A shared parking space access control method, system, product and medium

Abstract

The application discloses a shared parking space access control method, system, product and medium, and relates to the field of vehicle management.The method comprises the following steps: extracting a time interval from leaving to returning, counting the frequency of each time interval according to granularity, and obtaining return time length characteristic data.According to the current leaving time length of the empty owner parking space corresponding to the license plate, the proportion of the frequency of the forward-looking window period in the characteristic data is read to obtain the return tendency degree, and the conflict risk value of the return tendency degree of the parking space occupied by the shared vehicle is obtained.The parking space with a return tendency degree lower than a safety threshold and not occupied is marked as a hedging candidate parking space, the effective hedging depth is calculated, and the difference between the effective hedging depth and the conflict risk value and a redundancy constant are compared.If the difference is not less than the redundancy constant, access is allowed, and the hedging candidate parking space is selected to establish a hedging pairing mapping.When the owner vehicle returns and finds that the parking space is occupied, the hedging parking space number is read, and temporary parking guidance is sent to the owner terminal.The application can improve the resource utilization rate of the shared parking lot.

Description

Technical Field

[0001] This application relates to the field of vehicle management, and in particular to a method, system, product, and medium for access control of shared parking spaces. Background Technology

[0002] With the continuous growth of urban motor vehicle ownership, the contradiction between the supply and demand of parking spaces in residential communities and commercial buildings is becoming increasingly prominent. Some owners' designated parking spaces are vacant for a certain period of time due to daytime commuting, running errands, etc. In order to improve the utilization rate of parking spaces, the owner parking space sharing model is gradually being promoted and applied in smart parking management systems, which allows external vehicles to temporarily park in owners' parking spaces during vacant periods.

[0003] Existing shared parking space solutions typically use a time-segment declaration method for management. Owners pre-set the start and end times for sharing their parking spaces via their terminals. During the declared time period, the system adds the corresponding parking space to the shared resource pool and grants access to external vehicles. As the declared time period is about to end, it sends a departure reminder to the shared vehicles, requiring them to leave before the deadline to return the parking space to the owner.

[0004] However, residents' actual travel is affected by factors such as commuting fluctuations and unforeseen events, and there is often a discrepancy between their return time and the end of the pre-declared sharing period. When residents return earlier due to a shortened journey, their designated parking space may still be occupied by shared vehicles, and since the declared period has not yet ended, the system cannot trigger the departure process in advance. The returning residents' vehicles can only stay and wait inside the premises, causing congestion and reducing residents' willingness to participate. Summary of the Invention

[0005] This application provides a shared parking space access control method, system, product, and medium to improve the resource utilization rate of shared parking lots.

[0006] Firstly, this application provides a shared parking space access control method, applied to a shared parking space access control system. The method includes: extracting the time interval between each departure and return from the gate access records of each registered license plate within a preset historical period; statistically analyzing the frequency of occurrence within each time period according to a preset granularity to obtain return duration characteristic data corresponding to each registered license plate; based on the current departure duration of the registered license plate corresponding to each vacant owner's parking space, reading the frequency percentage of the time period covered by a preset look-ahead window from the return duration characteristic data of the registered license plate to obtain the return tendency of each vacant owner's parking space; summarizing the return tendency of all owner's parking spaces currently occupied by shared vehicles to obtain a conflict risk value; and classifying parking spaces with a return tendency value lower than a preset safety value as conflict risk values. Owner parking spaces that are not occupied by shared vehicles and meet the threshold are marked as hedging candidate parking spaces. The effective hedging depth is calculated based on the preset benchmark value and the return tendency of each hedging candidate parking space. When a shared vehicle triggers an entry request, the difference between the effective hedging depth and the conflict risk value is compared with the preset redundancy constant to obtain the first difference. When the first difference is not less than the preset redundancy constant, an access instruction is generated, and one of the hedging candidate parking spaces is selected to establish a hedging pairing mapping with the allocated shared parking space. When the entrance gate recognizes that the owner's vehicle has returned and the owner's parking space to which the owner's vehicle belongs is in a shared occupation state, the corresponding hedging parking space number in the hedging pairing mapping is read, and a temporary parking instruction containing the hedging parking space number is sent to the owner's associated terminal.

[0007] In the above embodiments, by extracting return time feature data from the gate passage records and calculating the return tendency based on the look-ahead window, the return probability of each vacant parking space is quantified into a comparable numerical index. On this basis, the return tendency of occupied parking spaces is summarized into a conflict risk value, and vacant parking spaces with low return tendency are used as hedging candidates to calculate the effective hedging depth. When a shared vehicle requests entry, the difference between the effective hedging depth and the conflict risk value is used to determine whether there is sufficient hedging margin to deal with the situation of the owner returning early. Only when the margin is sufficient is an access instruction generated and a hedging pairing mapping established. When the owner returns early, the system directly pushes a temporary parking space to the owner based on the hedging pairing mapping, so that the owner can complete the parking without waiting in the parking lot, ensuring the continuity of the owner's car use experience and the traffic efficiency of the parking lot channel under the parking space sharing mode.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the step of reading the frequency proportion of time periods covered by a preset look-ahead window from the return duration feature data of registered license plates to obtain the return tendency of each vacant owner's parking space specifically includes: using the current departure duration of the registered license plate corresponding to each vacant owner's parking space as the starting offset, and calculating the ending offset based on the starting offset and the window length of the preset look-ahead window; extracting the occurrence frequency corresponding to each time period between the starting offset and the ending offset from the return duration feature data of registered license plates; assigning attenuation weights to each time period according to the distance between each time period and the starting offset, wherein the attenuation weights decrease as the distance increases; calculating the weighted frequency total based on the occurrence frequency of each time period and the corresponding attenuation weight; summing the occurrence frequencies of all time periods in the return duration feature data to obtain the historical frequency total; and using the ratio of the weighted frequency total to the historical frequency total as the return tendency of the vacant owner's parking space.

[0009] In the above embodiments, the coverage of the forward window is determined by using the current departure time as the starting offset, so that the starting point for calculating the return tendency is aligned with the actual departure status of the vehicle. In each time period covered by the window, attenuation weights are assigned according to the distance between each time period and the starting offset, with the closer the time period, the higher the weight is given, so that the return events that may occur in the near future have a greater impact on the calculation results. The return tendency is obtained by comparing the total weighted frequency with the total historical frequency. This ratio is normalized to make the return tendency of different license plates comparable, providing input data with a unified dimension for subsequent conflict risk value aggregation and hedging depth calculation, thereby improving the accuracy of access control decisions.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the step of summarizing the return tendency of all owner parking spaces currently occupied by shared vehicles to obtain a conflict risk value, marking owner parking spaces with a return tendency lower than a preset safety threshold and not occupied by shared vehicles as hedging candidate parking spaces, and calculating the effective hedging depth based on a preset benchmark value and the return tendency of each hedging candidate parking space specifically includes: taking any pair of owner parking spaces from the owner parking spaces currently occupied by shared vehicles, and based on the appearance of each owner parking space within the time period covered by a preset forward window. The frequency distribution determines the overlap of return times for each pair of owner parking spaces; the return tendency of each owner parking space currently occupied by shared vehicles is accumulated to obtain the basic conflict amount; the coupling conflict increment is determined based on the overlap of return times for each pair of owner parking spaces; the basic conflict amount and the coupling conflict increment are added together to obtain the conflict risk value; each currently vacant owner parking space not occupied by shared vehicles is traversed, and owner parking spaces with a return tendency lower than a preset safety threshold are marked as hedging candidate parking spaces; the effective hedging depth is obtained by accumulating the difference between the preset benchmark value and the return tendency of each hedging candidate parking space.

[0011] In the above embodiments, when summarizing the conflict risk value, not only are the return tendency of each occupied parking space accumulated to obtain the basic conflict amount, but the overlap of return time periods is also determined by analyzing the frequency distribution of each pair of occupied parking spaces in the look-ahead window, and the coupled conflict increment is calculated accordingly. The two are then added together to obtain the conflict risk value. This method incorporates the joint probability of multiple occupied parking spaces returning simultaneously in the same time period into the risk assessment, avoiding the risk underestimation problem caused by simply accumulating the return probability of a single parking space. When calculating the effective hedging depth, the difference between the preset benchmark value and the return tendency of each hedging candidate parking space is accumulated, so that the hedging depth reflects the actual buffer capacity that each candidate parking space can provide, providing a more accurate basis for risk and capacity comparison for access decisions.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the step of generating an admission instruction when the first difference is not less than a preset redundancy constant, and selecting a hedging candidate parking space from the hedging candidate parking spaces to establish a hedging pairing mapping with the allocated shared parking space, specifically includes: generating an admission instruction in response to the first difference being not less than a preset redundancy constant; obtaining the spatial proximity between each hedging candidate parking space and the allocated shared parking space; comprehensively determining the pairing adaptation score of each hedging candidate parking space based on the return tendency of each hedging candidate parking space and the corresponding spatial proximity; selecting the hedging candidate parking space with the highest pairing adaptation score as the target hedging parking space; associating the target hedging parking space with the allocated shared parking space and writing it into the hedging pairing mapping table; removing the target hedging parking space from the hedging candidate parking spaces; and updating the effective hedging depth based on the difference between the preset benchmark value and the return tendency of the target hedging parking space.

