Smart city parking lot management method and device based on Internet of Things

Abstract

The invention provides a smart city parking lot management method and device based on the Internet of Things, and is suitable for the technical field of the Internet of Things, and the method comprises the steps: carrying out the coding processing of the dynamic operation information of a parking lot, and obtaining the dynamic operation coding information of the parking lot; performing spatial-temporal feature extraction on the parking lot static attribute information, the parking lot dynamic operation coding information and the smart city traffic flow information to obtain parking lot spatial-temporal correlation feature information; and according to the parking lot space-time correlation feature information and the vehicle user behavior information, generating parking lot parking space distribution information. According to the invention, fusion processing of multi-source heterogeneous cross-modal data is realized, time-space correlation rules of parking lots and urban traffic are deeply analyzed, conditions and real-time dynamic changes of the parking lots are accurately grasped, global scheduling is realized in combination with urban traffic information, the utilization rate of parking spaces is improved, the waiting time of vehicles is shortened, and the efficiency is improved. And the efficient collaborative management level of smart city parking lots is improved.

Description

Technical Field

[0001] This application belongs to the field of Internet of Things (IoT) technology, and in particular relates to a smart city parking management method and device based on IoT. Background Technology

[0002] With the rapid advancement of smart city construction, parking management, as a key component of the urban transportation system, is facing an urgent need for digital transformation. Currently, the application of IoT, big data, and AI technologies has significantly improved the automation level of parking management, but there is still much room for optimization. Multi-source data fusion and adaptability to complex scenarios are still under development, and more efficient and intelligent solutions are urgently needed to meet the ever-increasing urban parking demands.

[0003] In existing technologies, the occupancy status of parking spaces is usually obtained through geomagnetic sensors, and the vehicles entering the parking lot are scheduled in combination with the first-come, first-served parking space scheduling rules; or historical traffic flow data is used to predict parking spaces in a time period, so as to guide vehicles to park according to the predicted parking spaces.

[0004] However, in existing technologies, parking space allocation cannot adapt to sudden traffic conditions or traffic flow changes caused by large-scale events, and it is inflexible when dealing with complex constraints, which easily leads to problems such as imbalance in parking space resource allocation and excessively long vehicle waiting times. Summary of the Invention

[0005] In view of this, the present application provides a smart city parking management method and device based on the Internet of Things, which aims to solve the problems in the prior art that it cannot adapt to parking space management under sudden traffic conditions, and that it has poor flexibility when dealing with complex constraints, thereby reducing parking management efficiency and user experience.

[0006] The first aspect of this application provides a smart city parking management method based on the Internet of Things, comprising:

[0007] Acquire static attribute information of parking lots, dynamic operation information of parking lots, vehicle user behavior information, and smart city traffic information;

[0008] The static attribute information of the parking lot is processed into a structured form to obtain the static attribute structured information of the parking lot.

[0009] The dynamic operation information of the parking lot is encoded to obtain dynamic operation encoding information of the parking lot;

[0010] Based on multiple preset parking lot spatiotemporal feature mapping vectors, spatiotemporal features are extracted from the static attribute structured information of the parking lot, the dynamic operation coding information of the parking lot, and the smart city traffic information to obtain the spatiotemporal correlation feature information of the parking lot.

[0011] Based on the spatiotemporal correlation characteristics of the parking lot and the vehicle user behavior information, parking space allocation information is generated.

[0012] A second aspect of this application provides a smart city parking management device based on the Internet of Things, comprising:

[0013] The parking lot information and traffic information acquisition module is used to acquire static attribute information of parking lots, dynamic operation information of parking lots, vehicle user behavior information, and smart city traffic information.

[0014] The parking lot static attribute structured information generation module is used to perform structured processing on the parking lot static attribute information to obtain parking lot static attribute structured information;

[0015] The parking lot dynamic operation coding information generation module is used to encode the parking lot dynamic operation information to obtain parking lot dynamic operation coding information;

[0016] The parking lot spatiotemporal correlation feature information generation module is used to extract spatiotemporal features from the static attribute structured information, dynamic operation coding information and smart city traffic information of the parking lot based on multiple preset parking lot spatiotemporal feature mapping vectors, so as to obtain parking lot spatiotemporal correlation feature information.

[0017] The parking lot space allocation information determination module is used to generate parking lot space allocation information based on the parking lot spatiotemporal correlation feature information and vehicle user behavior information.

[0018] A third aspect of this application provides a terminal device, which includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the Internet of Things-based smart city parking management method described in the first aspect above.

[0019] Compared with the prior art, the beneficial effects of this application are as follows: This application realizes the fusion processing of multi-source heterogeneous cross-modal data, deeply analyzes the spatiotemporal correlation between parking lots and urban traffic, generates a parking space allocation method by combining vehicle user behavior information, thereby accurately grasping the parking lot conditions and real-time dynamic changes, and realizing global scheduling by combining urban traffic information, so as to improve the utilization rate of parking spaces, shorten vehicle waiting time, and improve the efficient collaborative management level of smart city parking lots. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment 1 of this application;

[0022] Figure 2 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment 2 of this application;

[0023] Figure 3 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment 3 of this application;

[0024] Figure 4 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment 4 of this application;

[0025] Figure 5 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment 5 of this application;

[0026] Figure 6 This is a schematic diagram illustrating the implementation process of the IoT-based smart city parking management method provided in Embodiment Six of this application;

[0027] Figure 7 This is a schematic diagram of the structure of a smart city parking management device based on the Internet of Things provided in an embodiment of this application;

[0028] Figure 8 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0029] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0030] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0031] Figure 1The following is a flowchart illustrating the implementation of the IoT-based smart city parking management method provided in Embodiment 1 of this application, detailed below:

[0032] Step S101: Obtain parking lot static attribute information, parking lot dynamic operation information, vehicle user behavior information, and smart city traffic information.

[0033] In this embodiment, the static attribute information of the parking lot can refer to the inherent attributes of the parking lot that do not change significantly over time, including basic information such as spatial layout, facility configuration and management rules. Spatial geometric data such as the floor plan layout, number of parking spaces, lane direction and entrance / exit location of the parking lot can be obtained through the parking lot's CAD drawings or GIS geographic information system. Three-dimensional structural information such as floor distribution and height can also be confirmed by combining smart city building planning documents. The type, quantity, location and technical parameters of facilities such as charging piles, monitoring equipment and gates can also be collected through Internet of Things sensors such as RFID and cameras. Basic information such as facility models can also be retrieved from the parking lot's equipment management system. It can also include retrieving rule data such as time-limited parking, vehicle type restrictions and payment methods from the parking lot's published fee standard documents and management regulations documents, or reading rule data such as time-limited parking, vehicle type restrictions and payment methods through the parking lot management system. Parking lot dynamic operation information refers to the real-time changing operational status data of the parking lot, reflecting the dynamic situation of parking space usage, vehicle flow, and equipment operation. This can be achieved through parking space detectors, such as geomagnetic sensors and ultrasonic detectors, collecting data on the occupancy status and duration of each parking space in real time. It can also be combined with license plate recognition systems to record the timestamps of vehicles entering and exiting parking spaces, forming a parking space usage log. Data such as vehicle entry and exit time, license plate number, and vehicle type can be obtained through entrance and exit cameras and license plate recognition equipment to statistically analyze traffic flow, queue length, and traffic direction. Surveillance cameras deployed within the parking lot can use video analytics to track vehicle trajectories in real time, calculating parameters such as vehicle speed and traffic density. Furthermore, on-board sensors, such as temperature and current sensors, can collect operating parameters of charging piles, barriers, and monitoring equipment, such as voltage, operating temperature, and switch status. IoT communication protocols, such as MQTT and LoRa, can be used to upload device status data to a management platform for real-time monitoring of equipment malfunctions and energy consumption. Vehicle user behavior information refers to the behavioral preferences, needs, and historical records of vehicle users during the parking process, which can be used to optimize parking space allocation strategies. Vehicle user behavior information can include user registration and reservation data, parking transaction and feedback data, and mobile terminal trajectory data.User registration and reservation data can be collected through parking lot apps or WeChat mini-programs, including user registration information such as frequently used car models, preferred parking areas, reserved parking time, duration, and destination. Historical reservation records can also be analyzed to identify user behavior patterns such as high-frequency parking periods and dwell times. Parking transaction and feedback data can be obtained through payment systems, including payment methods and parking fees. Feedback data on parking space location and facility satisfaction can be collected through user reviews and surveys. This data can also be linked to user historical parking records, such as entry / exit frequency and dwell time, using license plate recognition systems. Mobile terminal trajectory data, with user authorization, can be obtained through an app using GPS location data to analyze user routes and arrival times from their origin to the parking lot, aiding in parking demand prediction. Smart city traffic flow information refers to macro-level data related to traffic flow within the city, used to link parking lot dynamics with the urban transportation network. This can include traffic flow and road condition data, public transportation and travel demand data, and traffic policy and planning data. Traffic flow and road condition data can be obtained through real-time data such as traffic volume, average speed, and congestion index of main roads and intersections from equipment such as checkpoint cameras, microwave radar, and loop detectors used by urban traffic management departments; real-time road condition information, such as road construction and traffic accidents, can be obtained by connecting with navigation platforms such as Gaode and Baidu Maps; public transportation and travel demand data can be obtained through urban travel big data platforms, such as shared bicycle riding data and ride-hailing order data, to understand regional travel demand hotspots and predict potential traffic flow in parking lots; traffic policy and planning data can be obtained by synchronizing with temporary control policies issued by urban traffic management departments, such as tail number restrictions, traffic diversion plans for large-scale events, new road construction plans, and public transportation route adjustments.

