A parking recommendation method

Abstract

This application discloses a parking recommendation method, comprising: acquiring all parking record information of a user within a preset historical time period, each parking record information including the location information of the parking lot and the parking time period type corresponding to the parking time; when a user triggers a parking demand signal, matching the parking record information according to the parking time period type to which the trigger time belongs, to obtain several key parking lots; determining the largest circumcircle of the area formed by connecting all key parking lots as the core parking area, and determining all parking lots with available parking spaces in the core parking area as target parking lots; determining the parking recommendation value of the target parking lots based on the vehicle flow rate, number of available parking spaces, unit price, and distance value; selecting a preset number of target parking lots with the highest parking recommendation values ​​to form a recommended parking lot list and pushing it to the user's terminal. Thus, parking lots can be recommended in a personalized manner according to the user's parking time period type, improving parking efficiency.

Description

Technical Field

[0001] This invention relates to the field of parking recommendation technology, and more particularly to a parking recommendation method for parking lots. Background Technology

[0002] With social development, cars have become increasingly common and have gradually entered every household. While cars have brought many conveniences to people's lives, they have also brought a series of problems. Especially in bustling cities, on the one hand, there are many vehicles but few parking spaces. People can only confirm whether there is a parking space by driving to the site. When the parking spaces are full, it causes inconvenience for parking and leads to the phenomenon of random parking, which can easily cause traffic congestion and traffic accidents. Summary of the Invention

[0003] This application provides a parking recommendation method that can personalize parking recommendations based on the user's parking time period type, thereby improving parking efficiency.

[0004] This application provides a parking recommendation method for parking lots, including:

[0005] S101, obtain all parking record information of the user within a preset historical time period. Each parking record information includes the location information of the parking lot and the parking time period type corresponding to the parking time. The parking time period type includes weekday morning peak, weekday evening peak, weekday regular, and rest day regular.

[0006] S102, When a user triggers a parking request signal, the parking record information is traversed and matched according to the parking time period type to which the trigger time belongs, to obtain several key parking lots;

[0007] S103, the largest circumcircle of the area formed by connecting all key parking lots is determined as the core parking area, and all parking lots with vacant spaces within the core parking area are obtained and determined as target parking lots.

[0008] S104. Based on the vehicle turnover rate, number of vacant parking spaces, unit price, and distance value of the target parking lot, determine the recommended parking value for the target parking lot.

[0009] S105: Sort all target parking lot recommendation values ​​in descending order, select the top preset number of target parking lots to form a recommended parking lot list and push it to the user terminal.

[0010] Preferably, the step of matching parking record information based on the parking time period type to which the trigger time belongs to obtains several key parking lots, specifically including:

[0011] A1. Based on the parking time period type to which the trigger time belongs, match parking lots with the same parking time period type in all parking record information corresponding to the user, and record them as the first parking lot;

[0012] A2. The density value of user parking in each first parking lot within a preset historical time period is calculated according to the following formula:

[0013] ,in, Let be the density value of the i-th first parking lot. Let N be the number of times a user parks in the i-th parking lot within a preset historical time period, and let N be the total number of parking lots. This represents the total number of times a user parks in all first parking lots within a preset historical time period;

[0014] A3. Identify all parking lots with density values ​​greater than a preset threshold as key parking lots.

[0015] Preferably, the vehicle flow rate of the target parking lot is calculated according to the following formula: ,in, This represents the vehicle flow rate of the target parking lot within the current time window. This represents the vehicle flow rate of the target parking lot within the previous time window. These represent the number of vehicles entering and exiting the target parking lot within the current time window, respectively, where L is the length of the time window. The smoothing index, ranging from 0 to 1, is used to balance the impact of the current time window and historical time windows on vehicle flow rate.