[0013] In the above embodiments, after generating the admission instruction, the matching and adaptation score is determined by combining the return tendency of each hedging candidate parking space with the spatial proximity between each hedging candidate parking space and the allocated shared parking space, so that the selected target hedging parking space simultaneously meets the two conditions of low return risk and short-distance guidance. After the target hedging parking space is associated with the shared parking space and written into the mapping table, the target hedging parking space is removed from the candidate pool and the effective hedging depth is updated to ensure that the same hedging parking space is not repeatedly allocated to multiple shared parking spaces, thus maintaining the accuracy of the hedging resource pool. This mechanism ensures that the parking space pointed to by the temporary parking guidance received by the owner when returning to the site is closer to the owner's original parking space, shortening the owner's driving path within the site and reducing the probability of cross movement of vehicles within the site.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, after sending a temporary parking instruction containing the hedging parking space number to the owner's associated terminal, the method further includes: based on the return duration characteristic data of the registered license plates corresponding to each hedging candidate parking space, rereading the frequency proportion of the time periods covered by the preset forward window along consecutive time periods after the current time, to obtain the predicted return tendency of each hedging candidate parking space in each time period; excluding hedging candidate parking spaces whose predicted return tendency exceeds a preset safety threshold in each time period, and recalculating the predicted effective hedging depth corresponding to each time period based on the predicted return tendency of the remaining hedging candidate parking spaces and a preset benchmark value; subtracting the predicted effective hedging depth corresponding to each time period from the conflict risk value, and determining the time period when the difference first falls below a preset redundancy constant as the hedging exhaustion time period; determining the upper limit of the allowed parking duration based on the time interval between the hedging exhaustion time period and the current time period, and appending the upper limit of the allowed parking duration to the admission instruction.

[0015] In the above embodiment, the predicted return tendency of each hedging candidate parking space is recalculated for each consecutive time period after the current time, and candidate parking spaces exceeding the safety threshold are excluded for each time period to obtain the predicted effective hedging depth for each time period. By subtracting the predicted effective hedging depth from the conflict risk value for each time period, the time period when the hedging resources are insufficient to cover the conflict risk is determined, i.e., the hedging exhaustion period. The interval between the hedging exhaustion period and the current time period is used as the upper limit of the allowed parking time and attached to the access instruction so that shared vehicles know the maximum parking time when entering the site, avoiding the situation where temporary parking spaces cannot be provided for returning owners after the hedging resources are exhausted due to the shared vehicles parking for an extended period.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after sending a temporary parking instruction containing the offset parking space number to the owner's associated terminal, the method further includes: based on the return time characteristic data of the registered license plates corresponding to the parking spaces occupied by shared vehicles, rereading the frequency proportion of the time periods covered by the preset look-ahead window in consecutive time periods after the current time to obtain the predicted return tendency of each owner's parking space occupied by shared vehicles in each time period; summing the predicted return tendency of each owner's parking space occupied by shared vehicles in each time period to obtain the predicted conflict risk value corresponding to each time period; subtracting the effective offset depth from the predicted conflict risk value corresponding to each time period, and determining the time period when the difference is first not less than a preset redundancy constant as the expected recovery period; generating the expected waiting time based on the time interval between the expected recovery period and the current time period, and sending a queuing instruction containing the expected waiting time to the shared vehicle that triggered the entry request.

[0017] In the above embodiment, when the access conditions are not met, the changing trend of the conflict risk value is predicted time by time along future time periods. As time goes by, the departure of some shared vehicles in occupied parking spaces will gradually reduce the conflict risk value. The system determines the time period when the difference between the effective hedging depth and the predicted conflict risk value first meets the access conditions as the expected recovery period. Based on this, the expected waiting time is calculated and a queuing instruction is sent to the shared vehicles. This method provides quantifiable waiting expectations to shared vehicles, enabling drivers to judge whether to continue queuing or choose other parking lots based on the waiting time, thereby improving the information transparency of the parking lot's external services.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of comparing the difference between the effective hedging depth and the conflict risk value with a preset redundancy constant to obtain a first difference, the method further includes: when the first difference is less than the preset redundancy constant, generating a rejection instruction and preventing the entry gate from opening.

[0019] In the above embodiments, when the difference between the effective hedging depth and the conflict risk value does not meet the preset redundancy constant requirement, a rejection command is generated and the entrance gate is prevented from opening, so that shared vehicles cannot enter the parking lot when hedging resources are insufficient. This mechanism ensures that the number of owner parking spaces occupied by shared vehicles at any time does not exceed the upper limit of the hedging guarantee that the system can provide, preventing a situation where there are no available hedging parking spaces when owners return to the parking lot in a concentrated manner due to excessive acceptance of shared vehicles, thus maintaining the stability of the parking lot operation order and the basic protection of the owners' rights.

[0020] In a second aspect, embodiments of this application provide a shared parking space access control system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the shared parking space access control system to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a shared parking space access control system, cause the shared parking space access control system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a shared parking space access control system, cause the shared parking space access control system to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the shared parking space access control system provided in the second aspect, the computer storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0025] 1. By employing technical means such as extracting return duration feature data from gate passage records, calculating the return tendency of each vacant owner's parking space based on a forward-looking window, summarizing the return tendency of occupied parking spaces into a conflict risk value, constructing a hedging candidate pool with low-proportion vacant parking spaces and calculating the effective hedging depth, using the difference between the effective hedging depth and the conflict risk value for admission judgment and establishing a hedging pairing mapping, and sending temporary parking guidance to returning owners based on the hedging pairing mapping, the system can assess the owner's return risk and reserve space before shared vehicles enter the site. By offsetting parking space resources, when a resident actually returns to the site and their original parking space is occupied, the resident is directly guided to the offset parking space to complete parking based on a pre-established mapping relationship. This effectively solves the technical problem in existing technologies where the resident's exclusive parking space is still occupied by shared vehicles when they return, and the system cannot trigger the departure process in advance, resulting in the resident's vehicle being stuck in the site waiting, causing congestion in the passageway, and a decrease in the resident's willingness to participate. This enables preventive control and immediate resolution of return conflicts under the parking space sharing model, ensuring the continuity of parking when the resident returns and the traffic order of the parking lot passageway.

[0026] 2. By employing techniques such as using the current departure time as the starting offset to determine the coverage of the forward window, aligning the calculation starting point with the actual departure status of the vehicle; allocating attenuation weights according to the distance of each time period from the starting offset, giving recent time periods a higher weight coefficient than distant time periods; and comparing the total weighted frequency with the total historical frequency to obtain the normalized return tendency, the calculation result of the return tendency is dynamically correlated with the current actual departure duration of the vehicle and also has higher sensitivity to recent return events. Furthermore, the return tendency of each parking space is comparable under a unified dimension after normalization. This effectively solves the technical problem in existing technologies where there is a lack of means to dynamically quantify the probability of the owner's return, resulting in a lack of accurate risk input data for access decisions. Thus, it achieves dynamic alignment between the return risk assessment result and the real-time status of the vehicle, as well as standardized comparison of risk indicators between different parking spaces.

[0027] 3. By employing technical means such as determining the overlap of return periods based on the frequency distribution of each pair of occupied parking spaces within the forward window, adding the basic conflict quantity to the coupled conflict increment calculated based on the overlap to obtain the conflict risk value, and accumulating the difference between the preset benchmark value and the return tendency of each hedging candidate parking space to obtain the effective hedging depth, the calculation of the conflict risk value includes the joint probability of multiple parking spaces returning simultaneously in the same period in the evaluation scope rather than simply accumulating the probability of a single parking space. Furthermore, the effective hedging depth is accumulated based on the remaining distance of each candidate parking space from the benchmark value, thereby reflecting the actual available buffer capacity. This effectively solves the technical problems in the existing technology of insufficient risk assessment of multiple parking spaces returning simultaneously and the lack of accurate measurement methods for hedging resource capacity, which leads to deviations in access decisions. Thus, it achieves accurate matching and comparison of conflict risk and hedging capacity within the same quantitative framework. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a shared parking space access control method in an embodiment of this application;

[0029] Figure 2 This is another flowchart illustrating the shared parking space access control method in this application embodiment;

[0030] Figure 3 This is a schematic diagram of the physical device structure of a shared parking space access control system in the embodiments of this application. Detailed Implementation

[0031] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0033] To facilitate understanding, the application scenarios of the embodiments of this application are described below.

