A vehicle management system for intelligent parking

Abstract

The application discloses a vehicle management system for intelligent parking, comprising: a background monitoring device, which is used for acquiring basic information of each vehicle for entering a specified parking lot of a park in real time; a control device, which is used for determining whether the vehicle belongs to a reserved vehicle according to the basic information of the vehicle, and if yes, determining whether the area to which the current vehicle type belongs is idle, and if yes, sending a guiding instruction to a vehicle terminal. In addition, the management system analyzes historical data of parking space utilization of the parking lot in each period, thereby predicting the available situation of the parking space when a user who is ready to park arrives at the parking lot, and reasonably guiding the user to select a parking position according to the prediction result, and preferentially recommending a parking lot with higher stability, so that the accuracy of the recommendation can be improved, and the situation that a reserved vehicle is difficult to park when arriving at the parking lot due to a sudden change in the number of parked vehicles in the parking lot can be avoided.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent transportation, and in particular, relates to a vehicle management system for intelligent parking. BACKGROUND

[0002] With the development of economy, cars have become a common means of transportation for people, but the rapid growth of the number of cars has also led to many problems, such as the parking problem of cars, especially in areas with high population density in cities, parking difficulty has become a widespread problem.

[0003] In the prior art, when a car owner is looking for a parking space, he mainly relies on visual observation and cannot intuitively obtain the distribution of available parking spaces, so the efficiency is low, and it is easy to take a long time to find an available parking space, which affects the commuting efficiency, is not conducive to smooth traffic in the parking lot, and in addition, when the car owner has a specific destination, he cannot know whether the parking lot at the destination has a vacant parking space, and can only judge by experience, which is easy to cause the parking lot to have no space or few spaces when arriving, affecting the parking efficiency of the car owner, in order to accurately predict whether the parking lot is suitable for parking, the present application provides a vehicle management system for intelligent parking to solve the above problems. SUMMARY

[0004] The purpose of the present application is to provide a vehicle management system for intelligent parking, which solves the problem that in the prior art, when a vehicle is ready to park in a parking lot, it cannot be determined whether there is enough space for the parking lot to park when the vehicle arrives at the parking lot, and when there are multiple optional parking lots for parking, the parking resources cannot be reasonably utilized.

[0005] The purpose of the present application can be achieved by the following technical solutions:

[0006] In one possible implementation, the present application provides a vehicle management system for intelligent parking, comprising:

[0007] A background monitoring device is used to obtain basic information of each vehicle entering a specified park parking lot monitored by each monitoring unit in real time, the basic information including: vehicle identification, time information of vehicle entering the park, vehicle type;

[0008] A control device is used to determine whether the vehicle belongs to a reservation vehicle according to the basic information of the vehicle, if yes, to determine whether the area to which the current vehicle type belongs is idle, if yes, to send a guidance instruction to the vehicle terminal.

[0009] Optionally, if there are multiple public parking lots and public driving areas in the specified park, and the public parking lots meet the pre-reservation mechanism, the method further comprises:

[0010] The control device is also used for

[0011] receiving a reservation request of any first vehicle, the reservation request of the first vehicle comprising selected idle parking position information and current vehicle position information;

[0012] obtaining a predicted time t2 when the current first vehicle reaches the selected parking position according to the reservation request;

[0013] determining a reliability coefficient of the selected parking position in the reservation request according to the predicted time t2, parking strategy information established based on historical data, and sending response information of whether to agree to the first vehicle according to the reliability coefficient.

[0014] Optionally, determining the reliability coefficient of the selected parking position in the reservation request according to the predicted time t2, the parking strategy information established based on the historical data, and sending the response information of whether to agree to the first vehicle according to the reliability coefficient comprises: dividing a day into r time periods; obtaining average predicted parking numbers kj of the parking lot to which the selected parking position belongs in each time period from the parking strategy information, where 1≤j≤r, and r is a preset value.