[0016] Preferably, in step S104, the recommended parking value for the target parking lot is calculated according to the following formula:

[0017] Where S is the recommended parking value for the target parking lot. Let y be the vehicle flow rate of the target parking lot within the current time window, y be the current number of vacant parking spaces in the target parking lot, Y be the total number of parking spaces in the target parking lot, Mmax be the maximum charging unit price in the corresponding parking area, M be the charging unit price of the target parking lot, and D be the distance between the target parking lot and the user's current location. , , θ and θ represent the weighting factors of the impact of vehicle flow rate, number of available parking spaces, distance value, and unit price on the parking recommendation value, respectively.

[0018] Preferably, the method further includes:

[0019] S201, taking each key parking lot as the center, delineates a backup parking area according to a preset radius, with each key parking lot corresponding to one backup parking area;

[0020] S202, all parking lots with available spaces in the alternative parking areas are also identified as target parking lots, and each target parking lot is assigned an area label to distinguish its parking area; wherein, the parking area includes the core parking area and several alternative parking areas;

[0021] S203. Based on the total number of times a user parks in all parking lots in each parking area within a preset historical time period and the vehicle flow rate in each parking area within the current time window, the regional dynamic parking index of each parking area is obtained.

[0022] Preferably, the area dynamic parking index for each parking area is calculated according to the following formula:

[0023] ,in, The vehicle flow rate at the boundary of each parking area is denoted as T0, where T0 is the total number of times a user parks in all parking lots within that parking area during a preset historical time period, and t1 is the length of the preset historical time period. The preset influence weight value.

[0024] Preferably, after step S104, the method further includes:

[0025] The dynamic parking index of the parking area to which the target parking lot belongs is determined based on the area label of the target parking lot. The parking recommendation value of the target parking lot is multiplied by its corresponding dynamic parking index to obtain a new parking recommendation value, which replaces the original parking recommendation value.

[0026] Preferably, after step S105, the method further includes:

[0027] S301 periodically acquires the user's real-time location based on a preset unit time period;

[0028] S302, based on the distance relationship between the user's real-time location and the center point of each parking area, calculates the distance ratio between the user and each parking area, specifically: Where r is the distance ratio between the current user and each parking area, d1 is the actual distance between the current user and the center point of the corresponding parking area, and R is the radius of the corresponding parking area;

[0029] S303, based on the distance ratio between the user and each parking area, obtain the trend value between the user and each parking area;

[0030] S304, execute step S203 to obtain the current regional dynamic parking index of each parking area, and multiply it by its corresponding trend value to update and replace the original regional dynamic parking index.

[0031] S305, execute steps S104 to S105, update the recommended parking lot list every unit time period, periodically update the recommended parking lot list pushed to the user until the user is detected to have entered a target parking lot, then stop pushing.

[0032] Preferably, the trend value between the user and each parking area is obtained based on the distance ratio between the user and each parking area, specifically: TV = 1 / (1 + r), where TV is the trend value between the user and each parking area, and r is the distance ratio between the current user and each parking area.

[0033] Preferably, in S303, the following is further included:

[0034] C1. The following formula is used to obtain the trend value change index between users and various parking areas:

[0035] ,in, The index represents the trend value change of the user between the current location and various parking areas, where j is the sequence number of the corresponding unit time period.

[0036] C2. Sum the trend values ​​between the user and each parking area with their corresponding change indices, and replace the original trend values.

[0037] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0038] By leveraging the parking time period type corresponding to the user's triggered parking signal, core parking areas are matched against historical parking records to achieve personalized target parking lot search, improving parking efficiency and success rate, significantly reducing user search time, and enhancing user experience. By acquiring the vehicle flow rate, number of available parking spaces, unit price, and distance from the user's current location of the target parking lot, and combining these parameters to calculate parking recommendation values, a comprehensive and objective parking lot recommendation list can be provided to the user. This list is not only based on the user's personal historical parking habits but also considers real-time parking lot conditions, helping users quickly make the best choice among many parking lots, satisfying personalized needs while ensuring parking efficiency and cost-effectiveness.

