoiding real-time computation delay.

[0075] The solution yields the optimal number of temporary parking permits for each future decision-making period (e.g., dividing a 24-hour day into 48 half-hour periods), achieving dynamic optimization across all time periods.

[0076] S403: The optimal number of vehicles allowed to temporarily park during the current decision-making period is determined as the number of vehicles allowed to temporarily park.

[0077] Specifically, the current decision-making period is identified by the period index (k). The system automatically matches the corresponding period index based on the current time and extracts the optimal number of temporary parking permits for that period, ensuring that the decision is accurately aligned with the real-time scenario and avoiding resource waste or service breaches caused by using static values.

[0078] In an optional implementation, see Figure 5 As shown, Figure 5 The flowchart illustrates a method for determining the optimal number of temporarily parked vehicles according to Embodiment 1 of this application. The step of solving the constrained optimization model to obtain the optimal number of temporarily parked vehicles for each time period within a future timeframe includes steps S501-S504: S501: Construct various simulation conditions, where each simulation condition corresponds to a set of predicted parking demand and parking space supply change data.

[0079] Specifically, the core explanatory variables for the simulation conditions include four items: predicted temporary parking demand (MM), predicted monthly pass demand (QQ), current time period index (k), and current remaining parking spaces (...). ).

[0080] The reasonable range of variables is set based on the historical fluctuation of user volume over the past year, satisfying constraints such as "temporary parking demand + monthly pass demand ≤ total number of parking spaces"; random combination adopts a grid learning-like approach, with multiple parameters forming a grid and adjusting the simulation. Most variables are sampled independently, while some demand-related variables (such as temporary parking and monthly pass demand) have summation-related constraints, ultimately generating tens of thousands or even more simulation scenarios to ensure coverage of various operational situations.

[0081] S502: Construct a training dataset based on the simulation conditions and the constraint optimization model.

[0082] Specifically, for each simulation scenario ( , , , To obtain the current time period under this scenario, run the complete MILP constrained optimization model to solve it. The optimal number of vehicles allowed to park temporarily ( ), which is the explained variable.

[0083] The predicted temporary parking demand in the i-th simulation scenario (corresponding to the specific value of the basic variable MM in the i-th scenario, i.e. the total number of temporary parking vehicles expected to arrive within the future time window in this scenario). The predicted monthly card demand in the i-th simulation scenario (corresponding to the specific value of the basic variable QQ in the i-th scenario, i.e. the total number of monthly card users expected to arrive within the future time window in this scenario). This is the index of the current time period in the i-th simulation scenario (corresponding to the specific value of the basic variable k in the i-th scenario, i.e., the time period number corresponding to this scenario); The number of remaining parking spaces in the i-th simulation scenario (corresponding to the basic variable) The specific value for the i-th scenario is the real-time number of remaining parking spaces in the current time period under that scenario. The "input variable group" for each scenario ( , , , ) + Output the optimal admission number ( ")" serves as a set of training data to ensure a one-to-one correspondence between input and output.

[0084] S503: Based on the training dataset, train a statistical decision model.

[0085] Specifically, due to the explained variable ( For non-negative integers, the Poisson regression model from the generalized linear model (GLM) is selected, and the model formula is:

[0086] The characters in the formula are defined as follows: Natural logarithm function (connection function used to map linear prediction results to non-negative integer admission numbers); : The optimal number of temporary parking permits for the current time period (model output variable, non-negative integer); Model intercept term (a constant parameter obtained by fitting the training dataset); : Regression coefficients for predicting temporary parking demand (MM) (after fitting, reflecting the impact of MM on (the degree of positive impact) : Regression coefficients for predicting monthly card demand (QQ) (after fitting, reflecting QQ's influence on demand) (the degree of negative impact). The regression coefficients of the current time period index (k) (which, after fitting, reflect the time period characteristics on the regression coefficients) (the degree of influence) : Number of parking spaces remaining ( The regression coefficients (which, after fitting, reflect the impact of real-time parking space status on...) (the degree of positive impact) MM: Forecasted demand for temporary parking (the total number of vehicles expected to arrive for temporary parking within a future time window). QQ: Predicted monthly card demand (total number of monthly card users expected to reach within a future time window). k: Current time period index (same as k0, which is the sequence number corresponding to the current time period); : The number of parking spaces remaining at present (the number of parking spaces remaining in real time period k, which may be delayed by 10-60 minutes).

