Parking lot vehicle management method and system based on vehicle AI platform

By calculating the recommended parking spaces and congestion prediction coefficients in the parking area, and optimizing the parking space recommendation, the problems of congestion and high energy consumption caused by vehicle concentration in the prior art are solved, and the efficiency and safety of parking lots are improved.

CN118675365BActive Publication Date: 2024-11-19ROPEOK TECHNOLOGY GROUP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411148287.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-19
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

The existing parking lot vehicle AI management platform can only count the idle parking spaces in the parking lot and recommend them based on the distance between the vehicle and the idle parking spaces, resulting in concentrated congestion of vehicles in the parking area, increasing parking time and energy consumption, and the probability of accidents is high.

Method used

By obtaining the parking space area, obstacle location, idle parking space location, traffic flow and driving speed in each parking area, calculate the parking space recommendation and congestion prediction coefficient, determine the priority recommendation of each idle parking space, optimize the parking space recommendation, and reduce the vehicle concentrated.

Benefits of technology

It improves the efficiency of parking lot vehicles, reduces driver parking time and energy consumption, and reduces the probability of accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118675365B_ABST
    Figure CN118675365B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of parking lot vehicle management, and specifically to a parking lot vehicle management method and system based on a vehicle AI platform, the method comprising: obtaining the area of ​​each parking space in each parking area, the area of ​​a parking lane, the coordinates of the location of obstacles, and the coordinates of the location of all vacant parking spaces at each time, the traffic volume of each parking area at each time, and the vehicle driving speed; determining the parking space recommendation degree of each vacant parking space at each time; determining the parking space congestion prediction coefficient of each parking area at each time; determining the priority recommendation degree of each vacant parking space at each time, so as to recommend parking spaces to drivers and realize vehicle management in the parking lot. The present application can improve the efficiency of vehicle management in the parking lot and reduce the waiting time and energy consumption required for parking.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of parking lot vehicle management, and specifically to a parking lot vehicle management method and system based on a vehicle AI platform. Background Art

[0002] With the development of science and technology, all walks of life have gradually adopted information-based and digital AI management platforms. In recent years, with the introduction of parking lot vehicle AI management platforms, the management efficiency of parking lots for vehicles has been significantly improved.

[0003] Usually, the AI ​​management platform for parking vehicles mainly collects data on vehicles entering and exiting the parking lot and the data on vehicles actually parked in the parking lot, and then provides parking space availability information in each area for vehicles entering the parking lot, making it easier for drivers to find parking spaces. However, most of the current AI management platforms for parking vehicles can only count the available parking spaces in the parking lot, and recommend parking spaces based on the distance between the vehicle and the available parking spaces, which results in most of the recommended parking spaces for vehicles entering the same entrance being concentrated in a certain area, resulting in a large number of vehicles parked in the area at the same time, eventually forming congestion, which not only increases the time spent on parking and increases the possibility of accidents, but also further increases the energy consumption of the car, causing pollution to the environment. Summary of the invention

[0004] In order to solve the above technical problems, a parking lot vehicle management method and system based on a vehicle AI platform are provided to solve the existing problems.

[0005] The solution to the technical problem of this application is to provide a parking lot vehicle management method and system based on a vehicle AI platform, including the following steps:

[0006] In a first aspect, an embodiment of the present application provides a parking lot vehicle management method based on a vehicle AI platform, the method comprising the following steps:

[0007] Obtain the area of ​​each parking space in each parking area, the area of ​​the parking lane, the coordinates of the obstacle position, the coordinates of all vacant parking spaces at each time, the traffic volume and vehicle speed of each parking area at each time;

[0008] According to the changes in the traffic volume, driving speed and number of vacant parking spaces in each parking area at different times, as well as the vacant area around the vacant parking spaces, the parking space recommendation degree of each vacant parking space at each time is determined;

[0009] According to the changes in the traffic volume of each parking area at different times and the differences in the peak traffic periods, the parking space congestion prediction coefficient of each parking area at each time is determined;

[0010] Based on the parking space recommendation degree and the parking space congestion prediction coefficient, the priority recommendation degree of each vacant parking space at each moment is determined to recommend parking spaces to drivers and realize vehicle management in the parking lot.

