CN118430335B

CN118430335B - Parking lot vehicle traffic control method, device and storage medium

Info

Publication number: CN118430335B
Application number: CN202410877425.4A
Authority: CN
Inventors: 何毅斌; 钟淑容; 曾其勇; 王逸镇
Original assignee: China Merchants Jiyu Digital Technology Shenzhen Co ltd
Current assignee: China Merchants Jiyu Digital Technology Shenzhen Co ltd
Priority date: 2024-07-02
Filing date: 2024-07-02
Publication date: 2024-10-01
Anticipated expiration: 2044-07-02
Also published as: CN118430335A

Abstract

The application relates to the field of traffic control, and provides a parking lot vehicle traffic control method, a parking lot vehicle traffic control device and a storage medium. The method comprises the following steps: collecting real-time traffic flow videos entering and exiting a parking lot through a camera; analyzing the real-time traffic flow video, and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video; if the necessary lanes at the entrance and the exit of the parking lot are not congested, a common entrance and exit mode is started; if the necessary lanes at the entrance and the exit of the parking lot are congested, a rapid entrance mode or a rapid exit mode is started so that vehicles which enter and exit from the entrance and the exit of the parking lot rapidly enter and exit from the entrance and the exit of the parking lot; after the rapid entry mode is started, guiding the current entry vehicle to a target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is turned on, the departure vehicle is guided to the corresponding exit of the parking lot based on the dynamic exit guiding scheme. The scheme of the application can improve the traffic efficiency of the parking lot.

Description

Parking lot vehicle traffic control method, device and storage medium

Technical Field

The present application relates to the field of traffic control, and in particular, to a method, an apparatus, and a storage medium for controlling vehicle traffic in a parking lot.

Background

At present, a plurality of physical buildings such as hotels, extra-business buildings and the like, whether residential communities or cities, are provided with parking lots, so that the vehicle users can park the vehicles in daily life or in consumption. In these parking lots, a gateway is generally arranged at an entrance and a gateway, cameras are arranged beside the gateway, and vehicles need to enter or leave behind license plates of the vehicles acquired by the cameras. However, with some parking lots, as vehicles entering or leaving the parking lot are more and more, especially, in peak hours of business hours or holidays, due to various factors, such as scanning two-dimensional codes, checking license plates, checking fee and billing, inputting money and payment codes, cash change, and payment opening, etc., the existing parking lot vehicle passing control method requires that vehicle users drive in line at the entrance and exit of the parking lot, resulting in traffic congestion and greatly reduced passing efficiency.

Disclosure of Invention

The application provides a parking lot vehicle passing control method, a parking lot vehicle passing control device and a storage medium, which can improve the passing efficiency of a parking lot.

In one aspect, the present application provides a parking lot vehicle traffic control method, the method comprising:

Collecting real-time traffic flow videos entering and exiting a parking lot through a camera;

Analyzing the real-time traffic flow video, and judging whether congestion occurs in a necessary lane at the entrance and the exit of the parking lot according to the analysis result of the real-time traffic flow video;

If the necessary lanes at the entrance and the exit of the parking lot are not congested, a common entrance and exit mode is started;

If the necessary lanes at the entrance and the exit of the parking lot are congested, a rapid entrance mode or a rapid exit mode is started so that vehicles which enter and exit at present rapidly enter and exit from the entrance and the exit of the parking lot rapidly;

After the rapid entry mode is started, guiding the current entry vehicle to a target parking space based on the prediction of the parking space occupation condition of the parking lot;

And after the rapid departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm.

In another aspect, the present application provides a parking lot vehicle passage control apparatus, the apparatus comprising:

the acquisition module is used for acquiring real-time traffic flow videos entering and exiting the parking lot through the camera;

the judging module is used for analyzing the real-time traffic flow video and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video;

The first opening module is used for opening a common access mode if the traffic lane passing through the entrance and the exit of the parking lot is not congested;

the second opening module is used for opening a rapid entering mode or a rapid exiting mode if the necessary lanes at the entrance and the exit of the parking lot are congested, so that vehicles which enter and exit from the entrance and the exit of the parking lot rapidly;

the first guiding module is used for guiding the current entering vehicle to the target parking space based on the prediction of the parking space occupation condition of the parking lot after the rapid entering mode is started;

and the second guiding module is used for guiding the vehicles to be discharged to the corresponding outlet of the parking lot to be discharged based on a dynamic outlet lane allocation scheme or a departure path guidance planning algorithm after the rapid departure mode is started.

In a third aspect, the present application provides an apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the technical solution of the parking lot vehicle traffic control method as described above when executing the computer program.