[0015] marking a time period in which the predicted time t2 is located as a target time period; obtaining an average predicted parking number kj corresponding to the target time period from the parking strategy information 目标 , obtaining a maximum parking number kmax of the parking lot in the target time period, obtaining an actual average parking number kf of the parking lot in a time period before a time period corresponding to the current time, and obtaining an average predicted parking number kjf of the parking lot in the time period before the time period corresponding to the current time.

[0016] calculating a reliability coefficient Q of the parking lot corresponding to the current reservation information according to a formula Q=(kj 目标 / kmax)*(kf / kjf).

[0017] When the reliability coefficient Q>Qy 指定数 , the response information comprises information of not suggesting the first vehicle to park in the parking lot; and Qy 指定数 is a preset coefficient.

[0018] Optionally, obtaining the average predicted parking number kj of the parking lot to which the selected parking position belongs in each time period from the parking strategy information based on the parking strategy information established based on the historical data comprises:

[0019] in a time period within a day, obtaining actual parking numbers ksi of the parking lot at intervals of a preset time length tyl, where 1≤i≤n, and n represents a number of sampling times in the corresponding time period;

[0020] calculating a typical value ks of the n ksi;

[0021] Then, obtain the m typical values ​​ks corresponding to this time period over the past m days;

[0022] Continue to calculate the typical value kss of these m typical values ​​ks, and use the typical value kss as the average predicted number of parking spaces kj in the corresponding time period;

[0023] The parking strategy information records the average predicted number of parking spaces kj and the maximum number of parking spaces kma for each time period within a day on non-holidays, obtained through historical data, and the average predicted number of parking spaces kj and the maximum number of parking spaces kma for each time period within a day on holidays.

[0024] Optionally, the control device is further configured to:

[0025] The system monitors vehicle information that sends guidance commands in real time, extracts vehicle operation behavior information from the vehicle information, compares it with preset vehicle operation behavior information, and issues an alarm if there is any inconsistency.

[0026] Vehicle types include: concrete trucks, transport trucks, buses, private cars, and police vehicles;

[0027] The guidance instructions include sending driving information at each node of the guidance route.

[0028] Optionally, the designated park includes a portion of the electronic fence area and a portion of the non-electronic fence area, wherein the non-electronic fence area is the area of ​​public driving area and multiple public parking lots;

[0029] The control device interacts with the vehicle's terminal, which includes a positioning unit, a navigation unit, and a processing unit.

[0030] Optionally, parking strategy information based on historical data includes:

[0031] A reinforcement learning model is used to construct the actual number of parked cars in any time period before the current target time period, and the predicted number of parked cars in any time period within the next three days.

[0032] Optionally, if the response information includes information that it is not recommended to park in the designated parking lot, the vehicle terminal obtains available parking lots around the designated park; at the same time, it obtains parking information for each parking lot for each time period from the external network, and displays the distance between the parking lot to which the selected parking location belongs and the surrounding available parking lots.

[0033] Furthermore, when calculating the reliability coefficient Q, parking spaces reserved for specific individuals and vehicles with such spaces are not included in the calculation.

[0034] Furthermore, when obtaining the average number of parking spaces k for each time period, the standard deviation F of n ksi values ​​is also calculated, thereby obtaining the m standard deviations F obtained in the past m days;

[0035] Calculate the typical value Fd for these m standard deviations;

[0036] The typical value Fd for each time period was calculated using the method described above.

[0037] After logging into the platform via a terminal device, the user selects a destination, and the navigation unit obtains the available parking lots around the destination.

[0038] Obtain the typical value Fd corresponding to each parking lot that satisfies the reliability coefficient Q≤Qy;

[0039] The optional coefficient W for each parking lot is calculated according to the formula W=α1*Fd+α2*X; where α1 and α2 are preset coefficient values;

[0040] Parking lots are recommended in ascending order of their selectability coefficient W.