[0034] In this embodiment, the static attribute information of the parking lot may include parking lot spatial attribute information, parking lot facility attribute information, and parking rule attribute information. Parking lot spatial attribute information refers to the inherent layout and structural characteristics of the parking lot in physical space, reflecting its spatial geometry and functional zoning, and may include spatial layout data, three-dimensional structural information, and functional zoning data. Spatial layout data can be obtained from CAD drawings or BIM models, including the parking lot's plan outline, parking space arrangement, lane width and direction, and entrance / exit coordinates. Three-dimensional structural information can be extracted from architectural planning documents, including the number of floors, floor heights, and vertical transportation facilities such as elevators and ramps, their locations and parameters. Functional zoning data can be obtained through manual surveying or drawing annotations, defining the location and quantity of functional areas such as accessible parking spaces, charging parking spaces, and regular parking spaces; and recording the spatial distribution of safety facilities such as fire lanes and emergency exits. Parking lot facility attribute information refers to the types, technical parameters, and deployment status of various equipment and facilities within the parking lot, and is a basic attribute that does not dynamically change over time, including equipment type and parameters, facility deployment location, and auxiliary facility information. Equipment type and parameters can be obtained from equipment nameplates or purchase lists, including technical parameters such as model, power, and manufacturer of facilities like charging piles, barriers, surveillance cameras, and lighting fixtures. Facility deployment locations can be marked using RFID tags or GPS positioning devices, such as the specific coordinates of a charging pile next to a parking space, and the installation height and orientation of a camera. Auxiliary facility information can be obtained by collecting data on the locations of auxiliary facilities such as signs, speed bumps, and parking space markings within the parking lot. Parking rule attribute information refers to fixed management regulations, charging standards, and policy restrictions during parking lot operation, which do not change with real-time operating status. This can include charging and management rules, traffic control policies, and user service rules. Charging and management rules can be obtained by reading the parking lot's published price list, including charging standards and free time periods for different time periods and vehicle types. Traffic control policies can be obtained by connecting to the city's traffic management platform, such as dates for vehicle license plate number restrictions and times when high-emission vehicles are prohibited from entering; traffic control plans for special events can be obtained, such as parking lot closure plans around large events; and vehicle type restriction rules within the parking lot can be synchronized, such as no-parking zones for large vehicles. User service rules can be obtained by reading data such as parking reservation rules, cancellation procedures, and exception handling from the parking service agreement or APP documentation. Dynamic parking lot operation information can include parking space usage information, traffic flow information, and the status information of IoT devices in the parking lot. Parking space usage information refers to the occupancy status, usage duration, and vehicle-related data of each parking space in real-time or historical periods, reflecting the dynamic utilization of parking resources. This can include real-time parking space status, vehicle entry and exit records, and usage duration and frequency.Real-time parking space status can be monitored using geomagnetic sensors, ultrasonic detectors, or camera recognition technology to collect data on whether a parking space is occupied. For example, geomagnetic sensors detect changes in the magnetic field when a vehicle is parked, and cameras use image recognition to determine the vacancy status of parking spaces, transmitting the data to the parking management system. Vehicle entry and exit records can be generated using license plate recognition systems to record the timestamps and license plate numbers of vehicles entering and leaving parking spaces, combined with gate opening and closing signals to generate parking space usage logs. For unattended parking lots, the binding relationship between vehicles and parking spaces can be recorded through RFID tags or mobile app scanning. Usage duration and frequency can be calculated based on entry and exit times to determine the occupancy time of a single parking space, statistically analyze parking space turnover rates during peak hours, and analyze patterns in parking space usage frequency through historical data accumulation, such as fluctuations in occupancy rates during weekday morning and evening peak hours, providing a basis for dynamic scheduling. Parking lot traffic flow information refers to real-time data on vehicle movement within the parking lot, including traffic volume, driving trajectories, and queuing conditions, reflecting the dynamic operation status of traffic within the parking lot. This can include entrance and exit traffic volume, internal driving trajectories, and queuing and congestion status. The system includes several key data points: Traffic flow at entrances and exits can be monitored using license plate recognition cameras, infrared sensors, or inductive loops to track the number of vehicles entering and exiting the parking lot per unit time, vehicle type, and direction of entry and exit, generating traffic flow tidal data. Internal driving trajectories can be tracked using surveillance cameras deployed within the parking lot, employing video analytics technologies such as object detection and trajectory tracking algorithms to identify vehicle paths, speeds, and lane-changing behaviors, creating traffic flow heatmaps. For large parking lots, Wi-Fi probes or Bluetooth beacons can be used to locate vehicle positions and optimize route planning. Queuing and congestion status can be monitored using video surveillance or pressure sensors at entrances and exits to detect queue lengths and calculate waiting times. Sensors or cameras can be deployed in parking lot aisles to monitor internal congestion points in real time and trigger indicator lights or broadcast systems to guide traffic flow. Parking lot IoT device status information refers to the real-time operating parameters, fault status, and energy consumption data of various IoT devices within the parking lot, such as charging piles, barriers, and sensors. This information is used for equipment monitoring and maintenance and can include equipment operating parameters, fault and anomaly detection, and energy consumption and maintenance data. The equipment's operating parameters can be collected in real time by built-in sensors, such as current / voltage sensors in charging piles and motor speed sensors in barrier gates. This data is then uploaded to a cloud platform using IoT protocols such as MQTT and LoRa. For example, charging piles can transmit information such as charging power, charging time, and remaining battery power in real time. Fault and anomaly detection can be achieved through equipment status feedback signals, such as error codes when barrier gates malfunction, offline status of cameras, or sensor data thresholds. For instance, if the equipment temperature exceeds a safe range, faults can be automatically identified and alarms generated. Energy consumption and maintenance data can be collected from smart meters and water meters to analyze the power consumption patterns of charging piles and lighting systems. Maintenance records, such as maintenance times and parts replacement logs, can be retrieved from the equipment management system.Parking lot spatial attribute information can include parking lot location coordinates, parking space quantity, parking space type, parking lot access distribution, parking lot entrance / exit location, and parking lot entrance / exit quantity. Parking lot location coordinates refer to the precise coordinates of the parking lot in geographic space, used to locate its position and spatial relationship within the urban road network. The latitude and longitude coordinates of the parking lot's ground entrances / exits can be obtained through satellite positioning systems such as GPS and BeiDou, or the coordinate origin can be extracted from architectural plans or CAD files. Parking space quantity information refers to the total number of various types of parking spaces planned within the parking lot and their distribution across different floors, reflecting the parking lot's basic capacity. The number of parking spaces on each floor can be counted using CAD drawings or BIM models, marking the total number of parking spaces in different areas such as above-ground / underground, indoor / outdoor. Basic parking space data can also be retrieved from the parking management system database, such as the total number of parking spaces entered during system initialization, and the statistical results can be calibrated by combining real-time occupancy data. Parking space type information refers to the functional classification and specifications of parking spaces within the parking lot, such as the differences in type and size between ordinary parking spaces, charging parking spaces, and accessible parking spaces. This information can be obtained from design drawings. Parking lot access distribution information refers to the layout of vehicle traffic paths within the parking lot, including spatial parameters such as lane direction, width, slope, and turning radius. This can be achieved by generating a parking lot point cloud model using 3D laser scanning technology, extracting the spatial coordinates, width, and slope of the access paths, such as the inclination angle of underground parking garage ramps. Parking lot entrance / exit location information refers to the geographical coordinates and spatial orientation of entrances / exits connecting the parking lot to external roads, used for traffic flow guidance and organization. This can be achieved by locating the latitude and longitude of ground markers at entrances / exits using GPS. Parking lot entrance / exit quantity information refers to the total number of entrances / exits connecting the parking lot to the outside, including statistics on motor vehicle entrances / exits, non-motor vehicle entrances / exits, and pedestrian entrances / exits. This information can be extracted from architectural planning documents, retrieved from the parking management system, and verified by combining real-time traffic flow data to confirm the activation status of entrances / exits. Parking facility attribute information may include the number of parking barrier gates, their installation locations, the locations of parking space detection facilities, parking guidance equipment, and parking lighting facilities. Specifically, the number of parking barrier gates refers to the total number of barrier devices used to control vehicle entry and exit and for payment within the parking lot, including entrance and exit barriers and zone barriers. The installation location information for parking barrier gates refers to their specific installation coordinates and spatial relationships within the parking lot, including the road connection locations of entrance and exit barriers and the zone boundary locations of zone barriers within the parking lot.Parking space detection facility location information can refer to the installation location and coverage area of ​​sensors used to detect the occupancy status of parking spaces, such as geomagnetic sensors, cameras, and ultrasonic detectors. Parking guidance equipment location information can refer to the installation location of guidance screens, indicator lights, and other equipment used to indicate the availability of parking spaces, including the spatial distribution of zoned guidance screens and indicator lights in front of parking spaces. Parking lighting facility location information can refer to the installation location and coverage area of ​​lighting fixtures used for parking area illumination within the parking lot, including the distribution of regular lighting and emergency lighting. Parking rule attribute information can include parking fee rules, parking operation time information, and vehicle access restriction information. Specifically, parking fee rules can refer to the charging standards and rules set by the parking lot for different vehicle types, parking periods, and parking space types; parking operation time information can refer to the specific time periods during which the parking lot is open to the public, including daily start and end times, adjustments for special dates, and rules for 24-hour operation or time-segmented opening; vehicle access restriction information can refer to the restrictions set by the parking lot on the type, size, and access rights of vehicles entering, including prohibited vehicle types, size restrictions, and specific vehicle access restrictions. The status information of parking lot IoT devices can include the operational status information of parking lot gates, parking space detection devices, parking guidance devices, and parking lighting devices. Specifically, the operational status information of parking lot gates can refer to the real-time working status of the gate equipment, including gate opening / closing status, operating speed, motor temperature, fault alarms, and power status; the operational status information of parking space detection devices can refer to the working status of the parking space detectors, including online / offline status, signal strength, detection accuracy, battery level, and fault type; the operational status information of parking guidance devices can refer to the working status of devices such as parking guidance screens and indicator lights, including screen display content, brightness, communication connection status, and fault prompts; and the operational status information of parking lighting devices can refer to the working status of the parking lighting fixtures, including on / off status, brightness adjustment, energy consumption data, fault type, and lighting time settings.