[0034] This application's embodiments apply to shared parking scenarios where shared parking facilities employing a multi-level parking structure open their spaces to external shared vehicles during the owner's vehicle departure period. In a typical implementation environment, the parking facility adopts a honeycomb-honeycomb shell combined structure. The honeycomb shells are fixed within a steel structure frame using lifting points. The garage is configured with a multi-level layout, for example, four levels with twenty-five parking spaces per level, totaling one hundred parking spaces. It supports modular configurations such as "six plus six" or "three plus three" to adapt to different building volumes. During construction, the honeycomb shell units are installed one by one to the preset lifting point positions using a hoisting method, resulting in a short construction period. The parking facility is equipped with a handling unit and a vertical storage stacker crane. After the user parks their vehicle at the designated handover location at the entrance, the handling unit transports the vehicle to the target parking space assigned by the system for storage. When retrieving the vehicle, a reservation mode is used, and the handling unit transports the vehicle from the parking space to the exit handover location. The parking facility also includes charging spaces and car wash spaces. The charging spaces are equipped with fast-charging ports, and the car wash spaces are one meter wider than regular spaces with wash brushes positioned above them. The car wash system incorporates water recycling technology to achieve water recovery. Solar power generation modules are installed on the garage roof to supply power to the handling units, gates, and core control equipment during power outages, ensuring basic vehicle storage and retrieval operations. A warehouse control system is also deployed within the parking facility. This system centrally schedules the operation of the handling units, stacker cranes, charging piles, and car wash equipment, and interacts with the access control system described in this application. Access control equipment, including retractable smart fences or rising bollard arrays, is installed between the parking area and the residential area to physically isolate and control the passageway between the two areas during shared open hours.

[0035] In the aforementioned parking facilities, parking spaces belong to fixed owners. During daytime commuting hours, a large number of owners' vehicles leave the parking spaces, resulting in periodic tidal vacancy. The parking management system opens some of the vacant owner parking spaces to external shared vehicles within a limited time period to improve the utilization rate of parking spaces.

[0036] This sharing model faces multiple risks in actual operation. First, there is the risk of conflicting fulfillment times: when a resident returns early due to a shortened schedule before the shared parking period ends, their designated parking space remains occupied by shared vehicles, leaving the resident's vehicle without a parking spot and creating a deadlock. If multiple residents return around the same time, the deadlock expands, as there are insufficient available parking spaces to accommodate all returning resident vehicles, leading to congestion within the parking area and even spilling over onto external roads. Second, there is the risk of uncontrolled access boundaries: existing shared parking solutions typically manage shared vehicle access using fixed time slots or simple threshold controls, statically locking the start and end times of the shared period. This makes it impossible to dynamically perceive the return time patterns of registered license plates for each resident's parking space, assess the coupled and overlapping conflict risks caused by multiple resident vehicles returning at the same time, and lack the ability to utilize currently vacant parking spaces that will not be occupied by residents in the short term as a dynamic and elastic buffer to absorb conflict risks. If the system over-accepts shared vehicles, a failure to fulfill obligations will occur when residents return in large numbers, leading to a lack of available buffer parking spaces. Conversely, overly conservative restrictions on the number of vehicles allowed will result in a large number of vacant parking spaces being wasted, reducing sharing revenue and residents' willingness to participate. Third, there is the risk of physical security isolation: residential communities and commercial areas have different security levels. Simply allowing external shared vehicles to enter residents' parking areas at the software level, without coordinating with dynamic isolation and control of physical access routes, could allow outsiders to enter residential areas through parking areas, posing a security risk.

[0037] The technical solution provided in this application addresses the aforementioned risks of performance timing conflicts and loss of access control. It calculates the return tendency by extracting historical return duration characteristics of registered license plates for each vacant owner's parking space using a weighted frequency method within a look-ahead window. The cumulative return tendency value of each owner's parking space currently occupied by shared vehicles is combined with the coupling conflict increment caused by overlapping return times of multiple parking spaces to obtain a conflict risk value. Simultaneously, candidate parking spaces for hedging are selected from currently vacant owner's parking spaces with a return tendency value below a safety threshold, and the effective hedging depth is accumulated. The system compares the difference between the effective hedging depth and the conflict risk value with a preset redundancy constant to determine whether to generate an admission or rejection instruction. During admission, the system further selects the optimal hedging pairing for shared parking spaces based on the return tendency and spatial proximity, and updates the hedging depth. When an owner returns to the site early and their parking space is in a shared occupied state, the system pushes temporary parking guidance to the owner based on the hedging pairing mapping, allowing the owner to go directly to the hedging parking space to complete the parking without having to stay and wait in the site. This achieves dynamic risk control and parking space hedging buffer management for shared vehicle entry.

[0038] To facilitate understanding, the method provided in this implementation will be described in detail below, using the above scenario as an example. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a shared parking space access control method in an embodiment of this application.

[0039] S101. Extract the time interval between each departure and return from the gate passage records of each registered license plate within the preset historical period, and count the frequency of occurrence in each time period according to the preset granularity to obtain the return time feature data corresponding to each registered license plate.

[0040] In this context, "registered license plate" refers to the license plate number of a vehicle that has established an ownership relationship with a specific owner's parking space in the parking management system. "Preset historical period" indicates the time range pre-set by the system for collecting traffic behavior samples, such as the last ninety calendar days. "Barrier gate passage record" refers to the data entry containing the license plate number and timestamp recorded by the entrance / exit barrier gate equipment when a vehicle passes through. "Departure-to-return time interval" represents the time difference between the departure time and the subsequent first entry time of the same registered license plate. "Preset granularity" refers to the smallest unit length used to divide the time axis into consecutive equal-length time periods, such as fifteen minutes. "Return time characteristic data" represents discrete distribution data composed of preset granularity as the scale and the frequency of occurrence in each time period interval as the numerical value.

[0041] Specifically, the system reads all exit and entry events for each registered license plate within a preset historical period from the traffic record database. It pairs each exit record of the same registered license plate with the first entry record immediately following it on the timeline, calculating the difference in timestamps to obtain a sample return duration. The system divides the timeline into continuous time intervals starting from zero according to a preset granularity, assigning each return duration sample to its corresponding interval and accumulating its frequency. Once all samples are archived, they constitute the return duration feature data for that registered license plate.

[0042] In some embodiments, the extraction and statistics of return duration feature data can be achieved in multiple ways: Optionally, the system arranges all passage events of the same registered license plate in ascending order of time, sequentially pairs adjacent exit and entry events to calculate the duration difference, eliminates abnormal pairings exceeding a preset upper limit, and bins the remaining samples according to a preset granularity and accumulates the frequency of each bin; Optionally, the system traverses the historical period on a daily basis, extracts the duration samples of each entry and exit pairing from the daily passage records, and summarizes them to count the frequency according to a preset granularity; Optionally, the system traverses the historical period week by week in a sliding window manner, counts the frequency in each week, and then takes the average value weighted by distance from the current time as the final frequency. It is understood that other methods can also be used to achieve the pairing and frequency statistics of return duration samples, which are not limited here.

[0043] S102. Based on the current departure time of the registered license plate corresponding to each vacant owner's parking space, read the frequency ratio of the time period covered by the preset look-ahead window from the return time feature data of the registered license plate to obtain the return tendency of each vacant owner's parking space.

[0044] Vacant owner-occupied parking spaces refer to owner-owned parking spaces where no vehicles are currently parked and the corresponding registered license plate has been driven away. Current departure time represents the time elapsed since the last departure of the registered license plate to the current system time. The preset look-ahead window is a fixed-length time interval extending along the positive time axis from a certain offset. Return tendency is a metric representing the probability that the vehicle will return to the parking lot within the future time period spanned by the look-ahead window.

[0045] Specifically, the system uses the current time elapsed since departure as the starting offset, adds the window length to obtain the ending offset, and extracts the frequency of occurrence of each time period between the two offsets from the return time feature data. Attenuation weights are assigned according to the distance of each time period from the starting offset, with smaller distances resulting in higher weights. Each frequency is multiplied by its corresponding weight and then summed to obtain the weighted total frequency. The frequencies of all time periods are accumulated to obtain the historical total frequency. The ratio of the weighted total frequency to the historical total frequency is the return tendency.

[0046] In some embodiments, the return tendency can be calculated in several ways: Optionally, the system applies an exponential decay function to each time period within the look-ahead window, with the decay factor increasing with the distance from the initial offset, and the return tendency is obtained by dividing the total weighted frequency by the total historical frequency; Optionally, the system applies a linearly decreasing weight to each time period, with the time period closest to the initial offset having a weight of one, decreasing arithmetically to a preset minimum value at the end of the window, and then calculating the weighted ratio to obtain the return tendency; Optionally, the system first performs subset filtering on the return duration feature data based on whether the current date is a weekday or a non-weekday, and only performs the above window extraction and weighting calculation on samples of the same type of date. It is understood that other methods can also be used to calculate the return tendency, which are not limited here.