[0039] It is not limited to the core parking area determined by time period type and parking record, but also determines the alternative parking area based on each key parking lot, thus forming multiple different parking areas and expanding the range of recommended parking for users. The key parking lots determined based on historical parking records improve the diversity and selectivity of parking lots. By introducing the regional dynamic parking index of the parking area to which the target parking lot belongs, the parking recommendation value of the target parking lot is adjusted, which not only considers the characteristics of the parking lot itself, but also combines the dynamic situation of the parking area.

[0040] By periodically monitoring the user's real-time location, the system can reflect the dynamic relationship between the user and various parking areas in real time. By introducing distance ratios and trend values, the system can dynamically adjust the regional dynamic parking index of parking areas, quantify the dynamic relationship between the user and parking areas, and flexibly adjust the recommended parking lot list strategy according to the user's actual driving situation. The dynamically updated parking lot recommendation list can more accurately meet the user's actual needs during the driving process and improve the user experience. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the parking recommendation method in an embodiment of the present invention. Detailed Implementation

[0042] To facilitate understanding of the present invention, a more complete description of this application will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the present invention.

[0043] It should be noted that the terms "vertical," "horizontal," "up," "down," "left," "right," and similar expressions used in this article are for illustrative purposes only and do not represent the only possible implementation.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0045] Example 1: Figure 1 This is a flowchart illustrating a parking recommendation method for a parking lot according to an embodiment of the present invention.

[0046] like Figure 1 As shown, a parking recommendation method for a parking lot includes the following steps:

[0047] S101, obtain all parking record information of the user within a preset historical time period. Each parking record information includes the location information of the parking lot and the parking time period type corresponding to the parking time.

[0048] The parking time period types are set based on expert experience or users' personal time patterns, specifically including weekday morning peak, weekday evening peak, weekday regular, and weekend regular. The preset historical time period is set to the past year, which can be adjusted according to actual conditions.

[0049] Specifically, the user's client stores all parking records for that user within a preset historical time period. Through big data analysis, the user's personal time patterns can be analyzed. For example, if user A's work hours on weekdays are from 7:00 to 9:00, then the parking records during this period correspond to user A's weekday morning rush hour; if user A's get off work hours on weekdays are from 17:00 to 19:00, then the parking records during this period correspond to user A's weekday evening rush hour; and the parking records for the remaining time periods on weekdays correspond to user A's regular weekday schedule.

[0050] S102, when a user triggers a parking request signal, the parking record information is traversed and matched according to the parking time period type to which the trigger time belongs, to obtain several key parking lots.

[0051] Specifically, based on the parking time period type to which the trigger time belongs, parking record information is traversed and matched to obtain several key parking lots, including:

[0052] A1. Based on the parking time period type to which the trigger time belongs, match parking lots with the same parking time period type among all the user's corresponding parking record information, and record them as the first parking lot.

[0053] A2. The density value of user parking in each first parking lot within a preset historical time period is calculated according to the following formula:

[0054] ,in, Let be the density value of the i-th first parking lot. Let N be the number of times a user parks in the i-th parking lot within a preset historical time period, and let N be the total number of parking lots. This represents the total number of times a user parks in all first parking lots within a preset historical time period.

[0055] A3. Identify all parking lots with density values ​​greater than a preset threshold as key parking lots.

[0056] In other embodiments of the present invention, the parking record information also includes attribute tag information for the parking lot. This attribute tag information is set based on an electronic map. Specifically, all key environmental information within a 3km radius of the parking lot is obtained from the electronic map, and attribute tag information is set according to the business type to which the key environmental information belongs. For example, if the key environmental information includes "XX Building" and "XX Hospital," then the corresponding business types are office building and hospital, respectively. The business type to which the key environmental information belongs is defined according to actual circumstances, and the present invention does not impose any limitations on this.

[0057] Specifically, the user terminal is also equipped with a voice recognition function. Step S102 further includes: when user voice information is recognized, the attribute tag information in the parking record information is traversed and text matching is performed to obtain several key parking lots. For example, when "hospital" is recognized, text matching is performed based on all attribute tag information in the parking record information. A large language model can be used for recognition. All parking lots corresponding to attribute tag information that are semantically similar to or the same as "hospital" are recorded as the first parking lot. Then, the contents of A2 to A3 are executed to obtain the key parking lots.