[0035] Step S102: The static attribute information of the parking lot is processed into a structured form to obtain the static attribute structured information of the parking lot.

[0036] In this embodiment, the latitude and longitude coordinates of the parking lot location can be obtained first through a satellite positioning system and a GIS platform, and then converted into standard geographic coordinates by combining the coordinate origin of the drawing. The total number and hierarchical distribution of parking spaces can be statistically analyzed from CAD drawings or BIM models. After on-site surveying and synchronization with management system data, calibration can be completed. The functional classification and dimensional parameters of parking space types can be determined based on drawing annotations and on-site markings, and the location data of charging piles and other equipment can be associated. 3D scanning and drawings can be used to extract parameters such as lane direction and width of parking lot access distribution information. Simultaneously, vehicle trajectory data can be used to verify its rationality, and positioning equipment can be employed. By combining the coordinates, orientation, and connection with surrounding roads of the parking lot entrances and exits marked on the drawings, and calibrating with equipment such as cameras, the total number and type of parking lot entrances and exits can be confirmed from planning documents and on-site counts. The system data can also be synchronized to verify their activation status. Finally, the processed parking lot location coordinates, number of parking spaces, parking space type, passage distribution, and entrance / exit location and quantity information can be integrated according to a unified data format, such as JSON or database table structure, clarifying the definition and relationship of each field. This yields structured information on the static attributes of the parking lot, providing a standardized and orderly data foundation for subsequent data processing and decision-making.

[0037] Step S103: Encode the parking lot dynamic operation information to obtain parking lot dynamic operation code information.

[0038] In this embodiment, firstly, for parking space usage information, data such as parking space occupancy status and vehicle entry / exit times are collected in real time using devices such as geomagnetic sensors and cameras. The parking space occupancy status is encoded as 0 / 1, for example, 0 for idle status and 1 for occupied status. This is combined with license plate recognition records to generate a parking space usage log sequence. Then, for parking lot traffic information, data such as traffic flow and vehicle queue length are collected using license plate recognition cameras and inductive loop detectors at entrances and exits. Vehicle types are classified into small cars, large cars, etc., and encoded into corresponding digital tags. Video analytics technology tracks vehicle trajectories and converts them into path codes, which are then used in the parking process. When processing parking lot IoT device status information, operating parameters can be obtained from sensors of devices such as barrier gate controllers, parking space detectors, and parking guidance screens. The barrier gate opening and closing status can be encoded into specific symbols, such as encoding raised gate as 1 and lowered gate as 0. The online / offline status of parking space detectors can be encoded into different values, and the device fault type can be encoded into a preset fault code. Finally, the above-mentioned encoded parking space usage information, traffic flow information, and IoT device status information can be integrated according to a unified data structure to form a standardized encoding sequence, thereby obtaining dynamic operation encoding information of the parking lot, which facilitates subsequent data storage, analysis, and algorithm processing.

[0039] Step S104: Based on multiple preset parking lot spatiotemporal feature mapping vectors, spatiotemporal feature extraction is performed on the static attribute structured information of the parking lot, the dynamic operation coding information of the parking lot, and the smart city traffic information to obtain the spatiotemporal correlation feature information of the parking lot.

[0040] In this embodiment, the multiple preset parking lot spatiotemporal feature mapping vectors can be manually set, or they can be pre-set mathematical vector structures used to map parking lot-related information to spatiotemporal feature dimensions, or they can be designed to take into account spatial characteristics, temporal characteristics, and urban traffic correlation characteristics. This can be achieved by first matching the spatial information such as location coordinates and parking space layout in the static attribute structured information of the parking lot with a preset parking lot spatiotemporal feature mapping vector to determine the relative position of the parking lot in the urban geographic space and its relationship with surrounding roads and buildings. At the same time, features that change over time can be extracted from the parking space usage and traffic flow changes in the dynamic operation coding information of the parking lot. Combined with the preset parking lot spatiotemporal feature mapping vector, the parking space turnover rate and traffic flow peak and trough patterns at different time periods can be analyzed. Then, data such as main road traffic flow and congestion index in the smart city traffic information can be mined to explore the impact trend of urban traffic situation on parking lot traffic flow. Then, the static attribute structured information of the parking lot, the dynamic operation coding information of the parking lot, and the smart city traffic information can be processed for prosperity and deterioration. Through spatiotemporal correlation analysis, the interaction relationship between parking space usage inside the parking lot and urban traffic flow in the spatiotemporal dimension can be determined. For example, how the congestion of surrounding roads affects the queuing time at the parking lot entrance, and the spatiotemporal distribution of parking space demand during weekday morning and evening peak hours, etc., can be obtained to obtain parking lot spatiotemporal correlation feature information that comprehensively reflects the spatiotemporal correlation characteristics of parking lots and urban traffic.

[0041] Step S105: Generate parking space allocation information based on the parking lot spatiotemporal correlation feature information and vehicle user behavior information.

[0042] In this embodiment, the spatiotemporal relationship between the parking lot and urban traffic can be analyzed based on the spatiotemporal correlation feature information of the parking lot. For example, by combining the correlation between the congestion time of surrounding roads and the parking space demand, the parking space usage trend of each area at different time periods can be predicted. If the surrounding main roads are congested during the evening rush hour on weekdays, parking spaces near the exit area may be in high demand. At the same time, vehicle user behavior information can be analyzed to understand user reservation habits, historical parking preferences, such as whether they prefer parking spaces near elevators, whether they are new energy vehicle users, and real-time demand. Then, the spatiotemporal correlation feature information of the parking lot can be combined with the vehicle user behavior information. By combining the spatiotemporal correlation features of parking lots with vehicle user behavior information, an iterative optimization mechanism can be used to generate parking space allocation information. Specifically, an initial allocation scheme library can be built based on historical data and real-time status, simulating different vehicle-parking space matching combinations. Each scheme corresponds to a potential parking space allocation possibility. Then, with the core objective of maximizing parking space utilization and minimizing vehicle waiting time, each scheme in the scheme library is evaluated. During the evaluation process, the static attribute structured information of the parking lot is comprehensively considered. For example, based on parking space type restrictions, charging spaces are prioritized for allocation to new energy vehicles. The travel path distance and time cost from the entrance to each parking space are calculated using channel distribution information to screen out feasible schemes that meet physical space and rule constraints. Then, based on the real-time changes in the dynamic operation coding information of the parking lot, the feasible schemes are dynamically adjusted. For example, when a continuous flow of vehicles leaving a certain area is detected, the system can adjust the schemes accordingly. When parking spaces become available, subsequent vehicles are prioritized for allocation to that area to reduce unnecessary driving within the parking lot. If a passageway experiences temporary congestion, the allocation priority of parking spaces through that passageway is reduced. Simultaneously, the plan is personalized based on vehicle user behavior information. For example, reserved vehicles and high-frequency users are prioritized for allocation to parking spaces in their preferred areas, while temporary parking users are allocated spaces near the exit to shorten departure time. During continuous iterative adjustments, the score of each plan is calculated under the conditions of meeting the objectives and constraints. Plans with higher scores are retained, while those with lower scores are continuously optimized. Finally, the plan with the best score is selected from the optimized plan library to determine the specific parking space number for each vehicle entering the parking lot. Based on the parking lot passageway distribution information, the shortest or optimal guidance path suggestion from the entrance to the parking space is planned, generating comprehensive parking space allocation information that adapts to real-time needs, thus achieving dynamic and intelligent allocation of parking resources.

[0043] The IoT-based smart city parking management method provided in this application embodiment realizes the fusion processing of multi-source heterogeneous cross-modal data, deeply analyzes the spatiotemporal correlation between parking lots and urban traffic, generates a parking space allocation method by combining vehicle user behavior information, thereby accurately grasping the parking lot conditions and real-time dynamic changes, and realizing global scheduling by combining urban traffic information, so as to improve parking space utilization, shorten vehicle waiting time, and improve the efficient and collaborative management level of smart city parking lots.

[0044] Figure 2 The flowchart illustrating the implementation of the IoT-based smart city parking management method provided in Embodiment 2 of this application is shown. The difference between this method and Embodiment 1 is that step S103 specifically includes:

[0045] Step S201: Perform time alignment processing on the parking space usage information, parking traffic flow information, and parking IoT device status information to obtain parking space usage time sequence information, parking traffic flow time sequence information, and parking IoT device status time sequence information.