[0087] By fitting the training dataset, we can obtain estimated values ​​for each parameter of the model. , , , , Tables 1 and 2 below present a summary of the results for a Poisson regression model trained on 31,000 simulated observation points. The estimated model parameters were obtained by fitting the training dataset of 31,000 simulated observation points. =0.0047、 =-0.0019、 =0.0009、 =0.0151. The model's fit was verified by indicators such as log-likelihood (-1.22E+05), bias (65603), and Pearson chi-square (6.55E+04) to ensure prediction accuracy.

[0088] Table 1: Regression Results of the Generalized Linear Model

[0089] Table 2: Model Coefficients Table

[0090] S504: Input the predicted results of the parking demand and the changes in parking space supply into the trained statistical decision model to obtain the number of vehicles permitted to park temporarily.

[0091] Specifically, the system calls lightweight prediction models, such as ARIMA (Auto Regressive Integrated Moving Average) and LSTM (Long Short-Term Memory), to quickly predict the demand for temporary parking (MM) and monthly pass (QQ) for future periods based on the current time and historical data. Simultaneously, it obtains the current period index (k0) and the real-time remaining parking spaces. ).

[0092] Substituting these four variable values ​​into the trained Poisson regression formula, we can calculate... The value is then used for exponentiation. =exp(log( The result is converted into the final non-negative integer number of vehicles allowed to temporarily stop, enabling real-time and near-optimal decision-making; where exp is the natural exponential function, used to restore the logarithmic prediction result to the actual number of vehicles allowed to stop.

[0093] In an optional implementation, see Figure 6 As shown, Figure 6 The flowchart of a training dataset construction method provided in Embodiment 1 of this application is shown, wherein the construction of the training dataset based on various simulation conditions and the constrained optimization model includes steps S601 to S603: S601: For each simulation condition, the optimal admission decision for that simulation condition is obtained by solving the constraint optimization model.

[0094] Specifically, the constrained optimization model is a mixed integer linear programming (MILP) model. Its objective function and constraints are the same as those described in S401. The solution process needs to balance the core contradiction between "accepting more temporary parking vehicles to increase revenue" and "reserving parking spaces to guarantee priority users" to ensure that revenue is maximized without violating service commitments.

[0095] The optimal admission decision is the maximum number of safe temporary parking spaces allowed in the simulation scenario. This avoids both accepting too many temporary parking vehicles, which would leave priority users with no parking spaces (service breach), and being overly conservative in rejecting temporary parking vehicles, which would result in idle parking spaces (revenue loss).

[0096] S602: Use the simulation conditions and their corresponding optimal admission decisions as a set of training data.

[0097] Specifically, each set of training data contains complete input variable information, namely, predicted temporary parking demand (MM), predicted monthly pass demand (QQ), current time period index (k), and current remaining parking spaces (s). k The specific value of the data and the unique corresponding output result (the optimal number of temporary parking permits NM) are used to ensure the integrity and validity of the data.

[0098] S603: Integrate the various training data to obtain the training dataset.

[0099] Specifically, during the integration process, invalid data generated by abnormal simulation scenarios (such as variable values ​​exceeding reasonable ranges or abnormal solution results) needs to be removed to ensure the consistency and representativeness of the dataset.

[0100] The resulting training dataset can reach tens of thousands to hundreds of thousands of data points (e.g., 31,000 simulation observation points), providing sufficient samples for training statistical decision-making models and ensuring the model's generalization ability and prediction accuracy.

[0101] To better illustrate the parking lot vehicle entry control method provided in this application, please refer to [link / reference]. Figure 7 As shown, Figure 7 This document illustrates the overall execution flowchart of a parking lot temporary parking vehicle entry control method provided in Embodiment 1 of this application. The flowchart fully presents the closed-loop process from request reception to instruction execution, corresponding one-to-one with the parking lot vehicle entry control scheme described above (core steps such as S101~S104). The process begins with "Start". First, it executes "Step 1: Receive temporary parking vehicle access request" (corresponding to S101 above, receiving temporary parking vehicle entry request); then it executes "Step 2: Obtain the real-time status of the parking lot (e.g., remaining parking spaces)" (corresponding to S102 above, obtaining the current status data of the parking lot); then it executes "Step 3: Predict future demand for monthly passes and temporary parking, and parking space supply" (corresponding to S301 above, predicting changes in parking demand and parking space supply based on the current status and reservation data); then it executes "Step 4: Construct and solve a constrained optimization problem with the goal of maximizing revenue" (corresponding to S401 above, constructing and solving a constrained optimization model with the goal of maximizing operating revenue); finally, it executes "Step 5: Determine the optimal number of temporary parking vehicles M' for the current time period" (corresponding to S403 above, determining the optimal number of temporary parking vehicles for the current decision period). The process enters the judgment stage: "Is there still room for entry?" (corresponding to the judgment of whether the parking lot is saturated based on the number of entry permits and the current parking data in S302 above). If the judgment result is "Yes", then "Generate 'Allow Entry' instruction"; if the result is "No", then "Generate 'Deny Entry' instruction"; finally, "Execute control instruction (e.g., open the barrier gate)" is executed, and the process ends with "End". This process fully implements the entry control logic described above.