[0011] Preferably, determining the parking space recommendation degree of each vacant parking space at each moment includes:

[0012] Analyze the changes in traffic volume, driving speed, and vacant parking spaces at each parking area at each time and at previous times to determine the regional congestion of each parking area at each time;

[0013] Analyze the free area around each vacant parking space at each time and the distance to obstacles to determine the parking spaciousness of each vacant parking space at each time;

[0014] The ratio between the parking spaciousness of each vacant parking space at each moment and the area congestion degree of the parking area where the vacant parking space is located at the corresponding moment is used as the parking space recommendation degree of each vacant parking space at each moment.

[0015] Preferably, the determining of the regional congestion degree of each parking area at each time comprises:

[0016] The ratio between the average of the vehicle speed at each parking area at each time and the previous multiple times and the average of the vehicle flow is used as the vehicle flow variation coefficient at each parking area at each time;

[0017] The variance of the number of free parking spaces in each parking area at each moment and the number of previous moments is calculated, and the reciprocal of the sum of the variance and the traffic flow variation coefficient is used as the regional congestion degree of each parking area at each moment.

[0018] Preferably, the determining of the parking space spaciousness of each vacant parking space at each moment includes:

[0019] Analyze the area of ​​each vacant parking space at each moment, the area of ​​its adjacent vacant parking spaces, and the area of ​​the adjacent parking lanes, and determine the free area of ​​each vacant parking space at each moment;

[0020] Calculate the sum of the distances between the position coordinates of each free parking space at each moment and the coordinates of all obstacles in the parking area where it is located as the obstacle influence degree of each free parking space at each moment;

[0021] The free area and the obstacle influence are combined to determine the parking space spaciousness of each free parking space at each moment.

[0022] Preferably, the determining of the free area of ​​each vacant parking space at each moment includes:

[0023] The sum of the area of ​​each vacant parking space at each moment and the area of ​​its adjacent parking lane is recorded as the available area of ​​each vacant parking space at each moment;

[0024] The sum of the available areas of each vacant parking space and its adjacent vacant parking spaces at each moment is taken as the vacant area of ​​each vacant parking space at each moment.

[0025] Preferably, the determining of the parking space congestion prediction coefficient of each parking area at each moment includes:

[0026] Analyze the peak changes and trend changes of the traffic volume of each parking area at multiple times before each time, and determine the peak congestion degree and adjacent peak degree of each parking area at each time;

[0027] The peak congestion degree and the adjacent peak degree are combined to determine the parking space congestion prediction coefficient of each parking area at each moment.

[0028] Preferably, determining the peak congestion degree of each parking area at each time comprises:

[0029] Obtain all vehicle flows of each parking area at multiple moments before each moment, form a vehicle flow sequence of each parking area at each moment, and use a threshold segmentation algorithm to obtain the optimal threshold;

[0030] Recording the traffic flow in the traffic flow sequence that is greater than or equal to the optimal threshold as the peak traffic flow;

[0031] The number of all the peak traffic flows is used as the duration of the peak in each parking area at each moment;

[0032] Dividing the traffic flow sequence into multiple subsequences, and taking the average of the maximum values ​​of the peak traffic flows in all the subsequences as the peak traffic flow of each parking area at each moment;

[0033] The product of the peak traffic flow rate and the duration of the peak is taken as the peak congestion degree of each parking area at each moment.

[0034] Preferably, determining the proximity peak level of each parking area at each moment includes:

[0035] A trend decomposition algorithm is used on the traffic flow sequence to calculate the trend intensity;

[0036] The product of the mean of the first-order difference result of the traffic flow sequence and the trend strength is used as the traffic flow trend degree of each parking area at each moment;

[0037] The duration between each moment and the moment corresponding to the previous peak traffic flow is obtained, and the duration is integrated with the degree of traffic flow trend to determine the degree of adjacent peaks for each parking area at each moment.