[0047] In some embodiments, this step specifically includes:

[0048] The starting offset is the current departure time of the registered license plate corresponding to each vacant owner's parking space. The ending offset is calculated based on the starting offset and the window length of the preset look-ahead window. The frequency of occurrence of each time period between the starting offset and the ending offset is extracted from the return time feature data of the registered license plate. The attenuation weight is assigned to each time period according to the distance between each time period and the starting offset. The attenuation weight decreases as the distance increases. The weighted frequency total is calculated based on the frequency of occurrence of each time period and the corresponding attenuation weight. The frequency of occurrence of all time periods in the return time feature data is accumulated to obtain the historical frequency total. The ratio of the weighted frequency total to the historical frequency total is used as the return tendency of vacant owner's parking spaces.

[0049] The initial offset refers to the cumulative time elapsed from when a registered vehicle with a specific vacant parking space leaves the parking lot until the current time. This value is directly used as the index coordinate for locating the left endpoint of the extraction interval in the return time feature data. For example, if a vehicle leaves at 14:00 on the current day and the current time is 17:00, then the initial offset is 180 minutes. The final offset is the time coordinate value obtained by adding the initial offset to the preset lookahead window length. It is used to locate the right endpoint of the extraction interval. For example, if the lookahead window length is 60 minutes, then the final offset is 240 minutes. The decay weight is a multiplicative coefficient allocated based on the time interval distance between each time interval and the initial offset. The larger the distance, the smaller the coefficient, indicating that the time interval farther from the current time contributes less to the return probability judgment. For example, when using an exponential decay function, the weight of the first time interval is 1.0, the weight of the second time interval is 0.8, and the weight of the third time interval is 0.64. The weighted frequency total is obtained by multiplying the occurrence frequency of each time period within the extraction interval by the corresponding attenuation weight and then summing the results. The historical frequency total is the total number of occurrences obtained by directly summing the occurrence frequencies of all time periods in the return duration feature data, and is used as the normalization denominator to transform the weighted frequency total into a ratio. The return tendency is the dimensionless ratio obtained by dividing the weighted frequency total by the historical frequency total. The larger the value, the higher the probability that the registered license plate vehicle will return in the future time period covered by the look-ahead window.

[0050] The system performs the following calculation process for each vacant owner's parking space: First, it reads the current departure time of the vehicle with the corresponding registered license plate for that parking space and assigns it as the starting offset. Then, it adds the starting offset to the window length of a preset look-ahead window to obtain the ending offset, thus determining the closed interval for extracting data on the return time feature data time axis. Next, it locates the time period number corresponding to the starting offset and the time period number corresponding to the ending offset in the return time feature data for that registered license plate, and reads out the frequency of occurrence for each time period between the two numbers. Finally, it assigns attenuation weights to each time period within the extraction interval. Specifically, it calculates the number of time periods between each time period and the starting offset as the distance value, substitutes the distance value into a preset attenuation function to obtain the corresponding weight. The attenuation function uses an exponential form, meaning the weight is equal to the distance value raised to the power of the attenuation base. For example, if the attenuation base is 0.85 and the distance value is 3, the weight is 0.614. The system multiplies the frequency of occurrence in each time period by its corresponding attenuation weight and sums them up to obtain the weighted total frequency. Then, it directly sums all the occurrence frequencies from the first time period to the last time period in the registered license plate return duration feature data to obtain the historical total frequency. Finally, the weighted total frequency is divided by the historical total frequency to obtain the return tendency value of the vacant owner's parking space. This value reflects the intensity of the return of registered license plate vehicles within the time range from the current moment to the end of the lookahead window. The closer the value is to 1, the higher the return intensity and the greater the urgency of the parking space being reoccupied by the owner's vehicle.

[0051] S103. Summarize the return tendency of all owner parking spaces currently occupied by shared vehicles to obtain the conflict risk value. Mark the owner parking spaces with a return tendency lower than the preset safety threshold and not occupied by shared vehicles as hedging candidate parking spaces. Calculate the effective hedging depth based on the preset benchmark value and the return tendency of each hedging candidate parking space.

[0052] The conflict risk value represents the aggregated probability measure of the return of all owner-occupied parking spaces within the forward-looking window. The preset safety threshold is a critical value used to determine whether the short-term return risk of a particular vacant parking space is sufficiently low. Hedging candidate parking spaces represent vacant owner-occupied parking spaces with a return tendency below the safety threshold, serving as temporary buffer resources. The preset benchmark value is a reference constant representing the maximum hedging capacity that a single hedging candidate parking space can provide. The effective hedging depth refers to the total hedging capacity provided by all hedging candidate parking spaces.

[0053] Specifically, the system iterates through all owner-occupied parking spaces currently occupied by shared vehicles. For any two occupied parking spaces, it calculates the overlap of return periods based on the frequency distribution of each time period within their respective look-ahead windows. The return tendency of each occupied parking space is summed to obtain the base conflict quantity. The overlap of each pair of parking spaces is weighted and summed according to a preset coupling coefficient to obtain the coupling conflict increment. The two are then added together to obtain the conflict risk value. Subsequently, the system iterates through currently vacant owner-occupied parking spaces not occupied by shared vehicles, marking those with a return tendency below a preset safety threshold as hedging candidate parking spaces. For each hedging candidate parking space, the difference between a preset baseline value and its return tendency is calculated. All differences are summed to obtain the effective hedging depth.

[0054] In some embodiments, the calculation of conflict risk value and effective hedging depth can be achieved in several ways: Optionally, the system sums the minimum frequency values ​​of each pair of occupied parking spaces in each time period within the look-ahead window as the overlap, multiplies this by the coupling coefficient, and accumulates the sum to obtain the coupled conflict increment. This is then added to the basic conflict quantity to obtain the conflict risk value, and the hedging depth is directly accumulated based on the difference. Optionally, the system extracts the normalized frequency vectors of each pair of occupied parking spaces, calculates the inner product as the overlap, and obtains the conflict risk value accordingly. In the hedging depth calculation, a reduction coefficient related to the proximity of the safety threshold is introduced for each candidate parking space. Optionally, when calculating the basic conflict quantity, the system adjusts the return tendency of each occupied parking space according to the length of time the parking space has been occupied, with a higher weight for longer occupancy. The calculation method for the coupled conflict increment is the same as described above. It is understood that other methods can also be used to achieve the above aggregation calculation, which are not limited here.

[0055] In some embodiments, this step specifically includes: taking any pair of owner parking spaces from those currently occupied by shared vehicles; determining the overlap of the return time periods of the pair of owner parking spaces based on the frequency distribution of each owner parking space in the time period covered by a preset forward window; accumulating the return tendency of each owner parking space currently occupied by shared vehicles to obtain a basic conflict amount; determining the coupling conflict increment based on the overlap of the return time periods of each pair of owner parking spaces; adding the basic conflict amount and the coupling conflict increment to obtain a conflict risk value; traversing all currently vacant owner parking spaces that are not occupied by shared vehicles; marking owner parking spaces with a return tendency lower than a preset safety threshold as hedging candidate parking spaces; and accumulating the effective hedging depth based on the difference between the preset benchmark value and the return tendency of each hedging candidate parking space.

[0056] Owner-owned parking spaces currently occupied by shared vehicles refer to the set of owner-owned parking spaces that have been entered and parked by shared vehicles, and the registered license plates of these parking spaces are in a vacant state. Any pair of owner-owned parking spaces is a pair of units containing two owner-owned parking spaces selected from this set in a pairwise combination manner. For example, if there are three parking spaces A, B, and C in the set, then there are three pairs of combinations AB, AC, and BC. The overlap of return time periods is a numerical indicator that measures the degree of overlap in the return time distribution of the two registered license plates in a pair of owner-owned parking spaces within the look-ahead window coverage period. The higher the value, the stronger the trend of the two vehicles returning to the parking space at the same time. The basic conflict quantity is the value obtained by directly summing the return tendency of each owner-owned parking space currently occupied by shared vehicles. It represents the overall conflict intensity without considering the coupling relationship of return time between vehicles. The coupled conflict increment is the additional conflict compensation amount calculated based on the overlap of return time periods of each pair of owner-owned parking spaces. It reflects the superimposed conflict risk caused by multiple owner-owned vehicles returning to the parking space at the same time. The conflict risk value is a comprehensive risk measure obtained by adding the basic conflict quantity and the coupled conflict increment. The preset safety threshold is a pre-defined upper limit for the return tendency. Vacant owner parking spaces with a return tendency below this threshold are considered to have low owner vehicle return intensity in the short term. Hedging candidate parking spaces are currently vacant owner parking spaces that meet the condition of a return tendency below the preset safety threshold and are not occupied by shared vehicles; these serve as optional resources for subsequent hedging pairing. The effective hedging depth is the total hedging capacity value obtained by summing the hedging contributions of each hedging candidate parking space, where the hedging contribution of each candidate parking space equals a preset benchmark value minus the return tendency of that parking space.