[0058] S103, the largest circumcircle of the area formed by connecting all key parking lots is determined as the core parking area, and all parking lots with vacant spaces within the core parking area are obtained and determined as target parking lots.

[0059] S104. Based on the vehicle turnover rate, number of vacant parking spaces, unit price, and distance value of the target parking lot, determine the recommended parking value for the target parking lot.

[0060] Specifically, step S104 includes:

[0061] B1. The vehicle flow rate of the target parking lot is calculated using the following formula:

[0062] ,in, This represents the vehicle flow rate of the target parking lot within the current time window. This represents the vehicle flow rate of the target parking lot within the previous time window. These represent the number of vehicles entering and exiting the target parking lot within the current time window, respectively, and L is the length of the time window, for example, it can be set to 30 minutes. The smoothing index, ranging from 0 to 1, is used to balance the impact of the current time window and historical time windows on vehicle flow rate.

[0063] The time window is set to 30 minutes, and the current time window is set to the 30 minutes closest to the current time. It should be noted that for the first time window (t = 1), since there is no vehicle flow rate R(t-1) of the previous time window, the vehicle flow rate of yesterday corresponding to the first time window can be determined as R(t-1).

[0064] By incorporating vehicle flow rate data from the previous time window, we can smooth out data fluctuations in the current time window. Traffic flow often exhibits certain volatility, especially with significant changes occurring in short periods. By introducing data from the previous time window, we can reduce the impact of such short-term fluctuations on the calculation results. Considering data from the previous time window can provide more contextual information, thereby improving the accuracy of predictions and making the estimation of vehicle flow rate more stable and reliable.

[0065] A high vehicle turnover rate usually means that a large number of vehicles enter and exit the target parking lot in a short period of time. This may indicate that the target parking lot is very busy. For users looking for parking spaces, a parking lot with a high vehicle turnover rate may mean that it is more difficult to find an empty parking space, or that they need to move their vehicles more frequently to avoid blocking other vehicles, thus affecting the convenience and efficiency of parking.

[0066] B2. Obtain the total number of parking spaces and the number of currently available parking spaces in the target parking lot, the hourly rate, and the distance from the user's current location. Combined with the vehicle turnover rate of the target parking lot, calculate the recommended parking value for the target parking lot according to the following formula:

[0067] Where S is the recommended parking value for the target parking lot. Let y be the vehicle flow rate of the target parking lot within the current time window, y be the current number of vacant parking spaces in the target parking lot, Y be the total number of parking spaces in the target parking lot, Mmax be the maximum charging unit price in the corresponding parking area, M be the charging unit price of the target parking lot, and D be the distance between the target parking lot and the user's current location. , , θ and θ are the weighting factors of vehicle flow rate, number of available parking spaces, distance value, and unit price on the parking recommendation value, respectively. They are set according to actual conditions and expert experience, and this invention does not limit or elaborate on them.

[0068] It should be noted that, Specifically, by dividing the vehicle flow rate by the number of available parking spaces plus 1 (to avoid dividing by zero), we are essentially considering the vehicle flow rate relative to the number of available parking spaces. The fewer the available parking spaces, the greater the impact of the vehicle flow rate, which is reflected in the decrease of the exponential term, thereby lowering the recommended parking value. This links the vehicle flow rate of the target parking lot to the number of available parking spaces. It's understandable that... This can be used as an adjustment The dynamic parameter factor of size makes the parking recommendation value more in line with the actual situation.

[0069] S105: Sort all target parking lot recommendation values ​​in descending order, select the top preset number of target parking lots to form a recommended parking lot list and push it to the user terminal.

[0070] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages:

[0071] By matching the core parking area in historical parking records based on the parking time period type when the user triggers the parking signal, the system can identify core parking areas that match the current time period type by analyzing the user's personalized historical parking habits. This enables personalized parking spot search, improving parking efficiency and success rate, greatly reducing the time users spend searching for parking spaces, and enhancing user experience.