[0038] Preferably, determining the priority recommendation degree of each vacant parking space at each time to recommend a parking space to the driver includes:

[0039] The product of the parking space recommendation degree of each free parking space at each moment and the parking space congestion prediction coefficient of the parking area where the free parking space is located at the corresponding moment is used as the priority recommendation degree of each free parking space at each moment;

[0040] Arrange the priority recommendation of all vacant parking spaces at each time in descending order to recommend parking spaces to drivers.

[0041] In the second aspect, an embodiment of the present application also provides a parking lot vehicle management system based on a vehicle AI platform, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, it implements the steps of any one of the above-mentioned parking lot vehicle management methods based on the vehicle AI platform.

[0042] This application has at least the following beneficial effects:

[0043] The present application determines the parking space recommendation degree of each free parking space at each time according to the changes in the traffic volume, driving speed and number of free parking spaces in each parking area at different times, as well as the free area around the free parking space. The beneficial effect of this application is that it takes into account the congestion of the parking area, analyzes the influence of the free area and obstacles around the free parking spaces, and thus the impact on the recommendation of parking spaces; according to the changes in the traffic volume of each parking area at different times and the differences in the peak traffic periods, the parking space congestion prediction coefficient of each parking area at each time is determined. The beneficial effect of this application is that it takes into account the congestion of the parking area during the parking peak period and the parking space that is about to enter the parking lot at each time. The possibility of peak hours; based on the parking space recommendation degree and the parking space congestion prediction coefficient, the priority recommendation degree of each vacant parking space at each moment is determined to recommend parking spaces to drivers and realize vehicle management in the parking lot. Its beneficial effect is that it solves the problem that the current parking lot management platform can only recommend parking spaces by counting the vacant parking spaces in the parking lot and according to the distance between the vehicle and the vacant parking spaces, which leads to a small vacant parking space area and a large traffic volume in the parking area, resulting in a long waiting time for drivers to park, a high probability of accidents, and a large energy consumption. It improves the parking lot's efficiency in managing vehicles in the parking area and reduces the time and energy consumption required for drivers to park. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The parking lot vehicle management method based on the vehicle AI platform of the present application is further described in detail below in conjunction with the accompanying drawings.

[0045] Figure 1 A flowchart of the steps of the parking lot vehicle management method based on the vehicle AI platform provided in this application;

[0046] Figure 2 A schematic diagram of the parking spaces and parking lanes provided for this application;

[0047] Figure 3 A flowchart of the steps of the method for obtaining the parking space recommendation degree of each vacant parking space at each time provided by this application;

[0048] Figure 4 A flowchart of the steps of the method for obtaining the parking space congestion prediction coefficient of each parking area at each moment provided in this application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of this application more clear, the following is a further detailed description of the parking lot vehicle management method and system based on the vehicle AI platform proposed in this application in conjunction with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0050] Unless defined otherwise, 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 application belongs.

[0051] See also Figure 1 , which shows a flowchart of a parking lot vehicle management method based on a vehicle AI platform provided by an embodiment of the present application, the method comprising the following steps:

[0052] Step 1: Obtain the area of ​​each parking space in each parking area, the area of ​​the parking lane, the coordinates of the obstacle position, the coordinates of all vacant parking spaces at each time, the traffic flow and vehicle speed of each parking area at each time.

[0053] Usually, in order to maximize the use of parking space and facilitate management, parking lots are divided into multiple parking areas, each of which has multiple parking spaces.

[0054] The following data is obtained from the design drawing of the parking lot through the parking lot management system: the parking space area of ​​each parking space in each parking area, the area of ​​the parking lane and the coordinates of obstacles, where the obstacles include but are not limited to the parking lot load-bearing columns, parking lot area gates and other immovable obstacles;

[0055] By setting a gravity sensor in the parking space, when the gravity sensor recognizes that there is no gravity pressure from the vehicle, the parking space is determined to be in an idle state, and the position coordinates of the gravity sensor are used as the position coordinates of the idle parking space; the number of vehicles running in each parking area is recorded as the traffic flow of each parking area; using the parking lot management system, data is collected every t time, and the position coordinates and traffic flow of all idle parking spaces in each parking area at each time are obtained;

[0056] Through the surveillance cameras in the parking lot, we first use image recognition technology to obtain the first All vehicles at the moment are tracked separately by the target tracking algorithm, and the tracking time is , obtained through The positions and movement trajectories of all vehicles are calculated, and the ratio of the length of each vehicle's movement trajectory to the tracking time is calculated as the driving speed of each vehicle. The average driving speed of all vehicles is taken as the first The vehicle travel speed at each time is obtained.