[0046] In this embodiment, time alignment processing can be achieved by unifying the time base and time granularity. The GPS clock synchronization function of each device in the parking lot can be used to uniformly calibrate the timestamps of the data collected by the parking space detector, license plate recognition camera, and IoT device sensor to the UTC standard time, accurate to the millisecond level. At the same time, the original discrete data can be resampled at a fixed time interval of 1 minute. The vehicle entry and exit records in the parking space usage information, the traffic flow statistics in the traffic flow information, and the parameter collection data in the IoT device status information are integrated into time series with equal time intervals, ensuring that different types of data are comparable in the time dimension. Thus, parking space usage time series information, parking traffic flow time series information, and parking IoT device status time series information are obtained.

[0047] Step S202: Perform logical detection and correction processing on the parking space usage time sequence information, parking traffic flow time sequence information, and parking IoT device status time sequence information to obtain parking space usage correction information, parking traffic flow correction information, and parking IoT device status correction information.

[0048] In this embodiment, for parking space usage time sequence information, if there are no data records for 10 consecutive minutes, it is determined to be data missing, which can be supplemented by using linear interpolation to estimate the occupancy status of the preceding and following time points. If an abnormally frequent switching of "occupied-idle-occupied" is detected for the same parking space within 1 minute, it is considered noise data, and the majority status within that time period is used for correction. For parking lot traffic flow time sequence information, if there is a significant difference between the number of vehicles entering and leaving the entrance and exit statistics over a long period of time, such as when the difference exceeds 10% of the total number of parking spaces, the license plate recognition system log is checked to correct missed or false detection records. If the traffic flow speed exceeds 3 times the parking lot speed limit, the data point is directly removed and marked. Regarding the status and timing information of parking lot IoT devices, when the parameters reported by the device exceed the safety threshold but no alarm is triggered, such as when the temperature of the barrier gate motor exceeds the safety threshold but no alarm is triggered, it is determined to be a data anomaly, and corrections are made in combination with data from adjacent devices and historical operating patterns; if the device status has not been updated for a long time, such as more than 1 hour, it is marked as offline and the fault investigation process is initiated, and finally the corrected parking space usage correction information, parking traffic flow correction information, and parking lot IoT device status correction information are obtained.

[0049] Step S203: Based on the preset parking lot dynamic operation information fusion mapping vector and multiple preset parking lot dynamic fusion weight information, the parking lot space usage correction information, parking lot traffic flow correction information and parking lot IoT device status correction information are fused to obtain the parking lot space traffic flow association fusion vector and the parking lot traffic flow device association fusion vector.

[0050] In this embodiment, the preset parking lot dynamic operation information fusion mapping vector and multiple preset parking lot dynamic fusion weight information can both be manually set. The preset parking lot dynamic operation information fusion mapping vector can be used to define the association dimensions of different data types. For example, parking space usage and traffic flow information are mapped through vehicle entry and exit times and parking space occupancy status, and traffic flow information and equipment status information are mapped through timestamps and spatial locations, such as the relationship between gate location and traffic flow path. Multiple preset parking lot dynamic fusion weight information can be used to quantify the influence of each data type. For example, during peak hours, the weight of traffic flow information is set to 0.5, the weight of parking space usage information is 0.3, and the weight of equipment status information is 0.2; during off-peak hours, the weights are adjusted to 0.3, 0.4, and 0.3. This can be achieved by fusing parking space usage correction information with parking flow correction information to generate a parking lot parking space and traffic flow association fusion vector that reflects the relationship between parking space demand and traffic flow changes; and by fusing parking flow correction information with parking lot IoT device status correction information to generate a parking lot traffic flow and equipment association fusion vector that reflects the impact of traffic flow on equipment operation, thereby realizing multi-dimensional association analysis of dynamic data.

[0051] Step S204: The parking space traffic flow association fusion vector and the parking traffic flow equipment association fusion vector are spliced ​​and format converted to obtain the parking lot dynamic operation coding information.

[0052] In this embodiment, the parking space and traffic flow association fusion vector and the parking traffic flow and equipment association fusion vector can be horizontally concatenated in chronological order to form a comprehensive vector sequence containing multi-source information of parking spaces, traffic flow, and equipment. Then, the data format of the comprehensive vector sequence is converted to a unified standard, such as normalizing numerical data to the [0,1] interval, encoding categorical data into one-hot vectors, such as encoding equipment fault types into one-hot vectors, and encapsulating them in JSON format, adding metadata such as timestamps and parking lot IDs, and finally generating structured parking lot dynamic operation coding information, which is convenient for subsequent feature extraction and calculation.

[0053] The IoT-based smart city parking management method provided in this application can significantly improve the accuracy and relevance of data by strengthening the time alignment, logical verification and deep fusion of dynamic data. This makes the spatiotemporal correlation feature information of the parking lot extracted later more reflective of the real operating rules, effectively reduce decision-making bias caused by data inconsistency or errors, further optimize the utilization rate of parking spaces and vehicle traffic efficiency, and enhance the real-time performance and intelligence level of the smart city parking management system.

[0054] Figure 3 The flowchart illustrating the implementation of the IoT-based smart city parking management method provided in Embodiment 3 of this application is shown. Its difference from Embodiment 2 described above lies in:

[0055] Multiple preset parking lot dynamic fusion weight information includes preset parking lot space vehicle flow dynamic fusion weight information and preset parking lot vehicle flow equipment fusion weight information;

[0056] Step S203 specifically includes:

[0057] Step S301: Perform format transformation processing on the parking space usage correction information, parking traffic flow correction information, and parking IoT device status correction information to obtain parking space usage correction vector, parking traffic flow correction vector, and parking IoT device status correction vector.

[0058] In this embodiment, format transformation processing is used to convert different types of correction information into a unified data representation form to meet the needs of subsequent fusion computing. For parking space usage correction information, data such as parking space occupancy status and vehicle entry / exit time can be converted into numerical vectors. For example, the parking space occupancy status "idle" is encoded as 0 and "occupied" is encoded as 1, and the vehicle entry / exit time is converted into minutes from midnight of the day. For parking traffic flow correction information, data such as traffic flow, vehicle type classification, and driving trajectory are numerated and vectorized. For example, vehicle type classification is mapped to different integer values, and driving trajectory is converted into the encoding of path node sequence. For parking IoT device status correction information, data such as device operating parameters and fault status are normalized to the [0,1] interval. For example, the device temperature is divided by the maximum safe temperature to obtain a normalized value, and the fault type is represented by one-hot encoding. Thus, the parking space usage correction information, parking traffic flow correction information, and parking IoT device status correction information are respectively converted into parking space usage correction vectors, parking traffic flow correction vectors, and parking IoT device status correction vectors with the same dimensional structure to ensure the consistency and compatibility of data formats.

[0059] Step S302: Based on the preset parking lot dynamic operation information fusion window, the parking lot space usage correction vector, parking lot traffic flow correction vector, and parking lot IoT device status correction vector are segmented to obtain multiple parking lot space usage correction sub-vectors, multiple parking lot traffic flow correction sub-vectors, and multiple parking lot IoT device status correction sub-vectors.

[0060] In this embodiment, the preset parking lot dynamic operation information fusion window can be manually set or set to a fixed time length, such as 1 minute. Each parking space usage correction vector, parking lot traffic flow correction vector, and parking lot IoT device status correction vector can be divided into multiple sub-vectors according to time sequence. Specifically, for the parking space usage correction vector, using a 1-minute window, continuous parking space occupancy status and vehicle entry / exit time data can be divided into multiple 1-minute sub-vector sequences; for the parking lot traffic flow correction vector, similarly using a 1-minute window, traffic flow statistics, vehicle type classification, and driving trajectory data can be segmented; for the parking lot IoT device status correction vector, device operating parameters and fault status data can be segmented according to the window time. In this way, multiple parking space usage correction sub-vectors, multiple parking lot traffic flow correction sub-vectors, and multiple parking lot IoT device status correction sub-vectors with the same time span are obtained, facilitating data fusion analysis at the same time scale.

[0061] Step S303: Based on the preset parking lot dynamic operation information fusion mapping vector, the multiple parking lot space usage correction sub-vectors, multiple parking lot traffic flow correction sub-vectors, and multiple parking lot IoT device status correction sub-vectors are mapped and calculated to obtain multiple parking lot space usage sub-vectors to be fused, multiple parking lot traffic flow sub-vectors to be fused, and multiple parking lot IoT device status sub-vectors to be fused.

[0062] In this embodiment, the preset parking space and traffic flow dynamic fusion weight information and the preset parking traffic flow equipment fusion weight information can both be manually set. They can be used to define association rules and transformation relationships between different data types. For example, the parking space occupancy status in the parking space usage correction sub-vector and the traffic flow data in the parking traffic flow correction sub-vector can be mapped through vehicle entry and exit times to calculate the impact coefficient of parking space occupancy changes on traffic flow. The driving trajectory data in the parking traffic flow correction sub-vector and the gate operation status in the parking IoT device status correction sub-vector can be correlated through spatial location to calculate the impact of traffic flow on the pressure and usage frequency of the gate equipment. Thus, multiple parking space usage correction sub-vectors, multiple parking traffic flow correction sub-vectors, and multiple parking IoT device status correction sub-vectors are respectively converted into multiple parking space usage sub-vectors to be fused, multiple parking traffic flow sub-vectors to be fused, and multiple parking IoT device status sub-vectors to be fused, which have physical meaning and association relationships, providing a basis for subsequent weighted summation. It can be that the preset dynamic operation information fusion mapping vector of parking lots is multiplied by multiple parking space usage correction sub-vectors, multiple parking traffic flow correction sub-vectors, and multiple parking IoT device status correction sub-vectors respectively, and the multiplication result is used as multiple parking space usage to be fused sub-vectors, multiple parking traffic flow to be fused sub-vectors, and multiple parking IoT device status to be fused sub-vectors.

