Intelligent parking real-time monitoring system based on edge calculation

Abstract

The invention relates to the technical field of parking lot management, in particular to an intelligent parking real-time monitoring system based on edge calculation, which comprises a signal stability calibration module, a parking flow monitoring module, a regional load calculation module, a vehicle guide adjustment module and a task scheduling optimization module. According to the invention, by recording the signal amplitude, the time sequence change rate and the signal fluctuation range, abnormal fluctuation data screening, signal stability enhancement, environmental interference influence reduction, standby infrared sensor time window adjustment, deviation over-limit data correction, signal compensation capability improvement and detection accuracy ensuring are realized; the method comprises the following steps: extracting traffic flow, parking space occupation data and parking time, optimizing short-time releasable parking space evaluation, adjusting parking space recommendation weight in a high-load area, optimizing path distribution, improving guidance rationality, reducing congestion, combining task execution time, influence range and resource priority, optimizing task scheduling, improving parking management fineness, and improving parking management efficiency. And dynamic monitoring and efficient guiding are realized.

Description

Technical Field

[0001] This invention relates to the field of parking management technology, and in particular to a smart parking real-time monitoring system based on edge computing. Background Technology

[0002] The field of parking management technology encompasses the planning and design of parking lots, vehicle guidance, parking space management, fee management, and the development and application of related monitoring systems. The core of this technology involves how to efficiently allocate parking resources, reduce the time vehicles spend searching for parking spaces, and improve the operational efficiency of parking lots. Traditional parking management mainly relies on manual management or simple electronic time-based charging systems. With the development of the Internet of Things, sensor technology, and automatic control systems, parking management is gradually moving towards automation and intelligence. Current parking management technologies primarily use video surveillance, ultrasonic sensors, and geomagnetic detection to obtain parking space occupancy information, and transmit this data to the management system via wireless communication technology to achieve functions such as vehicle guidance, parking space monitoring, and intelligent fee collection.

[0003] The intelligent parking real-time monitoring system refers to a technical system based on edge computing architecture that collects, analyzes, and feeds back real-time data on the usage status of each parking space in a parking lot. The system mainly covers technical aspects such as parking space status monitoring, data calculation and transmission, parking guidance, and information dissemination. The system collects parking space occupancy information through ultrasonic sensors, geomagnetic detectors, or cameras, and performs data analysis at local edge computing nodes to reduce data transmission latency and reduce the cloud computing burden. The calculated data is transmitted to the information processing platform via wireless communication technology and synchronized to the parking lot information display equipment to realize parking guidance. The system can also interact with mobile application terminals through wireless communication protocols to provide users with real-time parking information, improving the convenience and efficiency of parking management.

[0004] Existing technologies primarily rely on video surveillance, ultrasonic sensors, and geomagnetic detection to obtain parking space occupancy information, which is then transmitted to the management system via wireless communication. However, these technologies have limitations in data processing. Due to a lack of precise filtering and correction for signal fluctuations, sensors are susceptible to interference in complex environments, leading to a high rate of misjudgment and affecting the accuracy of parking space occupancy information. Existing technologies mainly focus on static monitoring of parking space occupancy status, lacking in-depth analysis of parking flow trends. This makes it difficult to predict changes in parking supply and demand during peak hours, reducing the flexibility of parking management. Parking guidance methods are relatively fixed and cannot dynamically adjust parking space recommendation strategies based on real-time load conditions in different areas, easily causing some areas to be overloaded while resources are not fully utilized. Existing task scheduling is mostly based on fixed rules, failing to fully consider task execution time, impact scope, and resource allocation priorities. This results in emergency tasks not being prioritized due to scheduling delays, affecting the overall efficiency of parking management. The lack of accurate assessment of short-term available parking spaces causes lag in parking space scheduling, affecting parking turnover speed and further exacerbating congestion in high-load areas. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a smart parking real-time monitoring system based on edge computing.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A smart parking real-time monitoring system based on edge computing includes:

[0007] The signal stabilization calibration module uses signal data from geomagnetic and ultrasonic sensors in the parking garage to filter abnormal fluctuation data, adjust the time window of the backup infrared sensor signal, analyze the influence of ambient temperature and magnetic field interference, adjust the signal compensation coefficient, and obtain the sensor signal correction value.

[0008] Based on the sensor signal correction value, the parking flow monitoring module organizes the vehicle entry and exit data of the parking lot entrance and exit, counts the number of vehicles entering, compares the parking space data of the parking area, and generates a parking flow trend value.

[0009] The area load calculation module compares the parking flow trend value, extracts the traffic flow per unit time and parking space occupancy data of the parking area, filters the vehicle stay time data, estimates the number of parking spaces that can be released in a short time, and obtains the parking area load coefficient.

[0010] The vehicle guidance adjustment module filters the parking area load coefficient, adjusts the recommended weight of parking spaces in high-load areas, modifies vehicle navigation path data, and optimizes path allocation in low-load areas to obtain parking guidance adjustment parameters.

[0011] The task scheduling optimization module filters the parking guidance adjustment parameters, analyzes the remaining execution time of the task, compares the task's impact range data, adjusts the task's urgency, and generates dynamic monitoring results of the parking status.

[0012] As a further aspect of the present invention, the sensor signal correction values ​​include geomagnetic signal correction values, ultrasonic signal correction values, infrared signal correction values, and signal compensation correction coefficients; the parking flow trend values ​​include vehicle entry volume, parking space occupancy rate, and entrance / exit flow data; the parking area load coefficient includes vehicle flow per unit time, parking space occupancy data, and the number of parking spaces available for short-term release; the parking guidance adjustment parameters include parking space recommendation weight, navigation path adjustment, and path allocation optimization; and the parking status dynamic monitoring results include task urgency, task impact range, and remaining task time.