[0057] Preferably, in this embodiment, the data is collected every 1 second. The time interval is 5s. As other implementation methods, the implementer can set it according to the actual situation.

[0058] At this point, the parking area of ​​each parking space in each parking area, the area of ​​the parking lane, the obstacle coordinates, the position coordinates of all vacant parking spaces at each moment, the traffic volume and vehicle speed of each parking area at each moment are obtained.

[0059] Step 2: Determine the parking space recommendation degree of each free parking space at each time according to the traffic volume, driving speed and the number of free parking spaces in each parking area at different times, as well as the free area around the free parking spaces.

[0060] Due to the large daily parking demand, in order to meet the greater parking demand, the vacant spaces in the parking lot are usually planned as parking spaces as much as possible, which leads to narrow driving spaces in each area of ​​the parking lot. When encountering the peak period of vehicle entry and exit, the traffic volume in the parking lot is large, and the driver has to drive slowly and needs to avoid the passing vehicles, which eventually leads to slow driving in each area of ​​the parking lot. It takes a long time for vehicles to enter and exit. Therefore, when recommending parking spaces, parking areas with less traffic in the area should be selected. In addition, since the planning of parking spaces is usually continuous, that is, the interval between parking spaces is small, when there are many parked vehicles, the smaller the vacant space around each parking space, the higher the possibility of accidents when parking vehicles, and the longer the time required to park vehicles, so parking spaces with larger vacant spaces should be recommended. For example, if the adjacent parking spaces of a parking space are all parked, the actual parking space of the parking space is smaller, and the possibility of accidents is greater. If the adjacent spaces of a parking space are not parked, or the fewer the cars are parked, the actual parking space of the parking space is larger, the possibility of accidents is smaller, and it is more recommended to park.

[0061] Based on the above analysis, the free area around the vacant parking spaces at each time and the traffic volume are analyzed to determine the parking space recommendation degree of each vacant parking space at each time, specifically:

[0062] The ratio between the average of the vehicle speed at each parking area at each time and the previous multiple times and the average of the vehicle flow is used as the vehicle flow variation coefficient at each parking area at each time;

[0063] Calculate the variance of the number of free parking spaces at each parking area at each moment and its previous moments;

[0064] The inverse of the sum of the variance and the traffic flow variation coefficient is used as the regional congestion degree of each parking area at each moment;

[0065] Preferably, in this embodiment, the vehicle speed, traffic volume and number of vacant parking spaces at each moment and the five moments before it are selected for calculation to obtain the traffic flow variation coefficient and variance. As other implementation methods, the implementer can set them according to the actual situation.

[0066] It should be noted that the smaller the traffic flow variation coefficient is, the slower the driving speed is and the larger the traffic volume is. The smaller the change in the number of vacant parking spaces is, the smaller the variance is and the greater the congestion of the obtained area is. This means that the traffic volume in the parking area where the parking space is located is large and the driving is slow. In addition, vehicles enter and exit the parking area more frequently, and the driver needs to wait longer to park. Therefore, it is less recommended to choose this area for parking, and the corresponding parking space recommendation is lower.