[0057] The system first determines the set of owner parking spaces currently occupied by shared vehicles. From this set, all possible owner parking spaces are paired up. For each pair of owner parking spaces, the system calculates the overlap of the return time period: it reads the frequency of the registered license plates of the two owner parking spaces in each time period within the coverage of the forward window. The system takes the smaller value of the frequency of the two parking spaces in the same time period and adds them up in each time period to obtain the total overlap frequency. Then, it divides the total overlap by the average of the sum of the frequencies of the two parking spaces. For example, if the frequency of parking space A in the forward window is 3, 5, 2, 1 in 4 time periods, and the corresponding frequency of parking space B is 1, 4, 6, 0, then the smaller value of each time period is 1, 4, 2, 0, which is added up to 7. The sum of the frequencies of A and B is 11, and the average is 11. The overlap is 7 divided by 11, which equals 0.636. Next, the system sums the return tendency of each owner's parking space currently occupied by shared vehicles to obtain the basic conflict amount. Then, it weights the overlap of all paired return time periods according to a preset coupling coefficient and accumulates them to obtain the coupled conflict increment. The basic conflict amount and the coupled conflict increment are added together to obtain the conflict risk value. The system then iterates through all owner's parking spaces that are currently vacant and not occupied by shared vehicles in the parking lot, comparing the return tendency of each parking space with a preset safety threshold. Parking spaces with a return tendency lower than the preset safety threshold are marked as hedging candidate parking spaces and included in the hedging candidate set. Finally, the system calculates the difference between a preset benchmark value and the return tendency of each parking space in the hedging candidate set as the hedging contribution of that parking space. All hedging contributions are accumulated to obtain the effective hedging depth, which represents the maximum conflict buffer capacity that the currently available hedging resources can bear.

[0058] S104. When a shared vehicle triggers an entry request, the difference between the effective hedging depth and the conflict risk value is compared with a preset redundancy constant to obtain a first difference. When the first difference is not less than the preset redundancy constant, an access instruction is generated, and a hedging pairing mapping is established between the hedging candidate parking space and the allocated shared parking space.

[0059] The entry request refers to the entry signal sent to the system by the license plate recognition device after collecting information when a shared vehicle arrives at the entrance gate. The first difference represents the effective hedging depth minus the conflict risk value. The preset redundancy constant refers to the system's preset safety margin value. The access command is a control signal used to indicate permission for the gate to open and allow passage. Hedging pairing mapping refers to establishing a one-to-one association between a candidate parking space number and a resident parking space number occupied by a shared vehicle.

[0060] Specifically, the system calculates the difference between the effective hedging depth and the conflict risk value to obtain a first difference value, which is then compared with a preset redundancy constant. If the first difference value is not less than the preset redundancy constant, the system generates an access command and triggers the gate to open. Simultaneously, the system obtains the spatial proximity between each hedging candidate parking space and the allocated shared parking space, calculates a matching score by combining the return tendency and spatial proximity, and selects the parking space with the highest score as the target hedging parking space. The target hedging parking space and the shared parking space are associated and written into the hedging matching mapping table, removed from the candidate pool, and the effective hedging depth is deducted based on the difference between the preset benchmark value and the return tendency of that parking space.

[0061] In some embodiments, the selection and pairing of target hedging parking spaces can be achieved in several ways: Optionally, the system calculates the spatial proximity by weighting the floor difference and horizontal distance between each candidate parking space and the shared parking space, multiplies the reciprocal of the return tendency by the spatial proximity to obtain the pairing adaptation score, selects the highest score to establish a mapping, and updates the hedging depth; Optionally, the system arranges each candidate parking space in ascending order of return tendency, filters the subset of spaces on the same or adjacent floors as the shared parking space, and selects the one with the lowest return tendency to establish a mapping; Optionally, the system calculates the spatial proximity score and return safety score of each candidate parking space respectively, sums them according to preset weights to obtain a comprehensive score, and selects the highest score to establish a mapping. It is understood that other methods can also be used to establish the hedging pairing mapping, which are not limited here.

[0062] In some embodiments, the step specifically includes: in response to a first difference not being less than a preset redundancy constant, generating an admission instruction, obtaining the spatial proximity between each hedging candidate parking space and the allocated shared parking space, comprehensively determining the pairing adaptation score of each hedging candidate parking space based on the return tendency of each hedging candidate parking space and the corresponding spatial proximity, selecting the hedging candidate parking space with the highest pairing adaptation score as the target hedging parking space, associating the target hedging parking space with the allocated shared parking space and writing it into the hedging pairing mapping table, removing the target hedging parking space from the hedging candidate parking spaces, and updating the effective hedging depth based on the difference between the preset benchmark value and the return tendency of the target hedging parking space.

[0063] The first difference is the arithmetic result of subtracting the conflict risk value from the effective hedging depth, used to measure whether the current hedging buffer margin is sufficient to cover the conflict risk. The preset redundancy constant is a pre-set fixed threshold, serving as the minimum threshold for determining whether the hedging buffer margin is sufficient. The access command is the gate opening control signal generated and sent to the entrance gate controller after the system determines that the hedging buffer is sufficient. Spatial proximity is a metric obtained by normalizing the physical distance between the candidate hedging parking space and the allocated shared parking space; the closer the distance, the higher the spatial proximity. For example, the spatial proximity is 0.9 when two parking spaces are 2 spaces apart, and 0.5 when they are 10 spaces apart. The pairing suitability score is a comprehensive evaluation value obtained by combining the return tendency and spatial proximity according to preset weights; the higher the score, the more suitable the candidate hedging parking space is as a hedging pairing object. The target hedging parking space is the one with the highest matching score among all hedging candidate parking spaces. Once selected, it establishes a one-to-one hedging binding relationship with the allocated shared parking space. The hedging pairing mapping table is a data structure that records the correspondence between the shared parking space number and its bound target hedging parking space number, used for quickly locating the hedging transfer target when the owner's vehicle returns to the parking lot. The preset baseline value is a pre-set constant representing the upper limit of the hedging contribution integrity, for example, a value of 1.0.

[0064] After comparing the first difference with the preset redundancy constant, the system immediately generates an admission command if the first difference is not less than the preset redundancy constant. The system then reads the location coordinates of all current candidate parking spaces and the location coordinates of the assigned shared parking spaces, calculates the physical distance between each candidate parking space and the assigned shared parking space, divides the physical distance by the maximum parking space spacing in the parking lot to obtain a normalized distance value, and then subtracts the normalized distance value from 1 to obtain the spatial proximity, ensuring that the smaller the physical distance, the larger the spatial proximity value. Next, the system calculates a matching score for each hedging candidate parking space. Specifically, it subtracts the candidate parking space's return tendency from a preset baseline value to obtain a return safety margin. This return safety margin is then multiplied by a preset first weighting coefficient, and the spatial proximity is multiplied by a preset second weighting coefficient. The sum of these two products yields the matching score for the candidate parking space. For example, if the return safety margin is 0.85, the first weighting coefficient is 0.7, the spatial proximity is 0.9, and the second weighting coefficient is 0.3, then the matching score is 0.85 multiplied by 0.7 plus 0.9 multiplied by 0.3, which equals 0.865. The system sorts all hedging candidate parking spaces in descending order of their matching scores, selects the top-ranked candidate parking space as the target hedging parking space, writes the target hedging parking space number and the allocated shared parking space number as key-value pairs into the hedging matching mapping table, and removes the target hedging parking space from the hedging candidate parking space set so it no longer participates in subsequent matching. Finally, the system subtracts the return tendency of the target hedging parking space from the preset benchmark value to obtain the hedging contribution of the parking space, subtracts the hedging contribution from the current effective hedging depth to obtain the updated effective hedging depth value, and writes it back to storage.

[0065] In some embodiments, when the first difference is less than a preset redundancy constant, a rejection instruction is generated and the entrance gate is prevented from opening.