[0041] The beneficial effects of this invention are:

[0042] 1. This invention analyzes historical data on parking space utilization in parking lots at different times to predict the availability of parking spaces when users arrive at the parking lot. Based on the prediction results, it provides reasonable guidance to users on choosing a parking location, avoiding situations where there are few or no parking spaces available when users arrive at the parking lot, thus preventing parking difficulties. It also makes reasonable use of available parking resources in the surrounding area and considers the impact of temporary factors on the prediction results, thereby improving the accuracy of the prediction.

[0043] 2. By introducing stability evaluation parameters for the number of parking spaces in each parking lot during different time periods, this invention prioritizes parking lots with higher stability when recommending parking lots, which can improve the accuracy of recommendations and avoid situations where reserved vehicles have difficulty parking when they arrive at the parking lot due to sudden changes in the number of parking spaces. Attached Figure Description

[0044] The invention will now be further described with reference to the accompanying drawings.

[0045] Figure 1 This is a schematic diagram of the framework structure of a vehicle management system for smart parking according to the present invention.

[0046] Figure 2 This is a flowchart illustrating the process by which the control center determines whether a parking space meets the parking requirements when a user's vehicle arrives at the parking lot. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Example 1

[0049] like Figure 1 As shown, this embodiment provides a vehicle management system for smart parking, which includes:

[0050] The background monitoring device is used to obtain in real time the basic information of each vehicle entering the designated park parking lot monitored by each monitoring unit. The basic information includes: vehicle identification, vehicle entry time information, and vehicle type.

[0051] The control device is used to determine whether a vehicle is a reserved vehicle based on its basic information. If so, it determines whether the area to which the current vehicle type belongs is available. If so, it sends a guidance command to the vehicle terminal.

[0052] For example, the control device is also used to: monitor vehicle information that sends guidance instructions in real time, extract vehicle operation behavior information from the vehicle information, compare it with preset vehicle operation behavior information, and issue an alarm message if there is any inconsistency.

[0053] Vehicle types include: concrete trucks, transport trucks, buses, private cars, and police vehicles;

[0054] The guidance instructions include sending driving information at each node of the guidance route.

[0055] Therefore, the vehicle management system in this embodiment can realize real-time monitoring of reserved vehicles, and can realize real-time comparison of vehicle operation behavior information for reserved vehicles entering the electronic fence area, ensuring security, and achieving high-level monitoring of the electronic fence area.

[0056] In a specific implementation, the designated park includes a part of the electronic fence area and a part of the non-electronic fence area. The non-electronic fence area is the area of ​​public driving area and multiple public parking lots.

[0057] The control device interacts with the vehicle's terminal, which includes a positioning unit, a navigation unit, and a processing unit.

[0058] In this embodiment, the control device can communicate in real time with the positioning unit, navigation unit, and processing unit in the vehicle terminal to ensure the real-time performance and accuracy of the interaction.

[0059] In the specific implementation process, if a designated park has multiple public parking lots and public driving areas, and the public parking lots meet the pre-reservation mechanism, the method further includes:

[0060] The control device is also used for,

[0061] Step 1: Receive a reservation request from any vehicle. The reservation request from the first vehicle includes: selected available parking location information and the current location information of the vehicle.

[0062] Step 2: Obtain the predicted time t2 when the first vehicle arrives at the selected parking location based on the reservation request;

[0063] Step 3: Based on the predicted time t2, the parking strategy information established from pre-historical data, and the reliability coefficient of the selected parking location in the reservation request, send a response message indicating whether the reservation is approved to the first vehicle according to the reliability coefficient.

[0064] Regarding step three, this embodiment can be further explained as follows:

[0065] A01. Divide a day into r time periods; obtain the average predicted number of parking spaces kj in each time period from the parking strategy information of the parking lot to which the selected parking location belongs, where 1≤j≤r, and r is a preset value;

[0066] A02. Mark the time period where the predicted time t2 is located as the target time period; obtain the average predicted number of parking spaces kj corresponding to the target time period from the parking strategy information. 目标 ; and obtain the maximum number of parking spaces in the parking lot during the target time period kma; obtain the actual average number of parking spaces in the parking lot during the time period before the current time corresponding to the current time period kf; obtain the average predicted number of parking spaces in the parking lot during the time period before the current time corresponding to the current time period kjf;

[0067] A03, According to the formula Q=(kj 目标 The reliability coefficient Q of the reservation information for the current parking lot is calculated using (kmax)*(kf / kjf).