In a fourth aspect, the present application provides a storage medium storing a computer program which, when executed by a processor, implements the steps of the technical solution of the parking lot vehicle traffic control method as described above.

According to the technical scheme provided by the application, when the traffic jam occurs in the necessary lanes at the entrance and the exit of the parking lot, the rapid entrance mode or the rapid exit mode is started; further, after the rapid entry mode is started, guiding the current entry vehicle to the target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm. On the one hand, when judging that the traffic lane at the entrance and exit of the parking lot is jammed, the rapid entrance mode or the rapid exit mode is started, so that the current vehicle can rapidly enter and exit the entrance and exit of the parking lot; on the other hand, based on the prediction of the parking space occupation condition of the parking lot, the current incoming vehicles are guided to the target parking space, so that long-time stay of the incoming vehicles in the parking lot is effectively avoided, traffic jam of lanes in the parking lot can be relieved objectively, and based on a dynamic exit lane distribution scheme or an exit path guidance planning algorithm, corresponding exit of the vehicles to be exited to the parking lot is guided, long-time stay of the exit vehicles in the parking lot is effectively avoided, and traffic jam of exit lanes of the parking lot can be relieved. In conclusion, the technical scheme of the application effectively improves the vehicle passing efficiency of the parking lot.

Drawings

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of a parking lot vehicle traffic control method provided by an embodiment of the present application;

fig. 2 is a schematic structural view of a parking lot vehicle traffic control device according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In this specification, adjectives such as first and second may be used solely to distinguish one element or action from another element or action without necessarily requiring or implying any actual such relationship or order. Where the environment permits, reference to an element or component or step (etc.) should not be construed as limited to only one of the element, component, or step, but may be one or more of the element, component, or step, etc.

In the present specification, for convenience of description, the dimensions of the various parts shown in the drawings are not drawn in actual scale.

In order to solve the above problems in the prior art, the present application provides a method for controlling vehicle traffic in a parking lot, which can improve traffic efficiency in the parking lot, and a flowchart thereof is shown in fig. 1, and mainly includes steps S101 to S106, and is described in detail as follows:

step S101: and acquiring real-time traffic videos entering and exiting the parking lot through a camera.

In the embodiment of the application, the video of the vehicle flow entering and exiting the parking lot can be acquired in real time through the camera arranged near the entrance and exit of the parking lot, wherein the video comprises real-time information such as the number of vehicles, the running speed, the vehicle density and the like of the entrance and exit lanes of the parking lot (namely, the entrance lane of the parking lot and the exit lane of the parking lot, wherein the entrance lane of the parking lot refers to the lane leading to the entrance of the parking lot, and the exit lane of the parking lot refers to the lane leading to the exit of the parking lot); of course, other sensors, such as non-visual sensors like ground feel, may also be used to assist in obtaining such information. The traffic flow condition can be analyzed together with the visual information collected by the camera and the information collected by other non-visual sensors such as the ground sense.

Step S102: and analyzing the real-time traffic flow video, and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video.

In the step S101, the real-time traffic video acquired by the camera may be analyzed by the local terminal or uploaded to the background (generally, a server with strong performance) for analysis by the background, which is not limited in the present application. According to the embodiment of the application, whether the traffic flow, the vehicle speed, the occupancy, the queuing length, the delay time and other data of multiple dimensions on the traffic lane of the entrance and the exit of the parking lot are jammed can be judged. For example, when the occupancy of the necessary lanes at the entrance/exit of the parking lot (i.e., the space occupation ratio of the vehicle on the road, reflecting the saturation of the road) is too high, the vehicle is difficult to normally pass and is easy to form congestion, so that whether the necessary lanes at the entrance/exit of the parking lot are congested can be judged from the occupancy; for another example, when the queuing length of the vehicles passing through the lane at the entrance and the exit of the parking lot is too long, the point becomes a congestion bottleneck, and the upstream road traffic is affected, so that whether the traffic passing through the lane at the entrance and the exit of the parking lot is congested can be judged from the length of the queuing length, and the like.

Step S103: if the necessary traffic lanes at the entrance and the exit of the parking lot are not congested, a common exit and entrance mode is started.

If the real-time traffic flow video is analyzed, and the traffic jam is not caused by the traffic lane passing through the entrance and the exit of the parking lot, the current vehicles entering and exiting the parking lot are not influenced, and the common entering and exiting mode can be started.

Step S104: if the traffic of the necessary traffic lane at the entrance and exit of the parking lot is blocked, the rapid entrance mode or the rapid exit mode is started so that the vehicles which go forward and exit rapidly come in and go out from the entrance and exit of the parking lot.