[0072] By acquiring data on vehicle traffic flow, number of available parking spaces, unit price, and distance from the user's current location at the target parking lot, and combining these parameters to calculate a parking recommendation value, the system comprehensively considers parking convenience (distance), cost (unit price), availability (number of available parking spaces), and parking lot busyness (vehicle traffic flow). This provides users with a comprehensive and objective list of recommended parking lots. This list is based not only on the user's personal historical parking habits but also on real-time parking conditions, helping users quickly make the best choice among numerous parking lots. It satisfies personalized needs while ensuring parking efficiency and cost-effectiveness.

[0073] Example 2: Example 1 relied solely on core parking areas comprised of key parking lots identified through historical parking records to recommend parking lots. However, these core parking areas have limitations in their selection scope. Overly limited areas result in fewer parking lot choices. If too few target parking lots meet the criteria, calculating and ranking parking recommendation values ​​becomes largely meaningless, leading to resource waste and impacting parking recommendation efficiency. With the continuous changes in urban traffic conditions, a single parking recommendation method can no longer meet the parking needs of modern cities.

[0074] Therefore, the embodiments of this application are optimized based on the above embodiments.

[0075] In some embodiments, prior to step S104, the following steps are also included:

[0076] S201: Centered on each key parking lot, a backup parking area is delineated according to a preset radius, with each key parking lot corresponding to one backup parking area.

[0077] Among them, the alternative parking areas need to be deduplicated from the core parking areas. The deduplicated alternative parking areas are determined as new alternative parking areas to replace the original alternative parking areas, so as to ensure the independence of each alternative parking area and avoid the overlap between alternative parking areas and core parking areas.

[0078] S202, all parking lots with available spaces in the alternative parking areas are also identified as target parking lots, and each target parking lot is assigned an area label to distinguish its parking area (the core parking area and several alternative parking areas are collectively referred to as parking areas).

[0079] S203. Based on the total number of times a user parks in all parking lots in each parking area within a preset historical time period and the vehicle flow rate in each parking area within the current time window, the regional dynamic parking index of each parking area is obtained.

[0080] Specifically, the regional dynamic parking index for each parking area is calculated using the following formula: ,in, This represents the vehicle flow rate at the boundary of each parking area within the current time window, including both entry and exit. The specific calculation method can be found in step B1, simply replacing the target parking lot with the parking area. This embodiment will not elaborate further. T0 represents the total number of times the user parks in all parking lots within this parking area during the preset historical time period, and t1 represents the length of the preset historical time period. The preset influence weight value is used to adjust the degree of influence of vehicle flow rate on the regional dynamic parking index.

[0081] Therefore, the lower the vehicle flow rate in a parking area and the more times a vehicle stops, the higher the area's dynamic parking index.

[0082] In some embodiments, step S104 is followed by:

[0083] The dynamic parking index of the parking area to which the target parking lot belongs is determined based on the area label of the target parking lot. The parking recommendation value of the target parking lot is multiplied by its corresponding dynamic parking index to obtain a new parking recommendation value, which replaces the original parking recommendation value.

[0084] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages:

[0085] No longer limited to core parking areas determined by time period type and parking records, alternative parking areas are also determined based on each key parking lot, thus forming multiple different parking areas, expanding the range of recommended parking for users, and improving the diversity and selectivity of parking lots by determining key parking lots based on historical parking records;

[0086] By introducing the regional dynamic parking index of the parking area to which the target parking lot belongs, and taking into account the vehicle busyness (parking convenience) between different parking areas and the user's preference for the parking area, the parking recommendation value of the target parking lot is calculated by not only considering the characteristics of the parking lot itself, but also the dynamic situation of the parking area, so as to adapt to the ever-changing urban road environment and further improve the accuracy, dynamism and comprehensiveness of parking recommendations.