[0013] As a further aspect of the present invention, the acquisition step of the signal stabilization calibration module specifically includes:

[0014] The signal fluctuation screening submodule calculates the gradient change value between adjacent periods based on the signal data of geomagnetic sensors and ultrasonic sensors in the parking garage, including signal amplitude, time series change rate and signal fluctuation range, and filters abnormal fluctuation data to obtain abnormal signal marker data.

[0015] The time window adjustment submodule, based on the abnormal signal marker data, calls the backup infrared sensor signal, adjusts the time window and aligns it with the original signal, and performs correction based on the deviation exceeding the limit to obtain the time window correction signal;

[0016] The signal compensation calculation submodule corrects the signal based on the time window, calls up ambient temperature parameters and magnetic field interference data, analyzes the impact on signal deviation, and uses the following formula:

[0017]

[0018] The sensor signal correction value is calculated;

[0019] Among them, S c S is the sensor signal correction value. w For time window correction signal, G i T represents the signal gradient change value. f B is an environmental temperature influencing factor. f is the magnetic field interference factor, and n represents the number of adjacent cycles of the signal gradient change.

[0020] As a further aspect of the present invention, the acquisition steps of the parking flow monitoring module are specifically as follows:

[0021] The vehicle entry and exit counting submodule organizes the vehicle entry and exit data of the parking lot entrance and exit based on the sensor signal correction value, counts the number of vehicles entering per unit time, and combines the entrance and exit recognition data to eliminate duplicate counts and obtain the vehicle entry volume.

[0022] The parking space status assessment submodule, based on the vehicle entry volume, compares the parking space data of the parking area to analyze the current occupancy status of the parking spaces, and counts the cumulative number of vehicles entering and leaving, using the formula:

[0023]

[0024] Generate parking space occupancy rate;

[0025] Among them, O r For parking space occupancy rate, C e To accumulate the number of vehicles entering, C x To calculate the cumulative number of vehicles leaving, T s P represents the total number of available parking spaces. b Use fluctuation parameters for parking spaces;

[0026] The parking flow trend analysis submodule, based on the parking space occupancy rate, statistically analyzes the changes in occupancy over a time series, calculates the rate of change in combination with the original parking data, analyzes the growth and decline trends of parking demand, and obtains the parking flow trend value.

[0027] As a further aspect of the present invention, the acquisition step of the regional load calculation module is specifically as follows:

[0028] The vehicle flow extraction submodule extracts vehicle entry and exit data per unit time based on the parking flow trend value, counts vehicle entry and exit frequency, and obtains vehicle flow per unit time.

[0029] The available parking space estimation submodule extracts vehicle dwell time data within the parking area based on the traffic flow per unit time, filters the usage duration of differentiated parking spaces, and analyzes the estimated number of parking spaces that can be released in a short period of time, using the formula:

[0030]

[0031] Generate a number of parking spaces that can be released in a short period of time;

[0032] Among them, S r To allow for the release of parking spaces in a short period of time, V t D represents the number of vehicles leaving per unit time. p F represents the average dwell time in a parking space. u M represents the parking full load ratio. t Release fluctuation parameters for parking spaces;

[0033] The load factor acquisition submodule, based on the number of short-term releaseable parking spaces and combined with parking space data in the parking area, identifies the rate of load change per unit time, analyzes the dynamics of the parking area load, and obtains the parking area load factor.

[0034] As a further aspect of the present invention, the step of obtaining the vehicle guidance adjustment module specifically includes:

[0035] The parking area load filtering submodule filters the parking area load coefficient, extracts real-time occupancy data of parking spaces within the parking area, filters high-load areas with load coefficients exceeding the threshold, and obtains the parking area load ranking result.

[0036] The recommendation weight adjustment submodule adjusts the recommendation weight of parking spaces in high-load areas based on the parking area load ranking results, identifies the influencing factors of parking space recommendations in high-load areas, and optimizes the priority of area parking space recommendations using the following formula:

[0037]

[0038] The adjusted parking space recommendation weights were calculated.

[0039] Where W' represents the adjusted parking space recommendation weight, W represents the original parking space recommendation weight, and L represents the current parking area load coefficient. max The maximum load factor in the parking area is represented by D, the original average parking time in the parking area is represented by T, and the parking duration at the current moment is represented by T.

[0040] The route optimization and allocation submodule modifies the vehicle navigation route data based on the adjusted parking space recommendation weights, optimizes the route allocation in low-load areas, removes the impact values ​​of short-term releaseable parking spaces, and obtains parking guidance adjustment parameters.

[0041] As a further aspect of the present invention, the step of obtaining the task scheduling optimization module specifically includes:

[0042] The parking guidance parameter filtering submodule filters the parking guidance adjustment parameters, extracts the real-time parking space occupancy rate, vehicle inflow and outflow rate and raw parking data of each parking area, analyzes the dynamic parking space utilization rate, and obtains the key parking guidance parameter combination.