If the real-time traffic video is analyzed, it is determined that the traffic jam occurs in the traffic lane of the entrance/exit of the parking lot, the normal entrance/exit mode is started to increase the traffic jam degree, and at this time, the fast entrance mode or the fast exit mode should be started, for example, the fast entrance/exit mode is not sensitive, so that the vehicles in the current entrance/exit of the parking lot can rapidly enter/exit from the entrance/exit of the parking lot, thereby reducing the current traffic jam of the traffic lane of the entrance/exit of the parking lot.

Step S105: when the rapid entry mode is started, the current entry vehicle is guided to the target parking space based on the prediction of the parking space occupation condition of the parking lot.

Although the rapid entrance mode is started, the current vehicle can rapidly enter from the entrance of the parking lot, and congestion of lanes at the entrance of the parking lot is objectively relieved. However, if the vehicle enters the parking lot and turns around without destination, and cannot enter the target parking space quickly, after the entering vehicle reaches a certain number, the traffic jam of the traffic lane in the parking lot can be caused with high probability, and even the traffic jam of the traffic lane of the parking lot entrance is caused again in the reverse direction. Therefore, in the embodiment of the application, after the rapid entry mode is started, the current entry vehicle can be guided to the target parking space based on the prediction of the parking space occupation condition of the parking lot. As an embodiment of the present application, guiding the current entering vehicle to the destination parking space based on the prediction of the parking space occupation situation of the parking lot can be achieved through steps S1051 to S1054, which is described in detail as follows:

step S1051: and acquiring the occupied state data of each parking space of the parking lot in real time.

In the embodiment of the application, equipment such as a sensor, a camera and the like erected on the parking spaces of the parking lot can be used for collecting the occupied state data of each parking space in real time. Since the state of the parking space is either occupied or free, i.e. belongs to a binary logic, a logic "1" may be used to indicate that the parking space is occupied and a logic "0" to indicate that it is free. In other words, for the ith parking space, its occupancy state data at time tCan be expressed as:

Step S1052: the occupied state data of each parking space of the parking lot is input into a trained parking space occupied prediction model, and the probability of each parking space of the parking lot being occupied at a certain moment in the future is predicted by the trained parking space occupied prediction model.

In the embodiment of the application, the trained parking space occupation prediction model is obtained after training a model based on a certain algorithm (such as LSTM, GRU, etc.), and the specific training method can refer to the description of the subsequent embodiment. It should be noted that, before the occupation state data of each parking space in the parking lot is input into the trained parking space occupation prediction model, the occupation state data of each parking space can be subjected to preprocessing operations such as cleaning, denoising, standardization and the like, and converted into time series data, namely, for the ith parking space, the occupation state time seriesCan be expressed as:

Probability of being occupied at time t for the ith parking space Can be expressed as:

wherein, And representing the trained parking space occupation prediction model.

Step S1053: and recommending a parking space set formed by the parking spaces closest to the destination position of the current entering vehicle and having the smallest occupied probability to the current entering vehicle according to the occupied probability of each parking space of the parking lot, the occupied state data of each parking space and the destination position of the current entering vehicle.

When the current entering vehicle enters the parking lot, its destination location (denoted by D) such as an elevator entrance of a certain building or the like can be acquired. Then, according to the probability that each parking space of the parking lot is occupiedOccupancy status data for each parking spaceAnd a destination position D of the current entry vehicle, recommending to the current entry vehicle a set of parking spaces (denoted by R, and assuming that the set has k parking spaces in total) that are closest to the destination position of the current entry vehicle and have the smallest probability of being occupied:

any one of the parking spaces in the above set R The following requirements are satisfied:

wherein, Representing free parking spacesThe distance to the destination location D is such that,Parking space representing parking lotProbability of occupation, min () is a minimum function, representing the minimum value found for the expression in brackets. In the above embodiment, "free parking space" of "all free parking spaces in parking lot" is based on the occupancy state data of the i-th parking space, i.e., the above embodimentTo judge.

Step S1054: and sending the parking space set which is closest to the destination position of the current entering vehicle and has the smallest occupied probability to a user of the current entering vehicle so as to guide the current entering vehicle to any one of the parking spaces in the parking space set.

Specifically, the set of parking spaces, i.e., R, closest to the destination location of the current oncoming vehicle and having the smallest probability of being occupied may be transmitted to the user of the current oncoming vehicle through an application program of the smart phone or the car navigation system, so as to guide the current oncoming vehicle to any one of the set of parking spaces 。

As another embodiment of the present application, guiding the current entering vehicle to the destination parking space based on the prediction of the parking space occupation situation of the parking lot can be achieved through steps S '1051 to S'1055, which are described in detail as follows:

step S'1051: and acquiring a parking space reservation request of the current entering vehicle.