[0063] Step S304: Based on the preset dynamic fusion weight information of parking space traffic flow, the multiple parking space traffic flow vectors to be fused and the multiple parking space traffic flow vectors to be fused are weighted and summed to calculate the parking space traffic flow association fusion vector.

[0064] In this embodiment, the preset dynamic fusion weight information for parking space traffic flow can be pre-set according to different time periods and parking lot operating status. For example, during peak hours, a higher weight is given to the parking lot traffic flow sub-vector to be fused, which can be set to 0.6 to highlight the impact of traffic flow changes on parking space allocation, while the weight of the parking space usage sub-vector to be fused is set to 0.4. During off-peak hours, the weight ratio is adjusted appropriately, such as setting it to 0.5 and 0.5. The parking space usage sub-vector to be fused and the parking lot traffic flow sub-vector to be fused for each corresponding time window can be weighted and summed according to the preset dynamic fusion weight information for parking space traffic flow to calculate the parking space traffic flow association fusion vector, which reflects the dynamic relationship between parking space demand and traffic flow changes, thus quantifying the interaction strength and trend of the two in different time windows.

[0065] Step S305: Based on the preset parking lot traffic flow device fusion weight information, the multiple parking lot traffic flow sub-vectors to be fused and the multiple parking lot IoT device status sub-vectors to be fused are weighted and summed to calculate the parking lot traffic flow device association fusion vector.

[0066] In this embodiment, the preset parking lot traffic flow equipment fusion weight information can be set according to the equipment type and the degree of influence of traffic flow on the equipment. For example, for barrier gate equipment, its operating status data is given a higher weight during peak traffic periods, which can be set to 0.7 to emphasize the key role of the barrier gate in traffic flow management. The weight of the parking lot traffic flow sub-vector to be fused can be set to 0.3. For parking space guidance equipment, the weight is adjusted according to its guidance efficiency on traffic flow at different times. The parking lot traffic flow sub-vector to be fused and the parking lot IoT device status sub-vector to be fused in each time window can be weighted and summed according to the parking lot traffic flow equipment fusion weight information to calculate the parking lot traffic flow equipment association fusion vector that reflects the influence of traffic flow on equipment operation, reflecting the dynamic correlation and interaction between equipment status and traffic flow changes.

[0067] The IoT-based smart city parking management method provided in this application is used to achieve deep integration and refined analysis of parking space usage, traffic flow, and equipment status information. It can more accurately capture the dynamic correlation and mutual influence of different data types in the time dimension, so that the generated parking space-traffic flow correlation fusion vector and parking flow-equipment correlation fusion vector have higher spatiotemporal resolution and data representation capabilities. This provides a richer and more accurate data foundation for subsequent extraction of parking lot spatiotemporal correlation feature information, further improves the scientific nature of parking space allocation schemes and the real-time response capability of intelligent management systems, effectively optimizes parking resource allocation and vehicle traffic efficiency, and promotes the development of smart city parking management towards a more refined and intelligent direction.

[0068] Figure 4The following is a flowchart illustrating the implementation of the IoT-based smart city parking management method provided in Embodiment 4 of this application. The difference between this method and Embodiment 1 is that:

[0069] Multiple preset parking lot spatiotemporal feature mapping vectors include preset parking lot spatiotemporal feature focusing mapping vectors, preset parking lot spatiotemporal feature label mapping vectors, and preset parking lot spatiotemporal feature aggregation mapping vectors;

[0070] Step S104 specifically includes:

[0071] Step S401: Based on the static attribute structured information of the parking lot, the preset parking lot spatiotemporal feature focusing mapping vector, the preset parking lot spatiotemporal feature label mapping vector, and the preset parking lot spatiotemporal feature aggregation mapping vector, obtain the parking lot static feature focusing mapping vector, the parking lot static feature label mapping vector, and the parking lot static feature aggregation mapping vector.

[0072] In this embodiment, the preset parking lot spatiotemporal feature focusing mapping vector, the preset parking lot spatiotemporal feature label mapping vector, and the preset parking lot spatiotemporal feature aggregation mapping vector can all be manually set. The pre-defined parking lot spatiotemporal feature focusing mapping vector can be multiplied with the static attribute structured information of the parking lot to extract key spatial and temporal features from the static attribute structured information. It can be used to filter spatial data such as parking lot location coordinates and parking space layout, retaining features highly correlated with urban traffic, such as the distance between the parking lot and the main road, and the orientation of the entrance and exit. The multiplication result is the parking lot static feature focusing mapping vector. The pre-defined parking lot spatiotemporal feature label mapping vector can be multiplied with the static attribute structured information of the parking lot to classify and label the rule information in the static attributes. For example, parking fee rules and vehicle access restriction information can be mapped into label vectors such as "feeding strategy" and "access control". The multiplication result forms the parking lot static feature label mapping vector. The pre-defined parking lot spatiotemporal feature aggregation mapping vector can be multiplied with the static attribute structured information of the parking lot to perform dimensionality reduction and aggregation of spatial distribution and facility configuration data in the static attributes. For example, parking space quantity information and channel distribution information can be aggregated into a comprehensive "spatial capacity" vector. Finally, the parking lot static feature aggregation mapping vector is obtained to achieve multi-dimensional feature extraction of static attributes.

[0073] Step S402: Based on the parking lot dynamic operation coding information, the preset parking lot spatiotemporal feature focusing mapping vector, the preset parking lot spatiotemporal feature label mapping vector, and the preset parking lot spatiotemporal feature aggregation mapping vector, obtain the parking lot dynamic operation feature focusing mapping vector, the parking lot dynamic operation feature label mapping vector, and the parking lot dynamic operation feature aggregation mapping vector.

[0074] In this embodiment, for the dynamic operation coding information of the parking lot, the dynamic operation coding information of the parking lot can be multiplied with a preset parking lot spatiotemporal feature focusing mapping vector to extract key spatiotemporal features from the dynamic data. For example, focusing on the change of parking space occupancy rate during peak hours from parking space usage time series information, and extracting the peak traffic flow at entrances and exits from traffic flow information, etc., can be used to obtain the parking lot dynamic operation feature focusing mapping vector. Alternatively, the dynamic operation coding information of the parking lot can be multiplied with a preset parking lot spatiotemporal feature label mapping vector to obtain the preset parking lot spatiotemporal feature label mapping vector. The status information in the dynamic operation coding information of the parking lot is tagged. For example, the operation status information of the barrier gate and the status information of the parking space detection equipment are mapped into tag vectors such as "equipment operation" and "parking space status", forming a dynamic operation feature tag mapping vector of the parking lot. The dynamic operation coding information of the parking lot can be multiplied with the preset spatiotemporal feature aggregation mapping vector of the parking lot to compress and aggregate the temporal features in the dynamic data. For example, the traffic flow data of different time periods can be aggregated into a comprehensive vector to represent "traffic flow tide", thus obtaining the dynamic operation feature aggregation mapping vector of the parking lot.

[0075] Step S403: Based on the smart city traffic information, the preset parking lot spatiotemporal feature focusing mapping vector, the preset parking lot spatiotemporal feature label mapping vector, and the preset parking lot spatiotemporal feature aggregation mapping vector, obtain the traffic feature focusing mapping vector, the traffic feature label mapping vector, and the traffic feature aggregation mapping vector.

[0076] In this embodiment, the preset parking lot spatiotemporal feature focusing mapping vector, the preset parking lot spatiotemporal feature label mapping vector, and the preset parking lot spatiotemporal feature aggregation mapping vector can all be manually set. The preset parking lot spatiotemporal feature focusing mapping vector can be used to extract traffic features closely related to parking lots, such as the main road congestion index and traffic flow of the surrounding road network, to obtain the traffic flow feature focusing mapping vector. The preset parking lot spatiotemporal feature label mapping vector can be used to classify traffic policies and planning data, for example, mapping policies such as license plate number restrictions and traffic control for large-scale events to label vectors such as "policy restriction" and "temporary control," forming the traffic flow feature label mapping vector. The preset parking lot spatiotemporal feature aggregation mapping vector can be used to aggregate macro-traffic data in a spatiotemporal dimension, such as aggregating traffic flow data from different areas into a comprehensive "urban traffic impact" vector, to obtain the traffic flow feature aggregation mapping vector, thus realizing the feature extraction of urban traffic information.

[0077] Step S404: Perform spatiotemporal interaction calculations based on the parking lot dynamic operation feature focusing mapping vector, parking lot dynamic operation feature label mapping vector, parking lot dynamic operation feature aggregation mapping vector, traffic flow feature focusing mapping vector, traffic flow feature label mapping vector, and traffic flow feature aggregation mapping vector to obtain the parking lot spatiotemporal correlation feature information.