[0068] A04, When the reliability coefficient Q > Qy 指定数 When this happens, the response information includes a recommendation not to park in that parking lot; where Qy 指定数 These are preset coefficients.

[0069] In this embodiment, parking strategy information based on historical data is used to obtain the average predicted number of parking spaces (kj) of the parking lot to which the selected parking location belongs in each time period, including:

[0070] Within a given time period of a day, the actual number of cars parked in the parking lot, ksi, is obtained sequentially at preset time intervals ty1, where 1≤i≤n, and n represents the number of samplings within the corresponding time period. The typical values ​​ks of these n ksi are calculated. Then, the m typical values ​​ks corresponding to this time period are obtained over the past m days. The typical value kss of these m typical values ​​ks is calculated, and this typical value kss is used as the average predicted number of cars parked in the corresponding time period, kj.

[0071] The parking strategy information records the average predicted number of parking spaces (kj) and the maximum number of parking spaces (kmax) for each time period within a day on non-holidays, obtained from historical data, and the average predicted number of parking spaces (kj) and the maximum number of parking spaces (kmax) for each time period within a day on holidays.

[0072] In this embodiment, the parking strategy information established based on historical data can be constructed using a reinforcement learning model to determine the actual number of cars parked in any time period before the current target time period, as well as the predicted number of cars parked in any time period within the next three days.

[0073] Of course, the aforementioned response information includes information that it is not recommended to park in the parking lot. In this case, the vehicle terminal obtains the available parking lots around the designated park; at the same time, it obtains the parking information of each parking lot for each time period from the external network, and displays the distance between the parking lot where the selected parking location is located and the surrounding available parking lots.

[0074] Example 2

[0075] According to another aspect of the embodiments of the present invention, the present invention also provides a vehicle management system for smart parking, such as... Figure 2 As shown, it includes:

[0076] License plate collection units are installed in each parking lot to collect license plate information of vehicles entering the parking lot;

[0077] The vehicle monitoring unit is used to monitor the location of vehicles entering the parking lot, thereby obtaining the parking location and movement path of the vehicles entering the parking lot;

[0078] Both the license plate acquisition unit and the vehicle monitoring unit are connected to the background monitoring device in Embodiment 1, such as through a remote connection.

[0079] Understandably, the vehicle monitoring unit consists of several cameras distributed throughout the parking lot.

[0080] In this embodiment, the vehicle terminal can be a driver's user terminal, or it can be a vehicle terminal itself, or a combination of a vehicle terminal and a user terminal to form a vehicle-side terminal device for vehicle navigation. In this embodiment, both the vehicle terminal and the user terminal include, during the vehicle navigation process, a positioning unit, a navigation unit, and a parking communication unit.

[0081] The positioning unit is used to obtain the vehicle's location;

[0082] The navigation unit provides navigation information to the vehicle and can also predict the time it will take for the vehicle to reach its destination based on the route.

[0083] The parking lot communication unit is used to establish communication connections between each parking lot and the control center.

[0084] The control device is used to receive reservation information sent by the vehicle terminal equipment. After receiving the reservation information, it determines whether the parking space meets the parking needs when the vehicle to which the vehicle terminal belongs arrives at the parking lot.

[0085] In this embodiment, the control device can be a control center within a designated park. The parking lot to which the user-selected parking space belongs can be a public parking lot within the designated park. If the parking lot is not within the designated park, the control device interacts with the control device of that parking lot and not with the control device within the designated park. This embodiment uses the interaction with the control device within the designated park as an example for explanation.