[0066] After calculating the first difference, the system compares it with a preset redundancy constant. If the first difference is less than the preset redundancy constant, it determines that the current hedging buffer is insufficient to accommodate new shared vehicles while maintaining a safety margin. In this case, the system immediately generates a rejection command. This rejection command is sent to the entrance gate controller via the communication channel. Upon receiving the rejection command, the gate controller keeps the gate arm in the lowered state and does not perform the lifting action, thus physically preventing the shared vehicle that triggered the entry request from entering the parking lot. At the same time, the system pushes the rejection reason identifier carried in the rejection command to the terminal associated with the driver of the shared vehicle, and the terminal interface displays the current status information that entry is not possible. This mechanism ensures that no new shared vehicles will enter and occupy the owner's parking space when the hedging buffer is insufficient, thereby avoiding the conflict scenario where there are no parking spaces available when the owner's vehicle returns due to the depletion of hedging resources. For example, when the effective hedging depth is 1.2, the conflict risk value is 0.9, and the preset redundancy constant is 0.5, the first difference is 0.3, which is less than 0.5, and the system immediately generates a rejection command and prevents the gate from opening.

[0067] "Owner's vehicle returning to the parking lot" refers to the event where the license plate recognition equipment at the entrance gate detects that a vehicle with a specific registered license plate has arrived at the entrance. "Shared occupancy status" indicates that the owner's parking space is currently occupied by a shared vehicle. "Returning owner associated terminal" refers to the mobile communication device linked to the owner's account for that registered license plate. "Temporary parking guidance" is a guidance message containing the offset parking space number and location information, instructing the owner to temporarily park their vehicle in that offset parking space.

[0068] Specifically, after the entrance gate recognizes the owner's license plate, the system queries the current status of the parking space corresponding to that license plate. If the status is shared occupancy, the system uses the parking space number as the key to query the corresponding offset parking space number in the offset matching mapping table, extracts the floor and area information of the offset parking space from the parking space database, assembles the above information into a temporary parking guidance message, sends it to the owner's associated terminal through the push channel, and simultaneously sends an opening command to the gate to allow the owner's vehicle to enter.

[0069] In some embodiments, temporary parking guidance can be pushed out in multiple ways: Optionally, the system queries the mapping table to obtain the parking space number, reads its floor and area number, generates a guidance message containing text description, pushes it to the owner's terminal, and opens the gate; Optionally, the system queries the mapping table to obtain the parking space number, calculates the driving route based on the path from the entrance to the parking space, encapsulates the parking space number and route instructions into a guidance message, pushes it to the terminal, and triggers in-park navigation; Optionally, after querying the mapping table, the system simultaneously sends a directional indicator signal to the in-park electronic guidance screen and pushes text guidance to the owner's terminal, guiding the owner to the parking space through a dual-channel approach. It is understood that other methods can also be used to generate and push temporary parking guidance, which are not limited here.

[0070] S105. When the entrance gate detects that the owner's vehicle has returned to the parking lot and the owner's parking space to which the vehicle belongs is in a shared occupied state, it reads the corresponding offset parking space number in the offset pairing mapping and sends a temporary parking instruction containing the offset parking space number to the owner's associated terminal.

[0071] In some embodiments, when the access control system interfaces with a transport unit scheduling system within a parking facility, the method further includes the following supporting steps at the physical execution layer:

[0072] When the occupancy time of a shared vehicle in a resident's parking space reaches a preset warning ratio exceeding the allowed parking time limit, the access control system sends a reminder to the driver's associated terminal to leave. If the shared vehicle remains after the occupancy time reaches the allowed parking time limit, the access control system sends a transfer command to the warehouse control system. The warehouse control system's dispatch and handling unit physically moves the shared vehicle from the occupied resident's parking space to a preset edge-separated independent recycling area, restoring the occupied resident's parking space to a usable state. The edge-separated independent recycling area is a designated independent area within the parking facility, physically isolated from the resident's parking space area. Shared vehicle drivers must go to this area to retrieve their vehicles. The aforementioned physical transfer operation is executed autonomously by the handling unit after receiving the transfer instruction. The handling unit moves the target vehicle from the owner's parking space to an vacant recycling station in the edge independent recycling area along a preset track or navigation path. After the transfer is completed, it sends a signal to the warehouse control system to indicate that the parking space is in place. The warehouse control system updates the status of the owner's parking space from shared occupied to vacant and synchronizes it to the access control system. The access control system then reinstates the parking space into the screening range of the offset candidate parking spaces or returns it to the owner's available status.

[0073] In some embodiments, when the access control system generates the access instruction and allows the shared vehicle to enter, the method further includes a physical security isolation linkage step:

[0074] The access control system sends isolation commands to the access isolation devices deployed within the parking facility. These devices include retractable smart fences or rising bollard arrays positioned between the parking area and the residential area. The isolation commands control the access isolation devices to alter the terrain of the access route, physically creating a dedicated passage leading only to the shared open area. Simultaneously, the fireproof roller shutters and access control devices leading to the residential elevator lobby are locked at the physical level, blocking access for outsiders from entering the residential area via the parking area. When all shared vehicles leave the parking facility and the shared open period ends, the access control system sends a release command to the access isolation devices, restoring normal traffic flow to the parking area. The control signals for the access isolation devices are automatically generated by the access control system based on the current number of shared vehicles and the shared open period status, requiring no manual intervention. When a new shared vehicle enters during the shared open period, the access control system checks whether the access isolation devices are already in an isolated state while issuing an access command. If not, it sends an isolation command before opening the entrance gate.

[0075] In some embodiments, when allocating parking spaces for shared vehicles, the access control system also selects whether to allocate the vehicle to a charging parking space or a car wash parking space based on the estimated parking duration and vehicle type:

[0076] When the estimated parking time of a shared vehicle is not less than a preset charging threshold and the shared vehicle is a new energy vehicle, the access control system prioritizes selecting a target parking space from those with fast charging interfaces and adds a charging service identifier to the access instruction. The warehouse control system then activates the charging pile to perform the charging operation after the transport unit delivers the vehicle to the charging parking space. When a shared vehicle driver selects a car wash service through a linked terminal upon entry, the access control system selects a target parking space from the car wash parking spaces. The warehouse control system then activates the car wash brushes and water circulation system to perform the car wash operation after the transport unit delivers the vehicle to the car wash parking space. The width of the car wash parking space is one meter wider than that of a regular parking space to accommodate the lateral movement of the car wash brushes. When the charging and car wash parking spaces are not used for charging or car wash services, they are also included in the selection range of hedging candidate parking spaces and participate in the calculation of return tendency and the accumulation of effective hedging depth.

[0077] In some embodiments, the access control system triggers an automatic charging process simultaneously when generating the access instruction: determining the corresponding billing rate based on the type of the allocated parking space, writing the entry time, maximum allowed parking time, rate standard, and parking space number into the billing record, reading the time difference between the entry time and the departure time when the shared vehicle leaves, calculating the chargeable fee according to the rate, and adding the corresponding service fee if charging or car wash services were used during the parking period, and deducting the total fee through the payment channel of the shared vehicle driver's associated terminal.

[0078] In some embodiments, the parking facility is equipped with solar power generation modules on its roof, which power the handling unit, gate equipment, and the core processor of the access control system via a battery pack. When the mains power is interrupted, the access control system switches to a battery-powered mode. In this mode, only the gate control, license plate recognition, hedging pairing mapping query, and basic handling unit scheduling functions are retained. The batch update calculation of return tendency and the time-by-time extrapolation of the predicted effective hedging depth are suspended to reduce power consumption and ensure that the basic access operations of the owner's vehicle are not affected.

[0079] The following provides a more detailed description of the process of the method provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the shared parking space access control method in this application embodiment.

[0080] S201. Based on the return time characteristic data of the registered license plates corresponding to each hedging candidate parking space, reread the frequency ratio of the time period covered by the preset forward window in the continuous time period after the current time to obtain the predicted return tendency of each hedging candidate parking space in each time period.

[0081] A continuous time period refers to a sequence of equally spaced time segments arranged sequentially along the positive time axis from the current moment, according to a preset granularity. For example, if the current moment is 20:00 and the preset granularity is 15 minutes, then the continuous time periods would be 20:00 to 20:15, 20:15 to 20:30, and so on. The predicted return tendency refers to the probability measure of returning to the market obtained by re-executing the lookahead window frequency percentage extraction using the starting point of a future time period as an assumption of the already lost time.

[0082] The system takes each hedging candidate parking space as an object, adds the current departure time to the offset between the first time period and the current time as the starting offset, and superimposes it with a preset look-ahead window length to obtain the ending offset. It then extracts the frequency of each time period between the starting and ending offsets from the return time feature data of the registered license plate corresponding to that parking space, weights it according to attenuation weights, and compares it with the total historical frequency to obtain the predicted return tendency for the first time period. Subsequently, the starting offset is incremented by one granularity, and the above extraction and weighting calculation are repeated to obtain the predicted return tendency for the second time period, the third time period, and up to the end of the preset calculation span. After all hedging candidate parking spaces have completed the time-by-time calculation, a two-dimensional data table is output, with the parking space number and time period number as dual indices and the predicted return tendency as the numerical value.