[0102] Example 2 See Figure 8 As shown, Figure 8 A schematic diagram of a parking lot vehicle entry control device according to Embodiment 2 of this application is shown, wherein the device includes: The access request receiving module 801 is used to receive the parking lot entry request from temporary parking vehicles; The parking lot data acquisition module 802 is used to acquire the current status data and reserved parking data of the parking lot; The data prediction module 803 is used to predict the parking demand and parking space supply changes of different types of vehicles in the current and future preset time periods based on the current status data and the reserved parking data. The access request response module 804 is used to determine whether to respond to the entry request of the temporary parking vehicle based on the predicted results of the parking demand and the changes in the parking space supply, as well as the current parking data of the parking lot.

[0103] In one optional implementation, the current status data includes historical parking records, real-time parking records, and historical departure data; The method of predicting changes in parking demand and parking space supply for different types of vehicles within the current and future preset time periods based on the current status data and the reserved parking data includes: Based on the reserved parking data, the arrival demand of reserved vehicles in the current and future time periods is predicted; Based on the historical parking records and the real-time parking records, predict the arrival demand of temporary parking vehicles in the current and future time periods; Based on the historical departure data, the departure trend of vehicles in the parking lot is predicted for the current and future time periods to determine the changes in parking space supply.

[0104] In an optional implementation, the prediction is achieved in the following way: Prediction is performed using a prediction model, wherein the prediction model is a long short-term memory network model, a gated recurrent unit model, or an XGBoost model; Specifically, the reservation parking data is input into the prediction model to obtain the arrival demand of the reserved vehicles; The historical parking records and the real-time parking records are input into the prediction model to obtain the arrival demand of temporary parking vehicles; The historical departure data is input into the prediction model to obtain the change in parking space supply.

[0105] In an optional implementation, the step of utilizing the predicted results of changes in parking demand and parking space supply, as well as the current parking data of the parking lot, to determine whether to respond to the entry request of the temporary parking vehicle includes: A constrained optimization model is constructed and solved using the predicted results of the parking demand and the changes in parking space supply to determine the number of vehicles permitted to park temporarily in the current and future preset time periods. Based on the number of vehicles permitted to park temporarily and the current parking data of the parking lot, determine whether the parking lot is saturated. If it is saturated, do not respond to the entry requests of the vehicles to park temporarily.

[0106] In an optional implementation, the step of constructing and solving a constrained optimization model using the predicted results of parking demand and parking space supply changes to determine the number of vehicles permitted for temporary parking includes: Based on the predicted changes in parking demand and parking space supply, a constrained optimization model is constructed, wherein the constrained optimization model takes maximizing operating revenue as the objective function and satisfies the parking demand of all reserved vehicles as the constraint condition. Solving the constrained optimization model yields the optimal number of temporary parking vehicles allowed for each time period within the future timeframe. The optimal number of vehicles allowed to temporarily park during the current decision-making period is determined as the number of vehicles allowed to temporarily park.

[0107] In an optional implementation, solving the constrained optimization model to obtain the optimal number of temporary parking vehicles allowed for each time period within the future time period includes: Multiple simulation conditions are constructed, each of which corresponds to a set of predicted parking demand and parking space supply change data. A training dataset is constructed based on the simulation conditions and the aforementioned constraint optimization model; Based on the training dataset, a statistical decision model is trained. The predicted results of the parking demand and the changes in parking space supply are input into the trained statistical decision model to obtain the number of vehicles permitted to park temporarily.

[0108] In an optional implementation, constructing the training dataset based on the simulation conditions and the constrained optimization model includes: For each simulation condition, the optimal admission decision for that simulation condition is obtained by solving the constraint optimization model. The simulation conditions and their corresponding optimal admission decisions are used as a set of training data. The training dataset is obtained by integrating the various training data.

[0109] Example 3 Based on the same application concept, see [link / reference] Figure 9 As shown, Figure 9 This illustration shows a structural schematic diagram of a computer device provided in Embodiment 3 of this application, wherein, as shown... Figure 9 As shown, the computer device 900 provided in Embodiment 3 of this application includes: The computer device 900 includes a processor 901, a memory 902, and a bus 903. The memory 902 stores machine-readable instructions that can be executed by the processor 901. When the computer device 900 is running, the processor 901 communicates with the memory 902 through the bus 903. When the machine-readable instructions are executed by the processor 901, the steps of the parking lot vehicle entry control method shown in Embodiment 1 above are executed.