[0086] like Figure 2 As shown, the method by which the control center determines whether the parking space meets the parking requirements when a user's vehicle arrives at the parking lot is as follows:

[0087] STP1: When a vehicle needs to find a parking lot, the user logs into the platform through a terminal device (such as a mobile phone), selects a parking lot as a reserve parking lot, and then sends a reservation information to the control center through the terminal device.

[0088] Stp2: Obtain the location of the vehicle that issued the reservation information through the positioning unit in the user terminal device, then obtain the location of the reserved parking lot in the designated park, and predict the time t1 required for the vehicle to enter the reserved parking lot from the current location through the navigation unit in the user terminal device.

[0089] Based on the current time and duration t1, obtain the predicted time t2 when the vehicle arrives at the designated parking lot;

[0090] Stp3: Divide the day into r time periods. These time periods can be divided evenly or reasonably according to the application environment of the parking lot. For example, parking lots in tourist attractions can be divided according to peak visitor entry and peak visitor exit periods.

[0091] Obtain the average predicted number of parking spaces kj in the reserve parking lot during each time period, where 1≤j≤r, and r is a preset value;

[0092] The method for calculating the average predicted number of parking spaces kj in each time period is as follows:

[0093] Within a time period of a day, the actual number of cars parked in the parking lot is obtained sequentially at preset time intervals ty1, where 1≤i≤n, and n represents the number of times sampling is performed within the corresponding time period, i.e., the number of times the actual number of cars parked is obtained within the corresponding time period.

[0094] Calculate the typical value ks for these n ksi;

[0095] Then, obtain the m typical values ​​ks corresponding to this time period over the past m days;

[0096] Continue to calculate the typical value kss of these m typical values ​​ks, and use the typical value kss as the average predicted number of parking spaces kj in the corresponding time period;

[0097] Typical values ​​refer to parameters that can represent the group characteristics of a set of parameters. Specifically, they can be the average, median, or the average of the remaining values ​​after removing the values ​​with large deviations from the set of parameters.

[0098] It should be noted that the m past days for sampling can be consecutive or non-consecutive. The specific timeframe can be adjusted based on the usage context of the parking lot. For ease of understanding, here is an example: For shopping mall parking lots, since the number of parking spaces differs between weekdays and weekends, the sampling process separates weekends and weekdays, and then takes m days for each weekday for sampling and calculation.

[0099] In the above steps, when obtaining the average number of parking spaces k for each time period, the standard deviation F of n ksi values ​​is also calculated, thereby further obtaining the m standard deviations F obtained in the past m days; and the typical value Fd of these m standard deviations is calculated.

[0100] The typical value Fd for each time period was calculated using the method described above.

[0101] Stp4: Obtain the time period within the preparatory parking lot where the predicted time t2 falls. To facilitate differentiation, mark this time period as the target time period.

[0102] Obtain the average predicted number of parked cars (kj) for the target time period;

[0103] Get the maximum number of parking spaces in the prepared parking lot, kmax (the number of active parking spaces that can be put into operation);

[0104] The kj and kmax parameters corresponding to each parking lot can be stored in the parking lot's control device and updated according to actual changes;

[0105] Get the actual average number of cars parked in the reserve parking lot during the time period preceding the current time period;

[0106] Obtain the average predicted number of parking spaces (kjf) in the pre-parking lot during the time period preceding the current time period;

[0107] It should be noted that when performing the above steps, parking spaces reserved for specific individuals and vehicles with such spaces are not taken into consideration. That is, the average number of parking spaces and the maximum number of parking spaces mentioned above do not include parking spaces reserved for specific individuals.

[0108] In one embodiment of the present invention, the actual average number of parking spaces kf is calculated in the same way as ks, that is:

[0109] Within a time period preceding the current time period, the actual number of cars parked in the parking lot is obtained sequentially at preset time intervals ty1. A typical value of these actual number of cars parked is obtained, and this typical value is used as kf.