[0078] In this embodiment, the dynamic operation feature focus mapping vector of parking lots can be aligned with the traffic flow feature focus mapping vector in terms of spatiotemporal dimensions to analyze the impact intensity of urban traffic congestion on parking lot traffic flow. The dynamic operation feature label mapping vector of parking lots can be associated with the traffic flow feature label mapping vector to mine association rules such as "main road congestion" and "parking lot entrance queue" labels. By weighted fusion of the dynamic operation feature aggregation mapping vector of parking lots and the traffic flow feature aggregation mapping vector, the impact weight of urban traffic on parking lot space usage at different time periods can be calculated. Finally, the spatiotemporal association feature information of parking lots containing spatiotemporal association strength, feature interaction rules, and influence weight matrix is ​​generated to comprehensively reflect the dynamic interaction relationship between parking lots and urban traffic.

[0079] The IoT-based smart city parking management method provided in this application accurately captures key spatiotemporal features in the data, enhances feature readability through labeling, and uses aggregation mapping to reduce the dimensionality of high-dimensional data. This makes the generated spatiotemporal correlation feature information of parking lots more hierarchical and interpretable, providing a deeper spatiotemporal feature foundation for subsequent generation of parking space allocation information by combining vehicle user behavior information. This effectively improves the smart city parking management system's ability to perceive spatiotemporal dynamic changes and its decision support level, and promotes deep collaboration between parking resource scheduling and urban traffic management.

[0080] Figure 5 This document illustrates a flowchart of the implementation of an IoT-based smart city parking management method provided in Embodiment 5 of this application. The difference between this method and Embodiment 4 above is that step S404 specifically includes:

[0081] Step S501: Based on the parking lot static feature focusing mapping vector, parking lot static feature label mapping vector, parking lot dynamic operation feature focusing mapping vector, and parking lot dynamic operation feature label mapping vector, obtain the parking lot static information feature mapping fusion vector and the parking lot dynamic information feature mapping fusion vector.

[0082] In this embodiment, the static feature focus mapping vector and the static feature label mapping vector of the parking lot can be concatenated dimensionally to obtain a static information feature mapping fusion vector of the parking lot that includes spatial key features and rule labels. The dynamic operation feature focus mapping vector and the dynamic operation feature label mapping vector of the parking lot can be concatenated dimensionally to form a dynamic information feature mapping fusion vector of the parking lot that integrates dynamic spatiotemporal key features and status labels, so as to achieve the preliminary feature fusion of dynamic and static information of the parking lot.

[0083] Step S502: Based on the parking lot static information feature mapping fusion vector, parking lot dynamic information feature mapping fusion vector, parking lot static feature aggregation mapping vector, and parking lot dynamic operation feature aggregation mapping vector, obtain the parking lot static information feature aggregation vector and the parking lot dynamic information feature aggregation vector.

[0084] In this embodiment, the parking lot static information feature mapping fusion vector and the parking lot static feature aggregation mapping vector can be weighted and fused. The weights can be set according to the parking lot's operating hours. For example, during peak hours, the aggregation mapping vector is given a higher weight to highlight the spatial capacity feature, resulting in a parking lot static information feature aggregation vector that comprehensively reflects the static spatial and rule features. The parking lot dynamic information feature mapping fusion vector and the parking lot dynamic operation feature aggregation mapping vector can be weighted and fused. For example, during off-peak hours, the weight of the dynamic feature mapping fusion vector can be increased to obtain a parking lot dynamic information feature aggregation vector that integrates dynamic spatiotemporal features and aggregation features, thereby achieving further aggregation of features.

[0085] Step S503: The static information feature aggregation vector and the dynamic information feature aggregation vector of the parking lot are fused to obtain the static and dynamic information feature fusion vector of the parking lot.

[0086] In this embodiment, after aligning the aggregated vector of static information features of parking lots with the aggregated vector of dynamic information features of parking lots in terms of spatiotemporal dimensions, a feature cross-fusion algorithm can be used to calculate the association weights between static spatial features such as the number of parking spaces and the location of entrances and exits and dynamic operational features such as the parking space occupancy rate and traffic flow. This generates a fusion vector of static and dynamic information features of parking lots that includes spatiotemporal interaction relationships, which can be used to reflect the spatiotemporal association between the static attributes and dynamic operation of parking lots.

[0087] Step S504: Calculate the parking lot dynamic and static traffic flow feature focus mapping vector based on the parking lot dynamic and static information feature fusion vector and the traffic flow feature focus mapping vector.

[0088] In this embodiment, the dot product operation can be performed on the fusion vector of dynamic and static information features of parking lots and the focused mapping vector of traffic flow features to extract the correlation strength between the dynamic and static features of parking lots and key features of urban traffic. For example, the correlation between the location information of parking lot entrances and exits and the traffic flow of main roads can be calculated. The result of the dot product operation is used as the focused mapping vector of dynamic and static traffic flow features of parking lots to highlight the key correlation between dynamic and static features and traffic flow.

[0089] Step S505: Calculate the parking lot dynamic and static traffic flow feature label mapping vector based on the parking lot dynamic and static information feature fusion vector and the traffic flow feature label mapping vector.

[0090] In this embodiment, the dynamic and static information feature fusion vector of the parking lot can be matched with the traffic flow feature label mapping vector to establish association rules such as "parking space shortage during peak hours" and "congestion on urban main roads". The dynamic and static traffic flow feature label mapping vector of the parking lot can be generated through the label mapping relationship in the vector space to realize cross-domain association of feature labels.

[0091] Step S506: Calculate the parking lot dynamic and static traffic flow feature interaction mapping vector based on the parking lot dynamic and static traffic flow feature focusing mapping vector and the parking lot dynamic and static traffic flow feature label mapping vector.

[0092] In this embodiment, a tensor product operation can be performed on the dynamic and static traffic flow feature focus mapping vector of the parking lot and the dynamic and static traffic flow feature label mapping vector of the parking lot to construct a spatiotemporal feature interaction matrix. The interaction intensity of key features under different labels can be analyzed, such as the interaction relationship between the traffic flow at the entrance and exit of the parking lot and the congestion index of the main road under the "policy restriction" label. The dynamic and static traffic flow feature interaction mapping vector of the parking lot can be obtained to comprehensively characterize the interaction relationship between dynamic and static features and traffic flow.

[0093] Step S507: Based on the interaction mapping vector of dynamic and static traffic flow features of the parking lot and the aggregation mapping vector of traffic flow features, the spatiotemporal correlation feature information of the parking lot is obtained.

[0094] In this embodiment, the dynamic and static traffic flow feature interaction mapping vector of the parking lot can be weighted and fused with the traffic flow feature aggregation mapping vector. The weight is dynamically adjusted according to the time window. For example, the weight of the traffic flow feature aggregation mapping vector is increased during weekday morning and evening peak hours. Finally, the spatiotemporal correlation feature information of the parking lot containing dynamic and static feature interaction and traffic flow influence is generated to realize the deep spatiotemporal correlation analysis between the parking lot and urban traffic.

[0095] The IoT-based smart city parking management method provided in this application can achieve deep integration of parking lot static attribute information, dynamic operation information and urban traffic information. By introducing the fusion of dynamic and static information features and the multi-layer interaction of traffic flow features, it can more comprehensively capture the dynamic relationship between parking lots and urban traffic in the spatiotemporal dimension. This makes the generated parking lot spatiotemporal correlation feature information not only include single-dimensional feature extraction, but also cover the interaction relationship between dynamic and static features and the multi-dimensional mapping with urban traffic. This provides a more accurate spatiotemporal feature basis for subsequent parking space allocation and effectively improves the adaptability of the smart city parking management system to complex traffic environments and the level of intelligent resource scheduling.

[0096] Figure 6 The flowchart illustrating the implementation of the IoT-based smart city parking management method provided in Embodiment Six of this application is shown. The difference between this method and Embodiment One is that step S105 specifically includes:

[0097] Step S601: Based on the preset vehicle user behavior feature extraction vector, perform feature extraction processing on the vehicle user behavior information to obtain vehicle user behavior feature information.

[0098] In this embodiment, the preset vehicle user behavior feature extraction vector can be manually set based on the characteristics of user behavior data, and can include feature extraction rules from multiple dimensions such as user registration and reservation data, parking transaction and feedback data, and mobile terminal trajectory data. By performing operations on the vehicle user behavior information and this vector, features such as commonly used vehicle models and preferred parking areas are extracted from user registration information, payment habits and parking duration patterns are mined from parking transaction data, and travel routes and arrival time patterns are analyzed from mobile terminal trajectory data, thereby obtaining comprehensive and representative vehicle user behavior feature information.

[0099] Step S602: Generate parking lot user behavior association features based on the parking lot spatiotemporal association feature information and vehicle user behavior feature information.

[0100] In this embodiment, the spatiotemporal correlation characteristics of parking lots can be analyzed in conjunction with vehicle user behavior characteristics. This involves combining the parking space usage trends and the impact of surrounding traffic on the parking lot reflected in the spatiotemporal correlation characteristics with the reservation habits and parking preferences in the vehicle user behavior characteristics. For example, if the spatiotemporal correlation characteristics of parking lots show high demand for parking spaces near the exit during weekday evening rush hour, and at the same time, the vehicle user behavior characteristics show that some users prefer to leave quickly, these two types of information can be merged to generate parking lot user behavior correlation characteristics that reflect the correlation between user behavior and the spatiotemporal state of the parking lot.

[0101] Step S603: Based on the parking lot user behavior association features and the preset parking lot spatiotemporal feature mapping weight vector, obtain the parking lot user behavior spatiotemporal association mapping features.