[0067] Furthermore, the free space around the vacant parking spaces is analyzed. The larger the actual parking space, the lower the possibility of accidents. Therefore, the parking space recommendation is made. Therefore, the parking spaciousness is constructed as follows:

[0068] The sum of the area of ​​each vacant parking space at each moment and the area of ​​its adjacent parking lane is recorded as the available area of ​​each vacant parking space at each moment;

[0069] The sum of the available areas of each vacant parking space and its adjacent vacant parking spaces at each moment is taken as the vacant area of ​​each vacant parking space at each moment;

[0070] It should be noted that, taking a vacant parking space as an example, the free area means the area where a vehicle can park in the parking space and move without colliding with other vehicles or obstacles. Figure 2 As shown, 1 represents an empty parking space, 2 and 3 are two empty parking spaces adjacent to the empty parking space 1, 4 is a parking lane, 5, 6, and 7 are the parking lane areas adjacent to the empty parking spaces 2, 1, and 3 respectively, the available area of ​​the empty parking space 1 is the sum of the area of ​​the empty parking space 1 and the parking lane area 6, and the free area of ​​the empty parking space 1 is the sum of the areas of the empty parking spaces 1, 2, and 3 and the adjacent parking lane areas 6, 5, and 7; if the parking space 3 is not an empty parking space, the free area of ​​the empty parking space 1 is the sum of the areas of the empty parking spaces 1 and 2 and the adjacent parking lane areas 6 and 5.

[0071] Calculate the sum of the distances between the position coordinates of each free parking space at each moment and the coordinates of all obstacles in the parking area where it is located as the obstacle influence degree of each free parking space at each moment;

[0072] Preferably, in this embodiment, the sum of the Euclidean distances between the position coordinates of each vacant parking space at each moment and the obstacle coordinates of the parking area where the vacant parking space is located is calculated as the obstacle influence degree of each vacant parking space at each moment.

[0073] The free area and the obstacle influence are combined to determine the parking spaciousness of each free parking space at each moment;

[0074] It can be understood that the specific fusion relationship can be an addition relationship, a multiplication relationship, etc. The specific relationship is determined according to the actual scenario, and this embodiment does not impose any special restrictions on this.

[0075] Preferably, in this embodiment, the sum of the parking vacant area and the obstacle impact degree is used as the parking spaciousness of each vacant parking space at each moment; in other embodiments, the product of the parking vacant area and the obstacle impact degree is used as the parking spaciousness of each vacant parking space at each moment.

[0076] It should be noted that the larger the free area, the larger the parking space of the corresponding free parking space, and the greater the obstacle impact, which means the corresponding free parking space is farther away from the obstacles in the parking area, and the smaller the possibility of being obstructed when parking. The larger the resulting parking spaciousness, the more suitable the free parking space is for being recommended to the driver for parking.

[0077] Further, based on the parking space spaciousness and the area congestion, a parking space recommendation is determined, specifically:

[0078] The ratio between the parking spaciousness of each vacant parking space at each moment and the area congestion degree of the parking area where the vacant parking space is located at the corresponding moment is used as the parking space recommendation degree of each vacant parking space at each moment;

[0079] It should be noted that the greater the parking space recommendation degree, the lower the area congestion, and the higher the parking space spaciousness around the vacant parking space, the more drivers should be recommended to park there.

[0080] Furthermore, the flowchart of the method for obtaining the parking space recommendation degree of each vacant parking space at each time provided in the embodiment of the present application is as follows: Figure 3 shown.

[0081] At this point, the parking space recommendation degree of each vacant parking space at each time is obtained.

[0082] Step 3: Determine the parking congestion prediction coefficient of each parking area at each time according to the changes in the traffic volume of each parking area at different times and the differences in the peak traffic periods.

[0083] Under normal circumstances, parking lots will have certain peak periods, such as office buildings with large parking volumes and fewer cars picked up during working hours, and large cars picked up during off-get off work hours and fewer parked cars, or commercial buildings, where parking volumes may be larger during lunch or dinner. During peak periods, the entire parking lot is busy, and most of the vacant parking spaces have low recommended degrees and little difference. At this time, if the recommendation is made based on the highest recommended degree, it may result in the recommended parking spaces being more difficult to park in and more time spent waiting, which cannot solve the problems of long parking times for drivers, high probability of accidents, and environmental pollution. Therefore, further analysis is needed.