[0043] The task urgency adjustment submodule calls the key parking guidance parameter combination, analyzes the remaining task execution time, compares the task's impact range data with the parking area's parking space utilization rate, and uses the following formula:

[0044]

[0045] Calculate the urgency coefficient of each task, sort and adjust the task priorities according to the urgency coefficient, and obtain a task urgency priority list;

[0046] Where U represents the task urgency coefficient, R j C represents the remaining execution time of task j. j Q is the estimated completion time for task j. j V represents the duration of the influence range of task j. j Let P be the fluctuation value of parking space occupancy in the parking area for task j. j Let be the impact of task j on the parking space utilization rate of the parking area, and m be the total number of tasks.

[0047] The resource demand allocation priority analysis submodule analyzes the matching degree between the remaining available resources in the parking area and the resources required by the task based on the task urgency priority list, sorts the resource demand allocation priority, and generates dynamic monitoring results of parking status.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0049] This invention records signal amplitude, time series rate of change, and signal fluctuation range, and compares the gradient change values ​​of adjacent period signals to achieve precise screening of abnormal fluctuation data, enhance the stability of sensor signals, reduce the impact of environmental interference on data acquisition accuracy, adjust the time window of backup infrared sensor signal data, and correct data with deviation values ​​exceeding limits, thereby improving data compensation capabilities. This ensures that the sensor signal maintains high reliability even under abnormal conditions. The influence of environmental temperature parameters and magnetic field interference data is analyzed, and the signal compensation coefficient is adjusted to ensure the sensor maintains high detection accuracy under different external conditions. Based on the corrected signal data, vehicle entry and exit information at parking lot entrances and exits is processed, parking space occupancy rates are calculated, and parking trends are analyzed to improve the perception of overall parking lot traffic dynamics. By comparing parking traffic trend values, vehicle flow and parking space occupancy data per unit time are extracted, and combined with vehicle dwell time data, the number of parking spaces that can be released in a short period is estimated, optimizing the real-time assessment capability of parking resources. Analyzing the rate of load change improves the accuracy of load trend prediction, enabling parking lot management to have stronger dynamic adjustment capabilities. It adjusts the recommended weight of parking spaces in high-load areas and optimizes route allocation in low-load areas based on vehicle navigation paths, improving the rationality of vehicle guidance and reducing congestion risks in high-load areas. Correcting the impact value of short-term releaseable parking space data makes navigation guidance more accurate, improving parking turnover efficiency. Analyzing the remaining execution time of tasks and comparing the task impact range data adjusts task urgency, making parking resource allocation more targeted. Combining resource demand allocation priorities ensures the reasonable sequencing of different types of tasks, improving the refinement of parking lot management and optimizing the dynamic monitoring capabilities of parking status. Attached Figure Description

[0050] Figure 1 This is a system flowchart of the present invention;

[0051] Figure 2 This is a flowchart illustrating the acquisition process of the signal stabilization calibration module in this invention.

[0052] Figure 3 This is a flowchart illustrating the acquisition process of the parking flow monitoring module in this invention.

[0053] Figure 4 This is a flowchart illustrating the acquisition process of the regional load calculation module in this invention.

[0054] Figure 5 This is a flowchart illustrating the acquisition process of the vehicle guidance and adjustment module in this invention.

[0055] Figure 6 This is a flowchart illustrating the acquisition process of the task scheduling optimization module in this invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0057] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0058] Please see Figure 1 The edge computing-based smart parking real-time monitoring system includes:

[0059] The signal stabilization calibration module records signal amplitude, time series change rate, and signal fluctuation range based on signal data from geomagnetic and ultrasonic sensors in the parking garage. It compares the gradient change values ​​of adjacent period signals, filters abnormal fluctuation data, adjusts the time window of backup infrared sensor signal data, corrects data with deviations exceeding limits, analyzes the influence of environmental temperature parameters and magnetic field interference data, adjusts the signal compensation coefficient, and obtains the sensor signal correction value.

[0060] The parking flow monitoring module, based on sensor signal correction values, organizes vehicle entry and exit data at parking lot entrances and exits, counts vehicle entry volume, compares parking space data in parking areas and calculates parking space occupancy rate, analyzes parking trends, and generates parking flow trend values.

[0061] The area load calculation module compares parking flow trend values, extracts vehicle flow per unit time and parking space occupancy data, filters vehicle dwell time data to estimate the number of parking spaces that can be released in a short time, analyzes the load change rate, and obtains the parking area load coefficient.

[0062] The vehicle guidance adjustment module filters the parking area load coefficient, adjusts the recommended weight of parking spaces in high-load areas, modifies vehicle navigation path data, optimizes path allocation in low-load areas, corrects the impact value of short-term releaseable parking space data, and obtains parking guidance adjustment parameters.

[0063] The task scheduling optimization module filters parking guidance adjustment parameters, analyzes the remaining execution time of tasks, compares the task impact range data to adjust the urgency of tasks, analyzes the priority of resource demand allocation, and generates dynamic monitoring results of parking status.

[0064] Sensor signal correction values ​​include geomagnetic signal correction values, ultrasonic signal correction values, infrared signal correction values, and signal compensation correction coefficients. Parking flow trend values ​​include vehicle entry volume, parking space occupancy rate, and entrance / exit flow data. Parking area load coefficients include vehicle flow per unit time, parking space occupancy data, and the number of parking spaces available for short-term release. Parking guidance adjustment parameters include parking space recommendation weights, navigation route adjustments, and route allocation optimization. Parking status dynamic monitoring results include task urgency, task impact range, and remaining task time.

[0065] Please see Figure 2 The specific steps for obtaining the signal stabilization calibration module are as follows:

[0066] The signal fluctuation screening submodule calculates the gradient change value between adjacent periods based on the signal data of geomagnetic sensors and ultrasonic sensors in the parking garage, including signal amplitude, time series change rate and signal fluctuation range, and filters abnormal fluctuation data to obtain abnormal signal marker data.