[0102] In this embodiment, the preset parking lot spatiotemporal feature mapping weight vector can be manually set. It can be used to quantify the influence of parking lot spatiotemporal correlation features on user behavior correlation features. The parking lot user behavior correlation features can be multiplied with the preset parking lot spatiotemporal feature mapping weight vector, and corresponding weights can be assigned according to the importance of different spatiotemporal features. For example, during peak hours, the weight of spatiotemporal features related to traffic congestion is increased, so that the mapping of parking lot user behavior correlation features in the spatiotemporal dimension is more in line with the actual situation, thereby obtaining the parking lot user behavior spatiotemporal correlation mapping features to highlight the influence of spatiotemporal factors on user behavior correlation features.

[0103] Step S604: Generate parking lot user behavior spatiotemporal correlation displacement features based on the parking lot user behavior spatiotemporal correlation mapping features and the preset parking lot user behavior correlation feature mapping displacement vector.

[0104] In this embodiment, the preset parking lot user behavior correlation feature mapping displacement vector can be manually set, and can be a parameter vector used to adjust the direction and amplitude of the parking lot user behavior spatiotemporal correlation mapping feature. The features can be dynamically adjusted according to actual needs and historical data by adding the parking lot user behavior spatiotemporal correlation mapping feature to the preset parking lot user behavior correlation feature mapping displacement vector. For example, when a deviation is found in a certain type of user behavior under specific spatiotemporal conditions, the displacement vector is used to correct the features, generating a parking lot user behavior spatiotemporal correlation displacement feature that more accurately reflects the spatiotemporal relationship between user behavior and the parking lot.

[0105] Step S605: Determine whether the number of times the spatiotemporal correlation displacement features of parking lot user behavior are generated is less than the preset threshold for the number of times the spatiotemporal correlation displacement features of parking lot user behavior are generated; if yes, proceed to step S606; if no, proceed to step S607.

[0106] In this embodiment, the preset threshold for the number of times the spatiotemporal correlation displacement features of parking lot user behavior are generated can be set manually and used to control the number of iterative optimizations. When the number of times the spatiotemporal correlation displacement features of parking lot user behavior are generated is less than the threshold, it indicates that the currently generated features may not be sufficiently optimized and there is room for further improvement. Therefore, iterative optimization is needed to obtain a more accurate feature representation.

[0107] Step S606: Use the spatiotemporal correlation displacement features of parking lot user behavior as parking lot user behavior correlation features, and return to step S603.

[0108] In this embodiment, the currently generated parking lot user behavior spatiotemporal correlation displacement features can be used as new parking lot user behavior correlation features. Then, the calculation of the mapping weight vector with the preset parking lot spatiotemporal features and subsequent processing are performed again. Through continuous iteration, the parking lot user behavior correlation features are gradually optimized, so that the generated feature information is more in line with the actual user behavior and parking lot spatiotemporal state.

[0109] Step S607: Decode the spatiotemporal correlation displacement features of the parking lot user behavior to generate parking space allocation information.

[0110] In this embodiment, the spatiotemporal correlation displacement features of parking lot user behavior, which have undergone multiple iterations of optimization, can be converted into practically usable parking space allocation information. Based on the user behavior preferences, parking lot spatiotemporal status, and other information contained in the features, combined with the static attribute structured information and dynamic operation coding information of the parking lot, the abstract features can be transformed into specific parking space allocation schemes, determining the parking space number of each vehicle entering the parking lot, and planning the guidance path from the entrance to the parking space.

[0111] The IoT-based smart city parking management method provided in this application can achieve accurate modeling of the spatiotemporal relationship between user behavior and parking lots, so that the generated parking space allocation information fully considers the personalized needs of users and the dynamic spatiotemporal changes of parking lots, effectively improving the rationality of parking space allocation and user parking experience, further enhancing the intelligence and humanization level of the smart city parking management system, and providing strong support for the efficient allocation of urban transportation resources.

[0112] Corresponding to the method in the above embodiments, Figure 7 The diagram shows a structural block diagram of a smart city parking management device based on the Internet of Things provided in this application embodiment. For ease of explanation, only the parts related to the embodiments of this application are shown. Figure 7 The example IoT-based smart city parking management device can be the implementing entity of the IoT-based smart city parking management method provided in the aforementioned embodiment 1.

[0113] Reference Figure 7 The IoT-based smart city parking management device includes:

[0114] The parking lot information and traffic information acquisition module 710 is used to acquire static attribute information of parking lots, dynamic operation information of parking lots, vehicle user behavior information, and smart city traffic information.

[0115] The parking lot static attribute structured information generation module 720 is used to perform structured processing on the parking lot static attribute information to obtain parking lot static attribute structured information.

[0116] The parking lot dynamic operation coding information generation module 730 is used to encode the parking lot dynamic operation information to obtain parking lot dynamic operation coding information;

[0117] The parking lot spatiotemporal correlation feature information generation module 740 is used to extract spatiotemporal features from the static attribute structured information, dynamic operation coding information and smart city traffic information of the parking lot based on multiple preset parking lot spatiotemporal feature mapping vectors, so as to obtain parking lot spatiotemporal correlation feature information.

[0118] The parking lot space allocation information determination module 750 is used to generate parking lot space allocation information based on the parking lot spatiotemporal correlation feature information and vehicle user behavior information.

[0119] The process by which each module in the IoT-based smart city parking management device provided in this application implements its respective function can be specifically referred to the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0120] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0121] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0122] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0123] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0124] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0125] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0126] The IoT-based smart city parking management method provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, in-vehicle devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, and personal digital assistants (PDAs). This application does not impose any restrictions on the specific type of terminal device.

[0127] For example, the terminal device may be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a set-top box (STB), customer premises equipment (CPE), and / or other devices used for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved Public Land Mobile Network (PLMN) networks.

[0128] As an example and not a limitation, when the terminal device is a wearable device, the term "wearable device" can also refer to any device that utilizes wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices worn directly on the body or integrated into a user's clothing or accessories. Wearable devices are not merely hardware devices; they achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined, wearable smart devices include those with comprehensive functions, large sizes, and the ability to perform complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those focused on a specific application function that require interaction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0129] Figure 8 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 8 As shown, the terminal device 8 in this embodiment includes: at least one processor 80 ( Figure 8 Only one is shown in the image), and a memory 81 is stored in which a computer program 82 that can run on the processor 80 is stored. When the processor 80 executes the computer program 82, it implements the steps in the various embodiments of the IoT-based smart city parking management method described above, for example... Figure 1 Steps S101 to S105 are shown. Alternatively, when the processor 80 executes the computer program 82, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 8The functions of modules 810 to 850 are shown.

[0130] The terminal device 8 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art will understand that... Figure 8 This is merely an example of terminal device 8 and does not constitute a limitation on terminal device 8. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmission devices, network access devices, buses, etc.

[0131] The processor 80 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0132] In some embodiments, the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. The memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 8. Furthermore, the memory 81 may include both internal and external storage units of the terminal device 8. The memory 81 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 81 can also be used to temporarily store data that has been sent or will be sent.

[0133] Furthermore, the functional units in the various embodiments of 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0134] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0135] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0136] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0137] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0138] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0139] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0140] 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.

[0141] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications 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, and should all be included within the protection scope of this application.

Claims

1. A smart city parking lot management method based on Internet of Things, characterized in that, The method comprises the following steps: acquiring parking lot static attribute information, parking lot dynamic operation information, vehicle user behavior information, and smart city vehicle flow traffic information; structurally processing the parking lot static attribute information to obtain parking lot static attribute structured information; encoding processing the parking lot dynamic operation information to obtain parking lot dynamic operation encoding information; based on a plurality of preset parking lot space-time feature mapping vectors, extracting space-time features from the parking lot static attribute structured information, the parking lot dynamic operation encoding information, and the smart city vehicle flow traffic information to obtain parking lot space-time correlation feature information; generating parking lot space allocation information according to the parking lot space-time correlation feature information and the vehicle user behavior information.

2. The smart city parking lot management method based on the Internet of Things according to claim 1, wherein the parking lot static attribute information comprises parking lot space attribute information, parking lot facility attribute information, and parking rule attribute information; the parking lot dynamic operation information comprises parking lot space usage information, parking lot vehicle flow information, and parking lot Internet of Things device state information. 3.The IoT-based smart city parking lot management method of claim 2, wherein, The step of encoding processing the parking lot dynamic operation information to obtain parking lot dynamic operation encoding information comprises the following steps: time aligning the parking lot space usage information, the parking lot vehicle flow information, and the parking lot Internet of Things device state information to obtain parking lot space usage time sequence information, parking lot vehicle flow time sequence information, and parking lot Internet of Things device state time sequence information; logically detecting and correcting the parking lot space usage time sequence information, the parking lot vehicle flow time sequence information, and the parking lot Internet of Things device state time sequence information to obtain parking lot space usage correction information, parking lot vehicle flow correction information, and parking lot Internet of Things device state correction information; based on a preset parking lot dynamic operation information fusion mapping vector and a plurality of preset parking lot dynamic fusion weight information, fusing the parking lot space usage correction information, the parking lot vehicle flow correction information, and the parking lot Internet of Things device state correction information to obtain a parking lot space-vehicle flow correlation fusion vector and a parking lot vehicle flow-device correlation fusion vector; splicing and format converting the parking lot space-vehicle flow correlation fusion vector and the parking lot vehicle flow-device correlation fusion vector to obtain the parking lot dynamic operation encoding information.