[0083] S202. In each time period, exclude the hedging candidate parking spaces whose predicted return tendency exceeds the preset safety threshold, and recalculate the predicted effective hedging depth for each time period based on the predicted return tendency of the remaining hedging candidate parking spaces and the preset benchmark value.

[0084] Exclusion refers to removing hedging candidate positions whose predicted return tendency values ​​exceed a preset safety threshold from the set of effective hedging contributors for a given period, thus preventing them from participating in the accumulation calculation of hedging depth for that period. Predicted effective hedging depth refers to the predicted hedging capacity that can be collectively provided by all hedging candidate positions that still meet the safety threshold constraint within a future period. For example, if the predicted return tendency values ​​of three candidate positions are 0.05, 0.12, and 0.36 respectively within a given period, and the safety threshold is 0.3, then the first two positions are retained for calculation, while the third is excluded.

[0085] The system iterates through the two-dimensional data table generated in the previous step, selecting a subset of candidate parking spaces for hedging within each time period whose predicted return tendency does not exceed a preset safety threshold. For each parking space within the subset, the hedging contribution of a single parking space is obtained by subtracting the predicted return tendency of that parking space from the preset baseline value for that time period. The hedging contributions of all single parking spaces within the subset are then summed to obtain the predicted effective hedging depth for that time period. If the predicted return tendency of all candidate parking spaces for hedging exceeds the safety threshold within a certain time period, the predicted effective hedging depth for that time period is recorded as 0. The final output is a one-dimensional sequence indexed by the time period number and with the predicted effective hedging depth as the numerical value.

[0086] S203. Subtract the predicted effective hedging depth from the conflict risk value for each time period, and determine the time period when the difference first falls below the preset redundancy constant as the hedging exhaustion period.

[0087] The difference operation refers to the arithmetic operation of subtracting the conflict risk value from the predicted effective hedging depth value for the same time period to obtain the difference result. The hedging exhaustion period refers to the period when the difference value first falls from not lower than the preset redundancy constant to lower than the preset redundancy constant when traversing the difference value sequence of each time period in the positive direction of the time axis. This indicates that the hedging buffer margin is insufficient to cover the conflict risk for the first time in this period. For example, if the differences for the 1st to 4th periods are 1.2, 0.9, 0.6 and 0.3 respectively, and the preset redundancy constant is 0.5, then the 4th period is the hedging exhaustion period.

[0088] The system starts from the first time period in the time series, reads the predicted effective hedging depth for each time period, and subtracts the current conflict risk value to obtain the difference. This difference is compared to a preset redundancy constant. If the difference is not lower than the preset redundancy constant, the system continues to the next time period. If the difference is lower than the preset redundancy constant, the current time period is marked as a hedging exhaustion period, and the iteration terminates. If the difference is still not lower than the preset redundancy constant when the iteration reaches the end of the preset calculation span, it indicates that the hedging buffer is always sufficient within the calculation span, and the hedging exhaustion period is marked as the next time period after the end of the calculation span. This marking result serves as input for subsequent calculations of the upper limit of the allowed parking time.

[0089] S204. Determine the upper limit of the allowed parking time based on the time interval between the hedging exhaustion period and the current period, and attach the upper limit of the allowed parking time to the access instruction.

[0090] The time interval refers to the number of complete granularities contained between the start time of the hedging exhaustion period and the current time. Multiplying this number by the duration of a single granularity yields the span value of the time dimension. The maximum allowed parking time refers to the maximum allowed occupancy time allocated to a shared vehicle about to enter the site. The shared vehicle must leave before the expiration of this time; otherwise, the AGV physical transfer and recycling process will be triggered. For example, if the hedging exhaustion period is the 8th time interval with a granularity of 15 minutes, then the maximum allowed parking time is 120 minutes.

[0091] The system reads the difference between the time period number of the offset depletion period and the current time period number, multiplies the difference by a preset granularity of time length to obtain the maximum allowed parking time. The system writes this time value into the time limit field of the access command. The access command is sent to the entrance gate controller via the communication channel to trigger the gate opening action. At the same time, the time limit field is pushed to the shared vehicle driver's associated terminal to display the remaining parking time. When the gate opens, the gate controller writes the entry time and the maximum allowed parking time into the occupancy monitoring record. When the occupancy time reaches the preset warning ratio of the maximum threshold, a reminder to leave is triggered. If the vehicle has not left after reaching the maximum, the AGV is triggered to physically move the vehicle from the owner's parking space to the edge independent recycling area.

[0092] S205. Based on the return time characteristic data of the registered license plates corresponding to the parking spaces occupied by shared vehicles, the frequency ratio of the time periods covered by the preset look-ahead window is reread in each continuous time period after the current time to obtain the predicted return tendency of each owner's parking space occupied by shared vehicles in each time period.

[0093] A resident's parking space currently occupied by a shared vehicle refers to a resident's exclusive parking space marked as "shared occupied" at the current moment, where the registered vehicle has left but a shared vehicle from outside is parked in the space. This step calculates resident parking spaces currently occupied by shared vehicles to predict the probability trend of these occupied resident vehicles returning to the parking space in future time periods.

[0094] The system takes each owner's parking space currently occupied by a shared vehicle as the object, reads the current departure time of the registered license plate corresponding to each parking space, adds the offset of the first time period to the current time as the starting offset, and superimposes the lookahead window length to obtain the ending offset. It then extracts the frequency of each time period between the starting and ending offsets from the return time feature data of the registered license plate, weights it according to attenuation weights, and compares it with the total historical frequency to obtain the predicted return tendency of the first time period. The starting offset is incremented by granularity and the calculation is repeated until the end of the preset calculation span, resulting in a two-dimensional data table of the predicted return tendency of each occupied owner's parking space for each time period.

[0095] S206. The predicted return tendency of each owner's parking space currently occupied by shared vehicles is summed up to obtain the predicted conflict risk value for each time period.

[0096] Accumulation refers to adding the predicted return tendency values ​​of all owner parking spaces currently occupied by shared vehicles within the same time period to obtain the aggregated result for that time period. The predicted conflict risk value is an aggregated prediction measure of the probability of conflict when all occupied owner parking spaces return to their respective parking spaces in a future time period. For example, if the predicted return tendency values ​​of 3 occupied parking spaces are 0.2, 0.35, and 0.15 respectively in a certain time period, then the predicted conflict risk value for that time period is 0.7.

[0097] The system iterates through the aforementioned two-dimensional data table for each time period, reading the predicted return tendency values ​​of all occupied parking spaces within each period, and directly summing them to obtain the basic predicted conflict risk value for that period. In scenarios where the overlapping coupling effect of return periods needs to be considered, the predicted return tendency values ​​of any two occupied parking spaces within the same period are multiplied and weighted by a preset coupling coefficient. The weighted sum of all paired products is then added to the basic accumulated value as the coupling increment to obtain the final predicted conflict risk value. After calculation for all time periods, a one-dimensional sequence is output, indexed by the time period number and containing the predicted conflict risk value.

[0098] S207. Subtract the effective hedging depth from the predicted conflict risk value corresponding to each time period, and determine the time period when the difference is not less than the preset redundancy constant for the first time as the expected recovery period.

[0099] The expected recovery period refers to the period when, when traversing the difference sequence of each time period along the positive direction of the time axis, the difference between the effective hedging depth and the predicted conflict risk value first rises from below the preset redundancy constant to above the preset redundancy constant. This marks the first time that the hedging buffer margin has recovered to a sufficient level to accommodate new shared vehicles during this period. For example, if the differences for the 1st to 5th time periods are 0.1, 0.2, 0.3, 0.4 and 0.6 respectively, and the preset redundancy constant is 0.5, then the 5th time period is the expected recovery period.

[0100] The system starts from the first time period in the time series, reads the predicted conflict risk value for each time period, and subtracts the predicted conflict risk value from the current effective hedging depth to obtain the difference. This difference is compared to a preset redundancy constant. If the difference is lower than the preset redundancy constant, the system continues to the next time period; if the difference is not lower than the preset redundancy constant, the current time period is marked as a predicted recovery period and the iteration terminates. If the difference is still lower than the preset redundancy constant at the end of the preset calculation span, the predicted recovery period is marked as unpredictable, and the expected waiting time field in the subsequently generated queuing instructions is filled with a message indicating that the expected waiting time exceeds the estimated range.

[0101] S208. Generate an estimated waiting time based on the time interval between the expected recovery period and the current period, and send a queuing instruction containing the estimated waiting time to the shared vehicle that triggered the entry request.