[0084] Furthermore, there are certain differences in the peak hours in each area of ​​the parking lot. For example, the parking area closer to the mall entrance will attract more drivers because it is closer to the mall. During the parking peak period, the traffic volume will be larger and the duration will be relatively longer. However, the parking area used by residents on a daily basis has a larger traffic volume during the rush hour and the duration is relatively short. Therefore, the peak traffic characteristics of each parking area are analyzed to construct the parking space congestion prediction coefficient.

[0085] First, analyze the traffic flow changes and peak duration during peak traffic flow to determine the peak congestion, specifically:

[0086] Obtain all vehicle flows of each parking area at multiple moments before each moment to form a vehicle flow sequence of each parking area at each moment, and use a threshold segmentation algorithm to obtain the optimal threshold;

[0087] Recording the traffic flow in the traffic flow sequence that is greater than or equal to the optimal threshold as the peak traffic flow;

[0088] The number of all the peak traffic flows is used as the duration of the peak in each parking area at each moment;

[0089] Preferably, in this embodiment, all vehicle flows in each parking area within 1 day before each moment are selected to form a vehicle flow sequence for each parking area at each moment; the Otsu threshold segmentation method is used to obtain the optimal threshold, wherein the Otsu threshold segmentation method is a well-known technology and will not be described in detail here. As other implementation methods, the implementer adopts other methods of the prior art, such as frequency histograms, etc., and this embodiment does not impose any special restrictions on this.

[0090] Dividing the traffic flow sequence into multiple subsequences; taking the average of the maximum values ​​of the peak traffic flows in all the subsequences as the peak traffic flow of each parking area at each moment;

[0091] Preferably, in this embodiment, the traffic flow sequence is divided into a subsequence every hour, and the traffic flow sequence is divided into multiple subsequences. As other implementation methods, the implementer can divide them according to actual conditions.

[0092] The product of the peak traffic flow rate and the duration of the peak is taken as the peak congestion degree of each parking area at each moment;

[0093] In this embodiment, the peak congestion degree of each parking area at each time is: ,in, For the Moment The peak congestion of a parking area, For the Moment The peak traffic volume in each parking area, For the Moment The duration of the peak period in each parking area.

[0094] It should be noted that if there is a subsequence without peak traffic flow, the maximum value of the peak traffic flow in the subsequence is 0; the greater the peak congestion, the higher the traffic flow of the parking area during the peak period compared with other parking areas, and the longer the duration, and the less likely it is that the parking spaces in the parking area should be recommended for parking.

[0095] Furthermore, the changing trend of the traffic flow in each parking area at each time is analyzed before each time, and the adjacent peak degree is constructed to reflect the possibility of entering the peak period at each time, specifically:

[0096] Applying a trend decomposition algorithm to the traffic flow sequence and calculating the trend intensity;

[0097] Preferably, in this embodiment, the STL sequence decomposition algorithm is used to decompose the traffic flow sequence and calculate the trend strength, wherein the STL sequence decomposition algorithm and the trend strength calculation method are well-known technologies and will not be repeated here; as other implementation methods, the implementer may adopt other methods of the prior art, such as the X-11 sequence decomposition algorithm, etc., and this embodiment does not impose any special restrictions on this.

[0098] It should be noted that, in this embodiment, the STL sequence decomposition algorithm is used to decompose the traffic flow sequence to obtain a trend sequence and a residual sequence. The formula for the trend strength is: ,in, is the trend strength, is the variance of the residual sequence, is the variance of the trend series and the residual series, To find the maximum value.

[0099] The product of the mean of the first-order difference result of the traffic flow sequence and the trend strength is used as the traffic flow trend degree of each parking area at each moment;

[0100] Obtain the duration between each moment and the moment corresponding to the previous peak traffic flow; combine the duration with the degree of traffic flow trend to determine the degree of adjacent peaks for each parking area at each moment;

[0101] Preferably, in this embodiment, the sum of the duration and the traffic flow trend degree is used as the proximity peak degree of each parking area at each moment.