[0067] First, the parking status and movement of vehicles are continuously monitored using geomagnetic and ultrasonic sensors. By recording the signal amplitude and rate of change every minute, unusual patterns, such as abnormal vehicle movement, can be identified. For example, if the signal amplitude suddenly increases at a certain point in time, and the rate of change remains abnormal for the next few minutes, this period is marked as an abnormal signal. The threshold for anomalies can be set as the average of the signal amplitude and rate of change plus twice the standard deviation. This setting can effectively capture most abnormal activities while avoiding false alarms caused by normal fluctuations. In this way, when an abnormal signal is detected, the monitoring mechanism can be quickly invoked, such as sending a security alarm or adjusting the camera angle for tracking, thereby maintaining the safety and order of the parking lot and obtaining abnormal signal marking data.

[0068] The time window adjustment submodule uses abnormal signal marker data, calls backup infrared sensor signals, adjusts the time window and aligns it with the original signal, and performs correction based on deviation exceeding the limit to obtain the time window correction signal.

[0069] Precise adjustment of the time window can help managers better understand real-time changes in traffic flow. For example, if a surge in traffic flow is expected during a specific holiday, the time window can be adjusted based on infrared sensor data from the same time period in the past to ensure accurate data synchronization. The specific adjustment method can be achieved by analyzing the difference between the maximum deviation value in the marked data and the standard time window, and then dynamically adjusting the size of the time window, such as adjusting the original 5-minute time window to 3 minutes or 7 minutes, to adapt to different monitoring needs, ensure the continuity and integrity of the signal, and more accurately capture the actual entry and exit status of vehicles, assisting in making more reasonable parking lot scheduling decisions and obtaining time window correction signals.

[0070] The signal compensation calculation submodule corrects the signal based on a time window, calls environmental temperature parameters and magnetic field interference data, analyzes the impact on signal deviation, and uses the following formula:

[0071]

[0072] The sensor signal correction value is calculated;

[0073] Among them, S c S is the sensor signal correction value. w For time window correction signal, G i T represents the signal gradient change value. f B is an environmental temperature influencing factor. f is the magnetic field interference influence factor, and n represents the number of adjacent cycles of the signal gradient change;

[0074] In real-world scenarios, sensors are significantly affected by environmental factors, especially in high-temperature summer or low-temperature winter environments, where signal deviations occur, necessitating the calculation of the temperature influence factor T. f and magnetic field interference factor B f To perform signal correction, suppose a toll station measures a time window within a day and corrects the signal as S. w =75.3. Monitoring revealed that the gradient changes in the signal within adjacent periods were G1=3.2, G2=2.7, and G3=4.1, respectively. The temperature influence factor was T. f =0.15 (calculating the impact of temperature fluctuations on the sensor as the temperature drops from 35°C during the day to 20°C at night), magnetic field interference factor is B. f =9.6 (due to electromagnetic interference caused by nearby high-voltage power lines);

[0075] Calculate the sum of gradient changes:

[0076] Then calculate the signal correction part:

[0077] Considering the effect of temperature:

[0078] The corrected signal is calculated as: (65.3)×0.8696≈56.8;

[0079] Calculation of magnetic field interference correction part:

[0080] Final signal correction value: S c =56.8 + 3.1 = 59.9;

[0081] The results show that the final sensor signal correction value after temperature effect correction and magnetic field interference adjustment is 59.9, compared with the original signal of 75.3. The correction eliminates the error caused by environmental influence, making the data more accurate.

[0082] Please see Figure 3 The specific steps for obtaining data from the parking flow monitoring module are as follows:

[0083] The vehicle entry and exit counting submodule, based on sensor signal correction values, organizes vehicle entry and exit data at parking lot entrances and exits, counts the number of vehicles entering per unit time, and combines entrance and exit recognition data to eliminate duplicate counts and obtain the vehicle entry volume.

[0084] First, sensors installed at the entrance, such as infrared or geomagnetic sensors, monitor the moment a vehicle enters the parking lot in real time. Each time a vehicle passes, the sensor is triggered and sends a signal to the central processing unit. The central processing unit records the entry time of each vehicle and stores it in a database. The data is then used to calculate the total number of vehicles entering per hour or per day. For example, if the sensor is triggered 100 times in one hour, it indicates that 100 vehicles have entered the parking lot. This kind of statistics is particularly important for traffic flow analysis during peak and off-peak hours to obtain the number of vehicles entering.

[0085] The parking space status assessment submodule analyzes the current occupancy status of parking spaces based on vehicle entry volume, compares it with parking area data, and calculates the cumulative number of vehicles entering and leaving, using the following formula:

[0086]

[0087] Generate parking space occupancy rate;

[0088] Among them, O r For parking space occupancy rate, C e To accumulate the number of vehicles entering, C x To calculate the cumulative number of vehicles leaving, T s P represents the total number of available parking spaces. b Use fluctuation parameters for parking spaces;

[0089] Real-time vehicle flow data is compared with parking space status to determine parking space occupancy. Data on whether each parking space is occupied is collected using geomagnetic sensors located in the parking spaces. Whenever a vehicle enters or leaves a space, the geomagnetic sensor status changes, and the system updates the parking space occupancy status. The total number of vehicles entering is compared with the known number of parking spaces to calculate the occupancy rate. e The number of vehicles entering the country is the number after sensor correction, C x It is the number of vehicles leaving, T s It is the total number of parking spaces, P b This is a volatility parameter for parking space usage. For example, suppose a parking lot has 500 parking spaces, and 450 cars enter and 30 cars leave in one hour. The volatility parameter is assumed to be 0.05.