[0110] Example 4 Based on the same concept, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the parking lot vehicle entry control method described in any of the above embodiments.

[0111] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0112] The computer program product for controlling vehicle entry in a parking lot provided in this application includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0113] The parking lot vehicle entry control device provided in this application embodiment can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this application embodiment are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0114] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0117] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0119] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A parking lot vehicle entry control method characterized by comprising: The method comprises: receiving an entry request of a temporarily parked vehicle in a parking lot; obtaining current state data and reservation parking data of the parking lot; based on the current state data and the reservation parking data, predicting parking demand and parking space supply change of different types of vehicles in a current and future preset time period; using the prediction results of the parking demand and the parking space supply change, and the current parking data of the parking lot to determine whether to respond to the entry request of the temporarily parked vehicle.

2. The method of claim 1, wherein, The current state data includes historical parking records, real-time parking records and historical departure data; The prediction of the parking demand and the parking space supply change of different types of vehicles in a current and future preset time period based on the current state data and the reservation parking data comprises: predicting the arrival demand of the reserved vehicles in the current and future time period based on the reservation parking data; predicting the arrival demand of the temporarily parked vehicles in the current and future time period based on the historical parking records and the real-time parking records; predicting the departure trend of the vehicles in the parking lot in the current and future time period based on the historical departure data to determine the parking space supply change.

3. The method of claim 2, wherein, The prediction is realized by: using a prediction model to perform prediction, wherein the prediction model is a long short-term memory network model, a gated recurrent unit model or an XGBoost model; wherein the reservation parking data is input into the prediction model to obtain the arrival demand of the reserved vehicles; the historical parking records and the real-time parking records are input into the prediction model to obtain the arrival demand of the temporarily parked vehicles; the historical departure data is input into the prediction model to obtain the parking space supply change.

4. The method of claim 1, wherein, The use of the prediction results of the parking demand and the parking space supply change, and the current parking data of the parking lot to determine whether to respond to the entry request of the temporarily parked vehicle comprises: using the prediction results of the parking demand and the parking space supply change to construct a constraint optimization model and solve it to determine the number of temporarily parked vehicles admitted in the current and future preset time period; determining whether the parking lot is saturated according to the number of temporarily parked vehicles admitted and the current parking data of the parking lot, and not responding to the entry request of the temporarily parked vehicle if the parking lot is saturated.

5. The method of claim 4, wherein, The use of the prediction results of the parking demand and the parking space supply change to construct a constraint optimization model and solve it to determine the number of temporarily parked vehicles admitted comprises: based on the predicted parking demand and parking space supply change, constructing a constraint optimization model, wherein the constraint optimization model takes maximizing operating income as an objective function and takes satisfying the parking demand of all reserved vehicles as a constraint condition; solving the constraint optimization model to obtain the optimal number of temporarily parked vehicles admitted in each time period in the future time period; determining the optimal number of temporarily parked vehicles admitted in the current decision-making period as the number of temporarily parked vehicles admitted.

6. The method of claim 5, wherein, The solving of the constraint optimization model to obtain the optimal number of temporarily parked vehicles admitted in each time period in the future time period comprises: constructing multiple simulation conditions, wherein each simulation condition corresponds to a set of predicted parking demand and parking space supply change data; constructing a training data set based on each simulation condition and the constraint optimization model; training a statistical decision model based on the training data set; inputting the prediction results of the parking demand and the parking space supply change into the trained statistical decision model to obtain the number of admitted temporary parking vehicles.

7. The method of claim 6, wherein, The training data set is constructed based on each simulation condition and the constraint optimization model, including: for each simulation condition, obtaining an optimal admission decision of the simulation condition by solving the constraint optimization model; taking the simulation condition and the corresponding optimal admission decision as a set of training data; integrating each set of training data to obtain the training data set.

8. A parking lot vehicle access control apparatus characterized by comprising: The device includes: an admission request receiving module configured to receive an entry request of a temporary parking vehicle to a parking lot; a parking lot data obtaining module configured to obtain current state data and reservation parking data of the parking lot; a data prediction module configured to predict parking demand and parking space supply change of different types of vehicles in a current and future predetermined time period based on the current state data and the reservation parking data; an admission request response module configured to determine whether to respond to the entry request of the temporary parking vehicle by using the prediction results of the parking demand and the parking space supply change and current parking data of the parking lot.

9. A computer device, comprising: The computer device includes: a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the computer device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the parking lot vehicle entry control method in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to perform the steps of the parking lot vehicle entry control method in any one of claims 1 to 7.

Images