[0102] It should be noted that the greater the degree of proximity to the peak, the longer the time interval between the parking area and the previous peak. At this time, the traffic volume gradually increases, and the faster the increase rate, the more likely the parking area is to enter the next parking peak period, and the less recommended it is to park in the vacant parking space.

[0103] Further, based on the peak congestion degree and the adjacent peak degree, a parking space congestion prediction coefficient is determined, specifically:

[0104] The peak congestion degree and the adjacent peak degree are combined to determine the parking space congestion prediction coefficient of each parking area at each time;

[0105] Preferably, in this embodiment, the product of the peak congestion degree and the adjacent peak degree is used as the parking space congestion prediction coefficient of each parking area at each moment.

[0106] It should be noted that the greater the peak congestion coefficient of a parking space, the greater the possibility that the parking area where the parking space is located is about to enter the parking peak period, and the longer it takes to park. At this time, it is less recommended for drivers to park in the parking area.

[0107] Further, the flowchart of the steps of the method for obtaining the parking space congestion prediction coefficient of each parking area at each time provided in the embodiment of the present application is as follows: Figure 4 shown.

[0108] At this point, the parking space congestion prediction coefficient of each parking area at each time is obtained.

[0109] Step 4: Based on the parking space recommendation degree and the parking space congestion prediction coefficient, determine the priority recommendation degree of each vacant parking space at each moment, so as to recommend parking spaces to drivers and realize vehicle management in the parking lot.

[0110] Further, based on the parking space recommendation degree and the parking space congestion prediction coefficient, the priority recommendation order of each vacant parking space is arranged, and the priority recommendation degree of each vacant parking space is analyzed, specifically:

[0111] The product of the parking space recommendation degree of each free parking space at each moment and the parking space congestion prediction coefficient of the parking area where the free parking space is located at the corresponding moment is used as the priority recommendation degree of each free parking space at each moment;

[0112] It should be noted that the greater the priority recommendation degree of a parking space, the better the parking conditions in the parking area where the vacant parking space is located, the lower the traffic volume, the larger the vacant area, and the lower the possibility that the parking area will enter the parking peak period, and the more likely the vacant parking space should be recommended to the driver for parking.

[0113] When the driver enters the parking lot, the parking lot management system calculates the priority recommendation of each vacant parking space at each moment, and arranges the priority recommendation of all vacant parking spaces in descending order to recommend a parking space to the driver. The driver navigates to the vacant parking space according to the route of the selected vacant parking space to park.

[0114] Based on the same inventive concept as the above method, an embodiment of the present application also provides a parking lot vehicle management system based on a vehicle AI platform, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned parking lot vehicle management methods based on the vehicle AI platform are implemented.

[0115] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0116] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0117] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the present application. It should be pointed out that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application, shall fall within the protection scope of the technical solution of the present application.

Claims

1. A parking lot vehicle management method based on a vehicle AI platform, characterized in that: The method comprises the following steps: Obtain the area of ​​each parking space in each parking area, the area of ​​the parking lane, the coordinates of the obstacle position, the coordinates of all vacant parking spaces at each time, the traffic volume and vehicle speed of each parking area at each time; According to the changes in the traffic volume, driving speed and number of vacant parking spaces in each parking area at different times, as well as the vacant area around the vacant parking spaces, the parking space recommendation degree of each vacant parking space at each time is determined; According to the changes in the traffic volume of each parking area at different times and the differences in the peak traffic periods, the parking space congestion prediction coefficient of each parking area at each time is determined; Based on the parking space recommendation degree and the parking space congestion prediction coefficient, determining the priority recommendation degree of each vacant parking space at each time, so as to recommend parking spaces to drivers and realize vehicle management in the parking lot; The method of determining the parking space congestion prediction coefficient of each parking area at each time comprises: Analyze the peak changes and trend changes of the traffic volume of each parking area at multiple times before each time, and determine the peak congestion degree and adjacent peak degree of each parking area at each time; The peak congestion degree and the adjacent peak degree are combined to determine the parking space congestion prediction coefficient of each parking area at each time; Determining the peak congestion degree of each parking area at each time includes: Obtain all vehicle flows of each parking area at multiple moments before each moment, form a vehicle flow sequence of each parking area at each moment, and use a threshold segmentation algorithm to obtain the optimal threshold; Recording the traffic flow in the traffic flow sequence that is greater than or equal to the optimal threshold as the peak traffic flow; The number of all the peak traffic flows is used as the duration of the peak in each parking area at each moment; Dividing the traffic flow sequence into multiple subsequences, and taking the average of the maximum values ​​of the peak traffic flows in all the subsequences as the peak traffic flow of each parking area at each moment; The product of the peak traffic flow rate and the duration of the peak is taken as the peak congestion degree of each parking area at each moment; Determining the proximity peak level of each parking area at each moment includes: A trend decomposition algorithm is used on the traffic flow sequence to calculate the trend intensity; The product of the mean of the first-order difference result of the traffic flow sequence and the trend strength is used as the traffic flow trend degree of each parking area at each moment; The duration between each moment and the moment corresponding to the previous peak traffic flow is obtained, and the duration is integrated with the degree of traffic flow trend to determine the degree of adjacent peaks for each parking area at each moment.