[0090] Calculated by inserting formula

[0091] This indicates that there is overuse or errors in the parking data, requiring further investigation or parameter adjustment to obtain the parking space occupancy rate.

[0092] The parking flow trend analysis submodule is based on the parking space occupancy rate, statistically analyzes the changes in occupancy over a time series, calculates the rate of change in combination with the original parking data, analyzes the growth and decline trends of parking demand, and obtains parking flow trend values.

[0093] First, by comparing current data with historical data for the same period, such as comparing the daily occupancy rate of the current month with the same period last month, the analysis department can identify whether there is an increasing or decreasing trend in parking demand. For example, if it is found that the weekend occupancy rate this month is 10% higher than the same period last month, it indicates that parking facilities need to be increased or prices need to be adjusted to manage demand. This trend analysis is crucial for managers to formulate strategies and optimize the allocation of parking resources, and to obtain parking flow trend values.

[0094] Please see Figure 4 The specific steps for obtaining the regional load calculation module are as follows:

[0095] The vehicle flow extraction submodule extracts vehicle entry and exit data per unit time based on parking flow trend values, counts vehicle entry and exit frequency, and obtains vehicle flow per unit time.

[0096] The system effectively extracts vehicle entry and exit data per unit time and filters parking space occupancy status. Through a real-time traffic flow monitoring system, vehicle entry and exit data is collected by sensors at the entrance and exit. These sensors detect and record the entry and exit times of each vehicle. Based on timestamp data and information from parking space occupancy sensors, the system calculates hourly traffic flow and real-time parking space occupancy status. This data reflects the operational efficiency of the parking lot and is crucial for predicting peak hours and managing parking resources. This enables effective management and optimization of traffic flow. The process not only includes data collection and analysis but also real-time monitoring of parking space usage, ensuring dynamic adjustment and optimization of parking lot management and obtaining traffic flow data per unit time.

[0097] The available parking space estimation submodule extracts vehicle dwell time data within the parking area based on traffic flow per unit time, filters for differentiated parking space usage durations, and analyzes the estimated number of parking spaces that can be released in a short period of time, using the following formula:

[0098]

[0099] Generate a number of parking spaces that can be released in a short period of time;

[0100] Among them, S r To allow for the release of parking spaces in a short period of time, V t D represents the number of vehicles leaving per unit time. p F represents the average dwell time in a parking space. u M represents the parking full load ratio. t Release fluctuation parameters for parking spaces;

[0101] Based on traffic flow data per unit time and vehicle dwell time data within the parking area, the number of parking spaces expected to be released in a short period of time can be accurately calculated. For example, in the parking lot of a large shopping mall, data collected by the vehicle recognition system shows that the average dwell time of each vehicle is about 2 hours. During a typical weekday noon, traffic flow statistics show that about 50 vehicles leave the parking lot per hour. Assuming that the parking lot is 80% full, that is, 80% of the parking spaces are occupied most of the time.

[0102] Among them, S r V indicates the number of parking spaces that can be released in a short period of time. t =50 is the number of vehicles leaving per unit time, D p =2 is the average dwell time (in hours) in the parking space, F u =0.80 is the parking lot's occupancy rate, M t =0.05 is the fluctuation parameter for parking space release, taking into account abnormal fluctuations, such as holidays or special events;

[0103] Substitute the value into the formula:

[0104] The calculation process is as follows:

[0105] It is estimated that approximately 5,224 parking spaces will be released in the next hour. The calculation helps managers to dynamically adjust parking resources, especially during periods when heavy traffic is expected, to ensure that they can effectively cope with sudden parking demands.

[0106] This strategy not only provides an effective method for resource allocation, but also helps managers optimize parking lot operations, ensure the maximum utilization of parking resources, and reduce customer dissatisfaction caused by parking difficulties, thereby improving the overall service quality and operational efficiency of parking lots.

[0107] The load factor acquisition submodule is based on the number of parking spaces that can be released in a short time. Combined with parking space data in the parking area, it identifies the rate of load change per unit time, analyzes the dynamics of the parking area load, and obtains the parking area load factor.

[0108] Calculating the rate of load change per unit time and analyzing the load dynamics of parking areas involves comparing real-time data with historical load data to identify trends of load growth or decline. For example, if the current load factor continues to rise, it indicates increased parking demand, requiring the expansion of parking facilities or optimization of parking configuration. Conversely, if the load factor declines, it indicates the need to reduce the space reserved for parking to better utilize area resources. This analysis helps parking managers make data-driven decisions to ensure efficient use of parking resources, improve user satisfaction, and ultimately obtain the load factor of parking areas.

[0109] Please see Figure 5 The specific steps for obtaining the vehicle guidance and adjustment module are as follows:

[0110] The parking area load filtering submodule filters the parking area load coefficient, extracts real-time occupancy data of parking spaces within the parking area, filters high-load areas with load coefficients exceeding the threshold, and obtains the parking area load ranking results.