[0110] The reliability coefficient Q of the reserved parking lot corresponding to this reservation information is calculated according to the formula Q=(kj / kmax)*(kf / kjf);

[0111] Stp5. When the reliability coefficient Q>Qy, it is considered that parking spaces are scarce and inconvenient when the vehicle enters the preparatory parking lot, and it is not recommended to park in that parking lot.

[0112] Conversely, when the reliability coefficient Q≤Qy, it is assumed that there are sufficient parking spaces when the vehicle enters the reserved parking lot and can proceed to park.

[0113] Where Qy is a preset coefficient.

[0114] This invention analyzes historical data on parking space utilization in parking lots at different times to predict the availability of parking spaces when users arrive at the parking lot. Based on the prediction results, it provides reasonable guidance to users on choosing a parking location, avoiding parking difficulties and making reasonable use of surrounding available parking resources. It also considers the impact of temporary factors on the prediction results, thus improving the accuracy of the prediction.

[0115] Optionally, after logging into the platform through the vehicle terminal device, if the selected parking lot does not have good parking conditions, that is, when the reliability coefficient Q>Qy is true in the embodiment, the user can also select a destination. The navigation unit in the vehicle terminal obtains the available parking lots around the destination and interacts with the control centers of these available parking lots to obtain the reliability coefficient Q corresponding to each parking lot.

[0116] The system can display the reliability coefficient Q for each parking lot and the distance X between the parking lot and the destination, allowing users to select the appropriate option based on their needs.

[0117] In practical applications, the vehicle terminal can also sort the parking lots by combining the reliability coefficient Q and the distance X between the corresponding parking lot and the destination, intuitively showing the optimal selection of each parking lot.

[0118] The second method is illustrated below:

[0119] First, we determine whether the proximity coefficient Q satisfies Q≤Qy, then filter parking lots that meet this condition, and recommend parking lots in ascending order of distance X.

[0120] After logging into the platform through the vehicle terminal device, users can select a destination, and the navigation unit in the vehicle terminal will obtain the available parking lots around the destination.

[0121] Obtain the typical value Fd corresponding to each parking lot that satisfies the reliability coefficient Q≤Qy;

[0122] The optional coefficient W for each parking lot is calculated according to the formula W=α1*Fd+α2*X; where α1 and α2 are preset coefficient values;

[0123] Parking lots are recommended in ascending order of their selectability coefficient W. This invention introduces a stability evaluation parameter for the number of vehicles parked in each parking lot during different time periods. When recommending parking lots, it prioritizes those with higher stability, improving the accuracy of the recommendations and preventing situations where sudden changes in the number of vehicles parked in a parking lot cause parking difficulties for reserved vehicles upon arrival.