[0087] Example 3: The aforementioned examples aim to recommend a parking list once when a user triggers a parking demand signal. However, they do not take into account the unpredictable changes in the user's driving route due to the variability and complexity of actual traffic. When a user deviates from certain designated parking areas, the calculation of parking recommendation values ​​for all target parking lots in these areas may result in the final recommended parking list failing to adapt to the dynamic relationship between the user and the parking areas. The original recommendation strategy only makes a recommendation once when a user triggers a parking demand signal, which cannot adapt to the real-time changes in the relationship between the user and various parking areas during the driving process.

[0088] Therefore, the embodiments of this application are optimized based on the above embodiments.

[0089] In some embodiments, after step S105, i.e. after successfully pushing the recommended parking lot list when the user triggers a parking demand signal, the method further includes:

[0090] S301 periodically acquires the user's real-time location based on a preset unit time period, wherein the preset unit time period is set to 5 minutes, that is, the user's real-time location is acquired every 5 minutes.

[0091] S302, based on the distance relationship between the user's real-time location and the center point of each parking area, obtain the distance ratio between the user and each parking area, specifically: r = Where r is the distance ratio between the current user and each parking area, d1 is the actual distance between the current user and the center point of the corresponding parking area, and R is the radius of the corresponding parking area.

[0092] It should be noted that the center point of each parking area is the center of the corresponding circle.

[0093] S303, based on the distance ratio between the user and each parking area, obtain the trend value between the user and each parking area, specifically: TV = 1 / (1 + r), where TV is the trend value between the user and each parking area, and r is the distance ratio between the user and each parking area, so that the closer the user is to or near a certain parking area, the greater the trend value of that parking area.

[0094] S304, execute step S203 to obtain the current regional dynamic parking index of each parking area, and multiply it by its corresponding trend value to update and replace the original regional dynamic parking index.

[0095] The process involves multiplying the real-time trend value by the periodically calculated regional dynamic parking index to dynamically adjust the original index. Based on this adjusted index, a parking recommendation algorithm is periodically executed to obtain and update the recommended parking list for each time period. This adjustment method makes parking recommendations more flexible, adapting to real-time changes during user travel and ensuring the timeliness and accuracy of the recommended list.

[0096] S305, execute steps S104 to S105, update the recommended parking lot list every unit time period, thereby periodically updating the recommended parking lot list pushed to the user until the user is detected to have entered a target parking lot, and then stop pushing.

[0097] To better understand the parking recommendation method of this invention, as an example, the method flow specifically includes:

[0098] S401. User Xiao Zhang triggered a parking demand signal at 8:00 AM on Monday, which is during the weekday morning rush hour. Parking records for the past year were retrieved from Xiao Zhang's client. Based on the weekday morning rush hour, the key parking lots were identified as Parking Lot A, Parking Lot B, and Parking Lot C. The largest circumcircle of the area connecting the three parking lots was taken as the core parking area. A reserve parking area was delineated with each key parking lot as the center, resulting in three reserve parking areas. Finally, four parking areas were determined.

[0099] S402: Obtain all parking lots with available parking spaces in the four parking areas as target parking lots, obtain the regional dynamic parking index of the parking area to which the target parking lot belongs, multiply the parking recommendation value of the target parking lot by its corresponding regional dynamic parking index to obtain the parking recommendation value of the target parking lot, sort all target parking lots in descending order, select the top 3 parking lots to form a recommended parking lot list and push it to Xiao Zhang's user terminal.

[0100] S403 acquires Zhang's real-time location, the regional dynamic parking index of each parking area, and the parking recommendation value of each target parking lot every 5 minutes. It also obtains the real-time distance ratio and trend value between Zhang and each parking area, multiplies the trend value by the regional dynamic parking index of the corresponding parking area, updates the regional dynamic parking index of each parking area, and updates the parking recommendation value of each target parking lot. It then regenerates the recommended parking lot list and pushes it to the user's terminal. Zhang receives the recommended parking lot list periodically during the driving process and finally decides to drive into parking lot D, at which point the periodic push stops.