[0111] First, real-time occupancy data of all parking spaces within the parking area is acquired through a sensor network. This data includes vehicle entry and exit times. Then, the occupancy ratio of each area is calculated. For example, in a parking area with 100 spaces, if 80 spaces are currently occupied, the occupancy ratio is 80%. High-load areas with parking load coefficients exceeding a preset threshold (e.g., 50%) are filtered out. For instance, if the occupancy ratio of a parking area exceeds 50%, this area is marked as high-load. Based on the occupancy ratio, the load coefficients of different parking areas are sorted to identify those high-load areas that require special attention. This results in a parking area load ranking, which aims to optimize the traffic flow distribution within the parking lot, making vehicles more evenly distributed across different parking areas.

[0112] The recommendation weight adjustment submodule adjusts the recommendation weight of parking spaces in high-load areas based on the parking area load ranking results, identifies the influencing factors of parking space recommendations in high-load areas, and optimizes the priority of parking space recommendations in different areas using the following formula:

[0113]

[0114] The adjusted parking space recommendation weights were calculated.

[0115] Where W' represents the adjusted parking space recommendation weight, W represents the original parking space recommendation weight, and L represents the current parking area load coefficient. max The maximum load factor in the parking area is represented by D, the original average parking time in the parking area is represented by T, and the parking duration at the current moment is represented by T.

[0116] First, obtain the original recommendation weight W for the parking area. This weight can be calculated using historical vehicle parking data. For example, if the original recommendation weight W for a certain area is set to 0.9, it means that the system recommends parking spaces in that area by default. Then, calculate the load coefficient L for the current parking area. This coefficient is calculated based on the real-time parking space occupancy status within the parking area. For example, if there are 200 parking spaces in the area, and 160 of them are occupied, then... Determine the maximum load factor L in all parking areas. max For example, in another region, this value reaches a maximum of 0.95, i.e., L. max =0.95;

[0117] Calculate the historical average parking duration D for the parking area. This value can be obtained by statistically analyzing parking data from the past 30 days. For example, if the average parking duration of a certain parking area is 120 minutes, then D = 120. Obtain the parking duration T of vehicles in the area at the current moment. For example, if a car has been parked for 40 minutes, then T = 40.

[0118] Substitute specific values:

[0119]

[0120] W' = 0.9 × 0.1579 + 2.9268;

[0121] W' = 3.0689;

[0122] The adjusted parking space recommendation weight for this area was calculated to be 3.07 (rounded to two decimal places). The calculation results show that the recommendation weight for this area is higher than the original weight of 0.9. This indicates that although the current load in this area is high, some vehicles are about to leave due to the long average parking time, so it is suitable to continue recommending it. If the historical parking time of a certain area is short or the current parking duration is short, its recommendation weight will be lower, and vehicles will be guided to more suitable parking areas first, ultimately achieving dynamic load balancing and optimizing the parking space allocation strategy.

[0123] The route optimization and allocation submodule modifies the vehicle navigation route data based on the adjusted parking space recommendation weight, optimizes the route allocation in low-load areas, removes the impact value of short-term releaseable parking spaces, and obtains parking guidance adjustment parameters.

[0124] The optimal path weight of the recommended route is calculated, and the impact value of parking spaces released in a short period of time is removed to ensure that drivers are guided to the most suitable parking area. For example, if multiple parking spaces suddenly become available in a low-load area, the navigation route will be dynamically adjusted to guide the new vehicles to that area, thereby optimizing the use of parking spaces in the parking lot and finally obtaining parking guidance adjustment parameters, which helps to achieve more efficient and fair parking space allocation.

[0125] Please see Figure 6 The specific steps to obtain the task scheduling optimization module are as follows:

[0126] The parking guidance parameter filtering submodule filters parking guidance adjustment parameters, extracts real-time parking space occupancy rate, vehicle inflow and outflow rate and raw parking data for each parking area, analyzes dynamic parking space utilization, and obtains key parking guidance parameter combinations.

[0127] Based on real-time monitoring data of parking lots, statistical analysis is performed on the parking space occupancy rate, vehicle inflow and outflow rates, and historical parking data for each parking area. Real-time data streams are used to monitor the real-time status of each parking area. For example, in an area with 100 parking spaces, the real-time monitoring system reports that 80 spaces are occupied during peak hours, with an inflow rate of 1 vehicle per minute and an outflow rate of 0.5 vehicles per minute. Historical data shows that the occupancy rate for this area during the same period is 70%. Therefore, the analysis concludes that the parking space utilization rate in this area is abnormally high at the current time, requiring adjustments to the guidance strategy to direct vehicles to the area. The dynamic parking space utilization rate is calculated to determine the parking space utilization efficiency of each parking area. The utilization rate calculation formula is the number of real-time occupied parking spaces divided by the total number of parking spaces, yielding a result of 0.8, indicating a utilization rate of 80%, higher than normal. Further filtering is used to identify key parking guidance parameter combinations, including the total number of parking spaces in the area, the number of real-time occupied parking spaces, vehicle inflow rate, and outflow rate, providing decision support for real-time dynamic parking guidance.

[0128] The task urgency adjustment submodule calls key parking guidance parameter combinations, analyzes the remaining task execution time, compares the task's impact range data with the parking area's space utilization rate, and uses the following formula:

[0129]

[0130] Calculate the urgency coefficient of each task, sort and adjust the task priorities according to the urgency coefficient, and obtain a task urgency priority list;

[0131] Where U represents the task urgency coefficient, R j C represents the remaining execution time of task j. j Q is the estimated completion time for task j. j V represents the duration of the influence range of task j. j Let P be the fluctuation value of parking space occupancy in the parking area for task j. j Let be the impact of task j on the parking space utilization rate of the parking area, and m be the total number of tasks.