[0124] The above description is merely an example and illustration of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A vehicle management system for smart parking, characterized in that, include: The background monitoring device is used to obtain in real time the basic information of each vehicle entering the designated park parking lot monitored by each monitoring unit. The basic information includes: vehicle identification, vehicle entry time information, and vehicle type; vehicle types include: concrete trucks, transport trucks, buses, private cars, and police vehicles. The control device is used to determine whether a vehicle is a reserved vehicle based on its basic information. If so, it determines whether the area to which the current vehicle type belongs is available. If so, it sends a guidance instruction to the vehicle terminal. The guidance instruction includes sending driving information at each node of the guidance route. If a designated park has multiple public parking lots and public driving areas, and the public parking lots meet the pre-reservation mechanism, the system further includes: The control device is further configured to receive a reservation request from any first vehicle, the reservation request of the first vehicle including: selected available parking location information and current vehicle location information; Based on the reservation request, obtain the predicted time t2 when the first vehicle arrives at the selected parking location; Based on the predicted time t2, the parking strategy information established from pre-historical data, and the reliability coefficient of the selected parking location in the reservation request, a response message indicating whether or not the reservation is approved is sent to the first vehicle according to the reliability coefficient. The control device is the control center within the designated park, and the parking lot to which the user selects the parking space is a public parking lot within the designated park. Specifically, a day is divided into r time periods; the average predicted number of parking spaces kj in each time period is obtained from the parking strategy information of the parking lot to which the selected parking location belongs, where 1≤j≤r and r is a preset value; Mark the time period in which the predicted time t2 is located as the target time period; obtain the average predicted number of parking spaces kj corresponding to the target time period from the parking strategy information. 目标 ; and obtain the maximum number of parking spaces in the parking lot during the target time period kma; obtain the actual average number of parking spaces in the parking lot during the time period before the current time corresponding to the current time period kf; obtain the average predicted number of parking spaces in the parking lot during the time period before the current time corresponding to the current time period kjf; According to the formula The reliability coefficient Q of the reserved parking lot corresponding to this reservation information is calculated; When the reliability coefficient Q > Qy 指定数 When this happens, the response information includes a recommendation not to park in that parking lot; where Qy 指定数 These are preset coefficients; The system determines whether the coefficient Q satisfies Q≤Qy, filters parking lots that meet this condition, and recommends parking lots in ascending order of the distance X between the parking lot and the destination. After logging into the platform through the vehicle terminal device, the user selects a destination, and the navigation unit in the vehicle terminal obtains the available parking lots around the destination. Obtain the typical value Fd corresponding to each parking lot that satisfies the reliability coefficient Q≤Qy; specifically, when obtaining the average number of parking spaces k for each time period, the standard deviation F of n ksi values ​​is also calculated, and the m standard deviations F obtained in the past m days are obtained, and the typical value Fd of the m standard deviations is calculated; ksi is the actual number of parking spaces in the parking lot obtained at preset time intervals ty1 within a time period within a day; 1≤i≤n, where n represents the number of samplings in the corresponding time period; According to the formula The selectable coefficient W for each parking lot is calculated; where α1 and α2 are preset coefficient values. Parking lots are recommended in ascending order of their selectability coefficient W.

2. A vehicle management system for intelligent parking according to claim 1, characterized in that, Parking strategy information is established based on historical data. The average predicted number of parking spaces (kj) for each time period in the parking lot where the selected parking location is located is obtained from this information, including: Calculate the typical value ks for these n ksi; Then, obtain the m typical values ​​ks corresponding to this time period over the past m days; Continue to calculate the typical value kss of these m typical values ​​ks, and use the typical value kss as the average predicted number of parking spaces kj in the corresponding time period; The parking strategy information records the average predicted number of parking spaces (kj) and the maximum number of parking spaces (kmax) for each time period within a day on non-holidays, obtained from historical data, and the average predicted number of parking spaces (kj) and the maximum number of parking spaces (kmax) for each time period within a day on holidays.

3. A vehicle management system for intelligent parking according to claim 1, characterized in that, The control device is also used for: The system monitors vehicle information that sends guidance commands in real time, extracts vehicle operation behavior information from the vehicle information, compares it with preset vehicle operation behavior information, and issues an alarm if there is any inconsistency.

4. A vehicle management system for intelligent parking according to claim 1, characterized in that, The designated park area includes a portion of the electronic fence area and a portion of the non-electronic fence area. The non-electronic fence area is the area of ​​public driving area and multiple public parking lots. The control device interacts with the vehicle's terminal, which includes a positioning unit, a navigation unit, and a processing unit.

5. A vehicle management system for intelligent parking according to claim 2, characterized in that, Parking strategy information based on historical data includes: A reinforcement learning model is used to construct the actual number of parked cars in any time period before the current target time period, and the predicted number of parked cars in any time period within the next three days.

6. A vehicle management system for intelligent parking according to claim 1, characterized in that, If the response information includes a recommendation not to park in that parking lot, the vehicle terminal will obtain available parking lots around the designated park area. At the same time, it retrieves parking information for each parking lot at each time period from the external network, and displays the distance between the selected parking location and the surrounding available parking lots.

Images