[0101] Therefore, this parking recommendation method can combine Xiao Zhang's parking habits and the dynamic changes in the relationship between him and various parking areas during his journey to efficiently and accurately match the recommended parking lot list, saving Xiao Zhang time in finding parking spaces and improving commuting efficiency.

[0102] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages:

[0103] By periodically monitoring the user's real-time location, the system can reflect the dynamic relationship between the user and various parking areas in real time. By introducing distance ratios and trend values, the system can dynamically adjust the regional dynamic parking index of parking areas, quantify the dynamic relationship between the user and parking areas, and flexibly adjust the recommended parking lot list strategy according to the user's actual driving situation. The dynamically updated parking lot recommendation list can more accurately meet the user's actual needs during the driving process and improve the user experience.

[0104] Example 4: In Example 3, the trend value between the user and the corresponding parking area is determined based on the distance ratio between the parking area and the user in a single current time period. This has certain limitations and cannot adapt to the dynamic changes in the user's driving trend.

[0105] Therefore, the embodiments of this application are optimized based on the above embodiments.

[0106] In some embodiments, step S303 further includes:

[0107] C1. The following formula is used to obtain the trend value change index between users and various parking areas:

[0108] ,in, Let j be the index representing the trend change of the user's current location relative to various parking areas, and j be the sequence number of the current time unit. For example, if the current time unit is the 3rd time unit, i.e., the 3rd real-time acquisition of the user's location, then... This represents the index of the trend value change between the user and various parking areas, where 3 is the sequence number of the corresponding unit time period.

[0109] The formula takes into account the trend changes of the user and each parking area in the current unit time period j. This change index is used to quantify the change in the user's proximity to each parking area in a continuous unit time period.

[0110] C2. Sum the trend values ​​between the user and each parking area with their corresponding change indices, and replace the original trend values.

[0111] For example, if at the current moment, based on the distance ratio between the user and a certain pre-parking area, a trend value 'a' is obtained between the user and the pre-parking area, and the exponent of the change of the trend value 'a' between the user and the pre-parking area is 'b', then the value of 'a+b' replaces the original trend value 'a'.

[0112] Specifically, the trend value is adjusted using a change index. This index quantifies the changes in distance ratios across multiple time units. A higher change index indicates a greater likelihood that the user is currently approaching the corresponding parking area, resulting in a higher trend value. Therefore, the trend value is not only based on static distance ratios but also incorporates dynamic trends, accurately capturing the user's real-time approach intentions. This improves the differentiation between parking areas, thereby enhancing the accuracy and real-time nature of recommendations.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A parking recommendation method for parking lots, characterized in that, include: S101, obtain all parking record information of the user within a preset historical time period. Each parking record information includes the location information of the parking lot and the parking time period type corresponding to the parking time. The parking time period type includes weekday morning peak, weekday evening peak, weekday regular, and rest day regular. S102, When a user triggers a parking request signal, the parking record information is traversed and matched according to the parking time period type to which the trigger time belongs, to obtain several key parking lots; S103, the largest circumcircle of the area formed by connecting all key parking lots is determined as the core parking area, and all parking lots with vacant spaces within the core parking area are obtained and determined as target parking lots. S104, Based on the vehicle turnover rate, number of vacant parking spaces, unit price, and distance value of the target parking lot, determine the recommended parking value for the target parking lot; the vehicle turnover rate of the target parking lot is calculated according to the following formula: , This represents the vehicle flow rate of the target parking lot within the current time window. This represents the vehicle flow rate of the target parking lot within the previous time window. These represent the number of vehicles entering and exiting the target parking lot within the current time window, respectively, where L is the length of the time window. The smoothing index, ranging from 0 to 1, is used to balance the impact of the current time window and historical time windows on vehicle flow rate; the recommended parking value for the target parking lot is calculated using the following formula: S represents the recommended parking value for the target parking lot. Let y be the vehicle flow rate of the target parking lot within the current time window, y be the current number of vacant parking spaces in the target parking lot, Y be the total number of parking spaces in the target parking lot, Mmax be the maximum charging unit price in the corresponding parking area, M be the charging unit price of the target parking lot, and D be the distance between the target parking lot and the user's current location. , , , These are the weighting factors for the impact of vehicle turnover rate, number of available parking spaces, distance value, and unit price on parking recommendation value; S105: Sort all target parking lot recommendation values ​​in descending order, select the top preset number of target parking lots to form a recommended parking lot list and push it to the user terminal.