[0132] First, obtain the remaining execution time, estimated completion time, and duration of impact for each task. These parameters can be obtained in real time through the task scheduling system. For example, task A has a remaining execution time of 20 minutes, an estimated completion time of 18 minutes, and a duration of impact of 30 minutes; task B has a remaining execution time of 15 minutes, an estimated completion time of 14 minutes, and a duration of impact of 20 minutes; and task C has a remaining execution time of 10 minutes, an estimated completion time of 11 minutes, and a duration of impact of 15 minutes. All data is obtained and recorded through the real-time monitoring function of the scheduling system.

[0133] Based on the parking space utilization rate and vehicle fluctuation value in the key parking guidance parameter combination, the task impact range data is compared, and the urgency of the task on parking resource occupation is calculated. The parking space fluctuation value is calculated by the difference between historical data and real-time inflow and outflow rates.

[0134] For example, the parking space fluctuation value in the area affected by Task A is 0.1, and the utilization rate impact is 0.9;

[0135] The parking space fluctuation value in the area affected by Task B is 0.2, and the utilization rate impact is 0.85.

[0136] The parking space fluctuation value in the area affected by Task C is 0.15, and the utilization rate impact is 0.95.

[0137] Substitute the parameters:

[0138] R1 = 20, R2 = 15, R3 = 10 (remaining execution time);

[0139] C1 = 18, C2 = 14, C3 = 11 (estimated completion time);

[0140] Q1 = 30, Q2 = 20, Q3 = 15 (duration of the affected area);

[0141] V1 = 0.1, V2 = 0.2, V3 = 0.15 (parking space fluctuation value);

[0142] P1 = 0.9, P2 = 0.85, P3 = 0.95 (impact of parking space utilization rate);

[0143] Perform step-by-step calculations:

[0144] calculate |20-18|+|15-14|+|10-11|=2+1+1=4;

[0145] calculate 30 + 20 + 15 = 65;

[0146] Calculate Part 1:

[0147] calculate

[0148] Calculate the square root:

[0149] The final calculated value of U is: U = 0.0615 + 0.304 = 0.3655;

[0150] The task urgency coefficient U = 0.3655 is obtained. Based on this urgency coefficient, the tasks are sorted according to their urgency. For example, if the urgency coefficients of tasks A, B, and C are 0.3655, 0.25, and 0.4 respectively, the priority order is: task C > task A > task B. The final task urgency priority list is obtained and used for subsequent resource allocation priority analysis.

[0151] The resource demand allocation priority analysis submodule analyzes the matching degree between the remaining available resources in the parking area and the resources required by the tasks based on the task urgency priority list, sorts the resource demand allocation priority, and generates dynamic monitoring results of parking status.

[0152] The system calculates the matching degree between the remaining available parking resources in a parking area and the resources required for a task by using data provided by the real-time monitoring system. For example, if an area has 20 available parking spaces and an emergency task requires 10 spaces, the matching degree is calculated by dividing the number of spaces required for the task by the number of spaces remaining in the area. A result of 0.5 indicates that half of the resource requirements for the task can be met immediately. The system prioritizes the allocation of resources to each task, thereby effectively managing parking resources and ensuring that high-priority tasks receive the resources they need first. Finally, it generates dynamic monitoring results of parking status, reflecting the real-time allocation of resources in each area, providing a dynamic adjustment tool for parking lot management.

[0153] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A smart parking real-time monitoring system based on edge computing, characterized in that, The system includes: The signal stabilization calibration module uses signal data from geomagnetic and ultrasonic sensors in the parking garage to filter abnormal fluctuation data, adjust the time window of the backup infrared sensor signal, analyze the influence of ambient temperature and magnetic field interference, adjust the signal compensation coefficient, and obtain the sensor signal correction value. Based on the sensor signal correction value, the parking flow monitoring module organizes the vehicle entry and exit data of the parking lot entrance and exit, counts the number of vehicles entering, compares the parking space data of the parking area, and generates a parking flow trend value. The area load calculation module compares the parking flow trend value, extracts the traffic flow per unit time and parking space occupancy data of the parking area, filters the vehicle stay time data, estimates the number of parking spaces that can be released in a short time, and obtains the parking area load coefficient. The vehicle guidance adjustment module filters the parking area load coefficient, adjusts the recommended weight of parking spaces in high-load areas, modifies vehicle navigation path data, and optimizes path allocation in low-load areas to obtain parking guidance adjustment parameters. The task scheduling optimization module filters the parking guidance adjustment parameters, analyzes the remaining execution time of the task, compares the task's impact range data, adjusts the task's urgency, and generates dynamic monitoring results of the parking status.

2. The smart parking real-time monitoring system based on edge computing according to claim 1, characterized in that, The sensor signal correction values ​​include geomagnetic signal correction values, ultrasonic signal correction values, infrared signal correction values, and signal compensation correction coefficients. The parking flow trend values ​​include vehicle entry volume, parking space occupancy rate, and entrance / exit flow data. The parking area load coefficient includes vehicle flow per unit time, parking space occupancy data, and the number of parking spaces available for short-term release. The parking guidance adjustment parameters include parking space recommendation weight, navigation path adjustment, and path allocation optimization. The dynamic monitoring results of parking status include task urgency, task impact range, and remaining task time.