[0102] The estimated waiting time refers to the time span between the start time of the estimated recovery period and the current time, representing the estimated time required for a shared vehicle to wait from triggering an entry request to the restoration of access capability through the offset buffer. The queuing instruction is a notification message containing waiting time information sent to the driver's associated terminal after immediate entry is denied. For example, if the estimated recovery period is the 6th period with a granularity of 15 minutes, then the estimated waiting time is 90 minutes.

[0103] The system reads the difference between the expected recovery period number and the current period number, multiplies the difference by a preset granularity of time length to obtain the expected waiting time. The system encapsulates this waiting time value into a queuing instruction and sends it via the communication channel to the terminal associated with the driver of the shared vehicle who triggered the entry request. The terminal interface displays the expected waiting time and the expected entry time. Simultaneously, the system writes the shared vehicle's entry request into the queue along with an expected recovery period marker. When the difference between the effective hedging depth and the conflict risk value rises to a level not lower than a preset redundancy constant, the system resends the access instruction to the first shared vehicle in the queue according to the queue order and triggers the opening of the entrance gate.

[0104] The shared parking space access control system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a shared parking space access control system in an embodiment of this application.

[0105] It should be noted that, Figure 3 The structure of the shared parking space access control system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0106] like Figure 3 As shown, the shared parking space access control system includes a CPU 301, which can perform various appropriate actions and processes according to a program stored in ROM 302 or a program loaded from storage section 308 into RAM 303, such as executing the methods described in the above embodiments. RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via bus 304. I / O interface 305 is also connected to bus 304.

[0107] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including hard disks, etc.; and communication section 309 including network interface cards such as LAN (Local Area Network) cards, modems, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0108] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0109] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0110] Specifically, the shared parking space access control system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the shared parking space access control method provided in the above embodiment.

[0111] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the shared parking space access control system described in the above embodiments; or it may exist independently and not incorporated into the shared parking space access control system. The storage medium carries one or more computer programs, which, when executed by a processor of the shared parking space access control system, cause the shared parking space access control system to implement the shared parking space access control method provided in the above embodiments.

[0112] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. 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 scope of the technical solutions of the embodiments of this application.

[0113] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

Claims

1. A shared parking space access control method, characterized in that, The method, applied to a shared parking space access control system, includes: Extract the time interval between each departure and return from the gate passage records of each registered license plate within a preset historical period, and statistically analyze the frequency of occurrence in each time period according to a preset granularity to obtain the return time feature data corresponding to each registered license plate. Based on the current departure time of the registered license plate corresponding to each vacant owner's parking space, the frequency ratio of the time period covered by the preset look-ahead window is read from the return time feature data of the registered license plate to obtain the return tendency of each vacant owner's parking space. The return tendency of all owner parking spaces currently occupied by shared vehicles is aggregated to obtain a conflict risk value. Owner parking spaces with a return tendency lower than a preset safety threshold and not occupied by shared vehicles are marked as hedging candidate parking spaces. The effective hedging depth is calculated based on the preset benchmark value and the return tendency of each hedging candidate parking space. When a shared vehicle triggers an entry request, the difference between the effective hedging depth and the conflict risk value is compared with a preset redundancy constant to obtain a first difference. When the first difference is not less than the preset redundancy constant, an access instruction is generated, and a hedging pairing mapping is established between the hedging candidate parking space and the allocated shared parking space. When the entrance gate detects that a resident's vehicle has returned to the parking lot and the resident's parking space to which the vehicle belongs is in a shared occupied state, it reads the corresponding offset parking space number in the offset pairing mapping and sends a temporary parking instruction containing the offset parking space number to the resident's associated terminal.

2. The method according to claim 1, characterized in that, The step of reading the frequency percentage of the time period covered by the preset look-ahead window from the return duration feature data of the registered license plates to obtain the return tendency of each vacant owner's parking space specifically includes: The starting offset is calculated based on the current departure time of the registered license plate corresponding to each vacant owner's parking space, and the ending offset is calculated based on the starting offset and the window length of the preset forward window. Extract the frequency of occurrence of each time period between the starting offset and the ending offset from the return time feature data of the registered license plate; Attenuation weights are assigned to each time period based on its distance from the starting offset, and the attenuation weights decrease as the distance increases. The total weighted frequency is calculated based on the frequency of occurrence in each time period and the corresponding attenuation weight; The total historical frequency is obtained by summing the occurrence frequencies of all time periods in the return duration feature data; The ratio of the weighted total frequency to the historical total frequency is used as the return tendency of the vacant owner parking spaces.

3. The method according to claim 1, characterized in that, The steps of summarizing the return tendency of all owner parking spaces currently occupied by shared vehicles to obtain a conflict risk value, marking owner parking spaces with a return tendency below a preset safety threshold and not occupied by shared vehicles as hedging candidate parking spaces, and calculating the effective hedging depth based on a preset benchmark value and the return tendency of each hedging candidate parking space specifically include: Take any pair of owner parking spaces from those currently occupied by shared vehicles, and determine the overlap of the return time periods of the pair of owner parking spaces based on the frequency distribution of each owner parking space in the preset forward window during the time period covered. The basic conflict amount is obtained by summing up the return tendency of each owner's parking space that is currently occupied by shared vehicles. The coupling conflict increment is determined according to the overlap of the return time periods of each pair of owner's parking spaces. The basic conflict amount and the coupling conflict increment are added together to obtain the conflict risk value. Traverse all currently vacant parking spaces owned by owners that are not occupied by shared vehicles, and mark the parking spaces owned by owners whose return tendency is lower than the preset safety threshold as hedging candidate parking spaces. The effective hedging depth is obtained by accumulating the difference between the preset benchmark value and the return tendency of each of the hedging candidate parking spaces.

4. The method according to claim 1, characterized in that, The step of generating an admission instruction when the first difference is not less than the preset redundancy constant, and selecting one of the candidate parking spaces from the offsetting pool to establish an offsetting pairing mapping with the allocated shared parking space, specifically includes: In response to the first difference being not less than the preset redundancy constant, the admission instruction is generated; Obtain the spatial proximity between each of the hedging candidate parking spaces and the allocated shared parking spaces; The matching score of each hedging candidate parking space is determined by combining the return field tendency of each hedging candidate parking space with the corresponding spatial proximity. The candidate parking space with the highest matching score is selected as the target parking space for hedging. The target hedging parking space is associated with the allocated shared parking space and written into the hedging pairing mapping table. The target hedging parking space is removed from the hedging candidate parking spaces. The effective hedging depth is updated according to the difference between the preset benchmark value and the return tendency of the target hedging parking space.

5. The method according to claim 1, characterized in that, After the step of sending temporary parking instructions containing the offset parking space number to the owner's associated terminal, the method further includes: Based on the return time characteristic data of the registered license plates corresponding to each of the hedging candidate parking spaces, the frequency ratio of the time period covered by the preset forward window is reread in each continuous time period after the current time to obtain the predicted return tendency of each of the hedging candidate parking spaces in each time period. In each time period, the hedging candidate parking spaces whose predicted return tendency exceeds the preset safety threshold are excluded, and the predicted effective hedging depth for each time period is recalculated based on the predicted return tendency of the remaining hedging candidate parking spaces and the preset benchmark value. The difference between the predicted effective hedging depth for each time period and the conflict risk value is used to determine the time period when the difference first falls below the preset redundancy constant as the hedging exhaustion period. The upper limit of the allowed parking time is determined based on the time interval between the hedging depletion period and the current period, and the upper limit of the allowed parking time is appended to the access instruction.

6. The method according to claim 1, characterized in that, After the step of sending temporary parking instructions containing the offset parking space number to the owner's associated terminal, the method further includes: Based on the return time characteristic data of the registered license plates corresponding to the parking spaces occupied by shared vehicles, the frequency ratio of the time periods covered by the preset look-ahead window is reread in each continuous time period after the current time to obtain the predicted return tendency of each parking space occupied by shared vehicles in each time period. The predicted return tendency of each owner's parking space currently occupied by shared vehicles is summed up to obtain the predicted conflict risk value for each time period. The effective hedging depth is subtracted from the predicted conflict risk value corresponding to each time period, and the time period in which the first difference is not less than the preset redundancy constant is determined as the expected recovery period. Based on the time interval between the expected recovery period and the current period, an expected waiting time is generated, and a queuing instruction containing the expected waiting time is sent to the shared vehicle that triggered the entry request.

7. The method according to claim 1, characterized in that, After the step of comparing the difference between the effective hedging depth and the conflict risk value with a preset redundancy constant to obtain a first difference value, the method further includes: When the first difference is less than the preset redundancy constant, a rejection command is generated and the entrance gate is prevented from opening.

8. A shared parking space access control system, characterized in that, The shared parking space access control system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the shared parking space access control system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the shared parking space access control system, the shared parking space access control system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the shared parking space access control system, the shared parking space access control system performs the method as described in any one of claims 1-7.

Images