2. The parking lot vehicle management method based on the vehicle AI platform according to claim 1, characterized in that: Determining the parking space recommendation degree of each vacant parking space at each time includes: Analyze the changes in traffic volume, driving speed, and vacant parking spaces at each parking area at each time and at previous times to determine the regional congestion of each parking area at each time; Analyze the free area around each vacant parking space at each time and the distance to obstacles to determine the parking spaciousness of each vacant parking space at each time; The ratio between the parking spaciousness of each vacant parking space at each moment and the area congestion degree of the parking area where the vacant parking space is located at the corresponding moment is used as the parking space recommendation degree of each vacant parking space at each moment.

3. The parking lot vehicle management method based on the vehicle AI platform as claimed in claim 2, characterized in that: The determining of the regional congestion degree of each parking area at each time comprises: The ratio between the average of the vehicle speed at each parking area at each time and the previous multiple times and the average of the vehicle flow is used as the vehicle flow variation coefficient at each parking area at each time; The variance of the number of free parking spaces in each parking area at each moment and the number of previous moments is calculated, and the reciprocal of the sum of the variance and the traffic flow variation coefficient is used as the regional congestion degree of each parking area at each moment.

4. The parking lot vehicle management method based on the vehicle AI platform as claimed in claim 2, characterized in that: Determining the parking space of each vacant parking space at each time includes: Analyze the area of ​​each vacant parking space at each moment, the area of ​​its adjacent vacant parking spaces, and the area of ​​the adjacent parking lanes, and determine the free area of ​​each vacant parking space at each moment; Calculate the sum of the distances between the position coordinates of each free parking space at each moment and the coordinates of all obstacles in the parking area where it is located as the obstacle influence degree of each free parking space at each moment; The free area and the obstacle influence are combined to determine the parking space spaciousness of each free parking space at each moment.

5. The parking lot vehicle management method based on the vehicle AI platform according to claim 1, characterized in that: Determining the free area of ​​each free parking space at each time includes: The sum of the area of ​​each vacant parking space at each moment and the area of ​​its adjacent parking lane is recorded as the available area of ​​each vacant parking space at each moment; The sum of the available areas of each vacant parking space and its adjacent vacant parking spaces at each moment is taken as the vacant area of ​​each vacant parking space at each moment.

6. The parking lot vehicle management method based on the vehicle AI platform according to claim 1, characterized in that: Determining the priority recommendation degree of each vacant parking space at each time to recommend a parking space to the driver includes: The product of the parking space recommendation degree of each free parking space at each moment and the parking space congestion prediction coefficient of the parking area where the free parking space is located at the corresponding moment is used as the priority recommendation degree of each free parking space at each moment; Arrange the priority recommendation of all vacant parking spaces at each time in descending order to recommend parking spaces to drivers.

7. A parking lot vehicle management system based on a vehicle AI platform, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the parking lot vehicle management method based on the vehicle AI platform as described in any one of claims 1-6 are implemented.