2. The parking recommendation method for parking lots as described in claim 1, characterized in that, The process involves iterating through parking record information and matching it based on the parking time period type to which the trigger time belongs, resulting in several key parking lots, specifically including: A1. Based on the parking time period type to which the trigger time belongs, match parking lots with the same parking time period type in all parking record information corresponding to the user, and record them as the first parking lot; A2. The density value of user parking in each first parking lot within a preset historical time period is calculated according to the following formula: ,in, Let be the density value of the i-th first parking lot. Let N be the number of times a user parks in the i-th parking lot within a preset historical time period, and let N be the total number of parking lots. This represents the total number of times a user parks in all first parking lots within a preset historical time period; A3. Identify all parking lots with density values ​​greater than a preset threshold as key parking lots.

3. The parking recommendation method for parking lots as described in claim 1, characterized in that, Prior to S104, the method further includes: S201, taking each key parking lot as the center, delineates a backup parking area according to a preset radius, with each key parking lot corresponding to one backup parking area; S202, all parking lots with available spaces in the alternative parking areas are also identified as target parking lots, and each target parking lot is assigned an area label to distinguish its parking area; wherein, the parking area includes the core parking area and several alternative parking areas; S203. Based on the total number of times a user parks in all parking lots in each parking area within a preset historical time period and the vehicle flow rate in each parking area within the current time window, the regional dynamic parking index of each parking area is obtained.

4. The parking recommendation method for parking lots as described in claim 3, characterized in that, The regional dynamic parking index for each parking area is calculated using the following formula: ,in, The vehicle flow rate at the boundary of each parking area is denoted as T0, where T0 is the total number of times a user parks in all parking lots within that parking area during a preset historical time period, and t1 is the length of the preset historical time period. The preset influence weight value.

5. The parking recommendation method for parking lots as described in claim 4, characterized in that, Following S104, the method further includes: The dynamic parking index of the parking area to which the target parking lot belongs is determined based on the area label of the target parking lot. The parking recommendation value of the target parking lot is multiplied by its corresponding dynamic parking index to obtain a new parking recommendation value, which replaces the original parking recommendation value.

6. The parking recommendation method for parking lots as described in claim 3, characterized in that, Following S105, the method further includes: S301 periodically acquires the user's real-time location based on a preset unit time period; S302, based on the distance relationship between the user's real-time location and the center point of each parking area, calculates the distance ratio between the user and each parking area, specifically: Where r is the distance ratio between the current user and each parking area, d1 is the actual distance between the current user and the center point of the corresponding parking area, and R is the radius of the corresponding parking area; S303, based on the distance ratio between the user and each parking area, obtain the trend value between the user and each parking area; S304, execute step S203 to obtain the current regional dynamic parking index of each parking area, and multiply it by its corresponding trend value to update and replace the original regional dynamic parking index. S305, execute steps S104 to S105, update the recommended parking lot list every unit time period, periodically update the recommended parking lot list pushed to the user until the user is detected to have entered a target parking lot, then stop pushing.

7. The parking recommendation method for parking lots as described in claim 6, characterized in that, The trend value between the user and each parking area is obtained based on the distance ratio between the user and each parking area, specifically: TV = 1 / (1 + r), where TV is the trend value between the user and each parking area, and r is the distance ratio between the current user and each parking area.

8. The parking recommendation method for parking lots as described in claim 7, characterized in that, S303 also includes: C1. The following formula is used to obtain the trend value change index between users and various parking areas: ,in, The index represents the trend value change of the user between the current location and various parking areas, where j is the sequence number of the corresponding unit time period. C2. Sum the trend values ​​between the user and each parking area with their corresponding change indices, and replace the original trend values.

Images