4. The smart city parking lot management method based on the Internet of Things according to claim 3, wherein the plurality of preset parking lot dynamic fusion weight information comprises preset parking lot space-vehicle flow dynamic fusion weight information and preset parking lot vehicle flow-device fusion weight information; the step of fusing the parking lot space usage correction information, the parking lot vehicle flow correction information, and the parking lot Internet of Things device state correction information based on the preset parking lot dynamic operation information fusion mapping vector and the plurality of preset parking lot dynamic fusion weight information to obtain the parking lot space-vehicle flow correlation fusion vector and the parking lot vehicle flow-device correlation fusion vector comprises the following steps: The parking lot usage correction information, the parking lot traffic correction information, and the parking lot Internet of Things device state correction information are format-converted to obtain a parking lot usage correction vector, a parking lot traffic correction vector, and a parking lot Internet of Things device state correction vector; According to the preset parking lot dynamic operation information fusion window, the parking lot usage correction vector, the parking lot traffic correction vector, and the parking lot Internet of Things device state correction vector are segmented to obtain a plurality of parking lot usage correction sub-vectors, a plurality of parking lot traffic correction sub-vectors, and a plurality of parking lot Internet of Things device state correction sub-vectors; Based on the preset parking lot dynamic operation information fusion mapping vector, the plurality of parking lot usage correction sub-vectors, the plurality of parking lot traffic correction sub-vectors, and the plurality of parking lot Internet of Things device state correction sub-vectors are mapped and calculated to obtain a plurality of parking lot usage to-be-fused sub-vectors, a plurality of parking lot traffic to-be-fused sub-vectors, and a plurality of parking lot Internet of Things device state to-be-fused sub-vectors; Based on the preset parking lot dynamic operation information fusion mapping vector, the plurality of parking lot usage correction sub-vectors, the plurality of parking lot traffic correction sub-vectors, and the plurality of parking lot Internet of Things device state correction sub-vectors are mapped and calculated to obtain a plurality of parking lot usage to-be-fused sub-vectors, a plurality of parking lot traffic to-be-fused sub-vectors, and a plurality of parking lot Internet of Things device state to-be-fused sub-vectors; Based on the preset parking lot dynamic operation information fusion mapping vector, the plurality of parking lot usage correction sub-vectors, the plurality of parking lot traffic correction sub-vectors, and the plurality of parking lot Internet of Things device state correction sub-vectors are mapped and calculated to obtain a plurality of parking lot usage to-be-fused sub-vectors, a plurality of parking lot traffic to-be-fused sub-vectors, and a plurality of parking lot Internet of Things device state to-be-fused sub-vectors.

5. The Internet of Things-based smart city parking lot management method of claim 1, wherein The plurality of preset parking lot space-time feature mapping vectors include a preset parking lot space-time feature focusing mapping vector, a preset parking lot space-time feature label mapping vector, and a preset parking lot space-time feature aggregation mapping vector; The step of extracting space-time features based on the plurality of preset parking lot space-time feature mapping vectors from the parking lot static attribute structured information, the parking lot dynamic operation encoding information, and the smart city traffic information to obtain parking lot space-time correlation feature information specifically includes: According to the parking lot static attribute structured information, the preset parking lot space-time feature focusing mapping vector, the preset parking lot space-time feature label mapping vector, and the preset parking lot space-time feature aggregation mapping vector, a parking lot static feature focusing mapping vector, a parking lot static feature label mapping vector, and a parking lot static feature aggregation mapping vector are obtained; According to the parking lot dynamic operation encoding information, the preset parking lot space-time feature focusing mapping vector, the preset parking lot space-time feature label mapping vector, and the preset parking lot space-time feature aggregation mapping vector, a parking lot dynamic operation feature focusing mapping vector, a parking lot dynamic operation feature label mapping vector, and a parking lot dynamic operation feature aggregation mapping vector are obtained; According to the smart city traffic information, a preset parking lot space-time feature focusing mapping vector, a preset parking lot space-time feature label mapping vector, and a preset parking lot space-time feature aggregation mapping vector, a traffic flow feature focusing mapping vector, a traffic flow feature label mapping vector, and a traffic flow feature aggregation mapping vector are obtained; According to the parking lot dynamic operation feature focusing mapping vector, the parking lot dynamic operation feature label mapping vector, the parking lot dynamic operation feature aggregation mapping vector, the traffic flow feature focusing mapping vector, the traffic flow feature label mapping vector, and the traffic flow feature aggregation mapping vector, space-time interaction calculation is performed to obtain parking lot space-time correlation feature information. 6.The IoT-based smart city parking lot management method of claim 5, wherein, The step of performing space-time interaction calculation according to the parking lot dynamic operation feature focusing mapping vector, the parking lot dynamic operation feature label mapping vector, the parking lot dynamic operation feature aggregation mapping vector, the traffic flow feature focusing mapping vector, the traffic flow feature label mapping vector, and the traffic flow feature aggregation mapping vector to obtain parking lot space-time correlation feature information specifically includes: According to the parking lot static feature focusing mapping vector, the parking lot static feature label mapping vector, the parking lot dynamic operation feature focusing mapping vector, and the parking lot dynamic operation feature label mapping vector, a parking lot static information feature mapping fusion vector and a parking lot dynamic information feature mapping fusion vector are obtained; According to the parking lot static information feature mapping fusion vector, the parking lot dynamic information feature mapping fusion vector, the parking lot static feature aggregation mapping vector, and the parking lot dynamic operation feature aggregation mapping vector, a parking lot static information feature aggregation vector and a parking lot dynamic information feature aggregation vector are obtained; The parking lot static information feature aggregation vector and the parking lot dynamic information feature aggregation vector are fused to obtain a parking lot dynamic and static information feature fusion vector; According to the parking lot dynamic and static information feature fusion vector and the traffic flow feature focusing mapping vector, a parking lot dynamic and static traffic flow feature focusing mapping vector is calculated; According to the parking lot dynamic and static information feature fusion vector and the traffic flow feature label mapping vector, a parking lot dynamic and static traffic flow feature label mapping vector is calculated; According to the parking lot dynamic and static traffic flow feature focusing mapping vector and the parking lot dynamic and static traffic flow feature label mapping vector, a parking lot dynamic and static traffic flow feature interaction mapping vector is calculated; According to the parking lot dynamic and static traffic flow feature interaction mapping vector and the traffic flow feature aggregation mapping vector, parking lot space-time correlation feature information is obtained. 7.The IoT-based smart city parking lot management method of claim 1, wherein, The step of generating parking lot space allocation information according to the parking lot space-time correlation feature information and vehicle user behavior information specifically includes: According to a preset vehicle user behavior feature extraction vector, feature extraction processing is performed on the vehicle user behavior information to obtain vehicle user behavior feature information; According to the parking lot space-time correlation feature information and the vehicle user behavior feature information, parking lot user behavior correlation features are generated; According to the parking lot user behavior association feature and the preset parking lot space-time feature mapping weight vector, a parking lot user behavior space-time association mapping feature is obtained; According to the parking lot user behavior space-time association mapping feature and the preset parking lot user behavior association feature mapping displacement vector, a parking lot user behavior space-time association displacement feature is generated; Determine whether the generation number of the parking lot user behavior space-time association displacement feature is less than the preset parking lot user behavior space-time association displacement feature generation number threshold; If yes, the parking lot user behavior space-time association displacement feature is taken as the parking lot user behavior association feature, and the step of obtaining the parking lot user behavior space-time association mapping feature according to the parking lot user behavior association feature and the preset parking lot space-time feature mapping weight vector is returned to; If no, the parking lot user behavior space-time association displacement feature is decoded and processed to generate parking lot parking space allocation information.

8. The Internet of Things-based smart city parking lot management method of claim 2, wherein the parking lot space attribute information comprises parking lot location coordinate information, parking lot parking space quantity information, parking lot parking space type information, parking lot passage distribution information, parking lot entrance location information, and parking lot entrance quantity information; the parking lot facility attribute information comprises parking lot barrier gate quantity information, parking lot barrier gate quantity information installation location information, parking space detection facility location information, parking space guidance device location information, and parking space lighting facility location information; the parking rule attribute information comprises parking charging rule information, parking long operation time information, and vehicle access restriction information; and the parking lot Internet of Things device state information comprises parking lot barrier gate operation state information, parking lot parking space detection device operation state information, parking lot parking space guidance device operation state information, and parking space lighting device operation state information. Comprise: a parking lot information and vehicle flow traffic information acquisition module for acquiring parking lot static attribute information, parking lot dynamic operation information, vehicle user behavior information, and smart city vehicle flow traffic information; a parking lot static attribute structured information generation module for structuring the parking lot static attribute information to obtain parking lot static attribute structured information; a parking lot dynamic operation encoding information generation module for encoding the parking lot dynamic operation information to obtain parking lot dynamic operation encoding information; 9. An Internet of Things-based smart city parking lot management device, characterized by, a parking lot space-time association feature information generation module for extracting space-time features of the parking lot static attribute structured information, the parking lot dynamic operation encoding information, and the smart city vehicle flow traffic information based on a plurality of preset parking lot space-time feature mapping vectors to obtain parking lot space-time association feature information; a parking lot parking space allocation information determination module for generating parking lot parking space allocation information according to the parking lot space-time association feature information and the vehicle user behavior information. ​ ​ ​ ​ 10. A terminal device, comprising: The terminal device comprises a memory and a processor, the memory stores a computer program capable of running on the processor, and the processor implements the steps of the method according to any one of claims 1 to 8 when executing the computer program.

Images