3. The smart parking real-time monitoring system based on edge computing according to claim 1, characterized in that, The specific steps for obtaining the signal stabilization calibration module are as follows: The signal fluctuation screening submodule calculates the gradient change value between adjacent periods based on the signal data of geomagnetic sensors and ultrasonic sensors in the parking garage, including signal amplitude, time series change rate and signal fluctuation range, and filters abnormal fluctuation data to obtain abnormal signal marker data. The time window adjustment submodule, based on the abnormal signal marker data, calls the backup infrared sensor signal, adjusts the time window and aligns it with the original signal, and performs correction based on the deviation exceeding the limit to obtain the time window correction signal; The signal compensation calculation submodule corrects the signal based on the time window, calls up ambient temperature parameters and magnetic field interference data, analyzes the impact on signal deviation, and uses the following formula: The sensor signal correction value is calculated; Among them, S c S is the sensor signal correction value. w For time window correction signal, G i T represents the signal gradient change value. f B is an environmental temperature influencing factor. f is the magnetic field interference factor, and n represents the number of adjacent cycles of the signal gradient change.

4. The intelligent parking real-time monitoring system based on edge computing according to claim 3, characterized in that, The specific steps for obtaining information from the parking flow monitoring module are as follows: The vehicle entry and exit counting submodule organizes the vehicle entry and exit data of the parking lot entrance and exit based on the sensor signal correction value, counts the number of vehicles entering per unit time, and combines the entrance and exit recognition data to eliminate duplicate counts and obtain the vehicle entry volume. The parking space status assessment submodule, based on the vehicle entry volume, compares the parking space data of the parking area to analyze the current occupancy status of the parking spaces, and counts the cumulative number of vehicles entering and leaving, using the formula: Generate parking space occupancy rate; Among them, O r For parking space occupancy rate, C e To accumulate the number of vehicles entering, C x To calculate the cumulative number of vehicles leaving, T s P represents the total number of available parking spaces. b Use fluctuation parameters for parking spaces; The parking flow trend analysis submodule, based on the parking space occupancy rate, statistically analyzes the changes in occupancy over a time series, calculates the rate of change in combination with the original parking data, analyzes the growth and decline trends of parking demand, and obtains the parking flow trend value.

5. The smart parking real-time monitoring system based on edge computing according to claim 4, characterized in that, The specific steps for obtaining the regional load calculation module are as follows: The vehicle flow extraction submodule extracts vehicle entry and exit data per unit time based on the parking flow trend value, counts vehicle entry and exit frequency, and obtains vehicle flow per unit time. The available parking space estimation submodule extracts vehicle dwell time data within the parking area based on the traffic flow per unit time, filters the usage duration of differentiated parking spaces, and analyzes the estimated number of parking spaces that can be released in a short period of time, using the formula: Generate a number of parking spaces that can be released in a short period of time; Among them, S r To allow for the release of parking spaces in a short period of time, V t D represents the number of vehicles leaving per unit time. p F represents the average dwell time in a parking space. u M represents the parking full load ratio. t Release fluctuation parameters for parking spaces; The load factor acquisition submodule, based on the number of short-term releaseable parking spaces and combined with parking space data in the parking area, identifies the rate of load change per unit time, analyzes the dynamics of the parking area load, and obtains the parking area load factor.

6. The smart parking real-time monitoring system based on edge computing according to claim 5, characterized in that, The specific steps for obtaining the vehicle guidance adjustment module are as follows: The parking area load filtering submodule filters the parking area load coefficient, extracts real-time occupancy data of parking spaces within the parking area, filters high-load areas with load coefficients exceeding the threshold, and obtains the parking area load ranking result. The recommendation weight adjustment submodule adjusts the recommendation weight of parking spaces in high-load areas based on the parking area load ranking results, identifies the influencing factors of parking space recommendations in high-load areas, and optimizes the priority of parking space recommendations in these areas using the following formula: The adjusted parking space recommendation weights were calculated. Where W' represents the adjusted parking space recommendation weight, W represents the original parking space recommendation weight, and L represents the current parking area load coefficient. max The maximum load factor in the parking area is represented by D, the original average parking time in the parking area is represented by T, and the parking duration at the current moment is represented by T. The route optimization and allocation submodule modifies the vehicle navigation route data based on the adjusted parking space recommendation weights, optimizes the route allocation in low-load areas, removes the impact values ​​of short-term releaseable parking spaces, and obtains parking guidance adjustment parameters.

7. The smart parking real-time monitoring system based on edge computing according to claim 6, characterized in that, The specific steps for obtaining the task scheduling optimization module are as follows: The parking guidance parameter filtering submodule filters the parking guidance adjustment parameters, extracts the real-time parking space occupancy rate, vehicle inflow and outflow rate and raw parking data of each parking area, analyzes the dynamic parking space utilization rate and obtains the key parking guidance parameter combination. The task urgency adjustment submodule calls the key parking guidance parameter combination, analyzes the remaining task execution time, compares the task's impact range data with the parking area's parking space utilization rate, and uses the following formula: Calculate the urgency coefficient of each task, sort and adjust the task priorities according to the urgency coefficient, and obtain a task urgency priority list; Where U represents the task urgency coefficient, R j C represents the remaining execution time of task j. j Q is the estimated completion time for task j. j V represents the duration of the influence range of task j. j Let P be the fluctuation value of parking space occupancy in the parking area for task j. j Let be the impact of task j on the parking space utilization rate of the parking area, and m be the total number of tasks. The resource demand allocation priority analysis submodule analyzes the matching degree between the remaining available resources in the parking area and the resources required by the task based on the task urgency priority list, sorts the resource demand allocation priority, and generates dynamic monitoring results of parking status.

Images