In the embodiment of the application, the parking space reservation request of the current entering vehicle is request information which is sent to a parking lot management system in advance before the current entering vehicle arrives at a parking lot, and comprises the estimated arrival time T of the current entering vehicle, the stay time L of the current entering vehicle at the parking lot, the destination position D of the parking lot and the like; the background can acquire a parking space reservation request of the current entering vehicle through a parking lot management system.

Step S'1052: and acquiring the reserved state of the virtual parking space pool from a management system of the parking lot.

In the embodiment of the application, the parking lot management system maintains a virtual parking lot pool, and records the reserved state and the expected idle time of each parking lot. Since the state of the parking space is either reserved or unreserved, i.e. belongs to a binary logic, a logic "1" may be used to indicate that the parking space is reserved and a logic "0" indicates unreserved. In other words, for the ith parking space, its reservation stateCan be expressed as:

i th parking space estimated free time Can be expressed as:

wherein t represents the current time of day, Indicating the expected parking time of the jth vehicle in the ith parking spaceThe expression of (2) indicates: if the ith parking space is reserved by the jth vehicle at the current time t and the expected parking time of the jth vehicle isThen the ith parking space is predicted to be inIdle at the moment; if the ith parking space is not reserved at the current time t, the ith parking space is expected to be free at the current time t. For example, assume that the current time t is 10 am: 00, the ith parking space has been reserved by the jth vehicle, which is expected to park for 2 hours (i.e.=2 Hours), the estimated free time of the ith parking spaceIs 10 am: 00 +2 hours = 12 pm: 00. on the other hand, if the i-th parking space is 10:00 is not reserved, it predicts an idle timeIt is 10 am: 00.

Step S'1053: and determining a candidate parking space set according to the parking space reservation request of the current entering vehicle and the reservation state of the virtual parking space pool.

Specifically, according to the parking space reservation request of the current entering vehicle and the reserved state of the virtual parking space pool, k parking spaces with the expected idle time earlier than the expected arrival time of the current entering vehicle and closest to the destination position are taken as a candidate parking space set C:

for any candidate parking space in the candidate parking space set C The following conditions must be satisfied:

wherein, Representing candidate parking spacesIs used to determine the reservation state of the (c),Representing candidate parking spacesThe estimated probability of idleness at the estimated arrival time T of the current oncoming vehicle,Is a preset probability threshold value, and the probability threshold value is set,Representing free parking spacesDistance to destination D, andShould also satisfyAndMin () is a minimum function, representing the minimum value of the expression in brackets. The above conditions indicate that: for each candidate parking spaceIts estimated probability of idleness at the estimated arrival time T of the current oncoming vehicleShould be greater than or equal to a preset probability thresholdCan be taken into consideration of the allocation of the parking spaces. Therefore, the higher availability of the parking spaces allocated to the vehicles can be ensured when the vehicles arrive, and the risk of parking space reservation failure is reduced. The preset probability threshold value is setThe value of (2) can be adjusted according to the actual condition and management requirement of the parking lot. The higher the threshold, the more stringent the screening of candidate parking spaces, the higher the reservation success rate, but the number of available parking spaces may be reduced; the lower the threshold, the wider the candidate parking space selection range, the greater the number of available parking spaces, but the risk of reservation failure may also increase. Thus, the probability threshold is presetThe value of (c) requires a balance between availability and risk.

Step S'1054: and selecting a target parking space from the candidate parking space set, wherein the target parking space is a parking space with the smallest occupied probability at the expected arrival time of the current entrance vehicle, which is output by the trained parking space occupation prediction model.

When a target parking space is selected from the candidate parking space set, and the target parking space is the parking space with the smallest occupied probability at the predicted arrival time of the current oncoming vehicle output by the trained parking space occupation prediction model, which means that the selected parking space is certainly one parking space in the candidate parking space set C obtained in the foregoing embodiment, and secondly, the selected parking space also needs to satisfy the condition that the occupied probability at the predicted arrival time of the current oncoming vehicle output by the trained parking space occupation prediction model is smallest. If the target parking space is usedRepresentation, then:

wherein, From the above-mentioned trained parking space occupation prediction modelObtained, i.e ，As a mathematical function, means that the finding causes

Candidate parking space for obtaining minimum value。

Step S'1055: and sending the target parking space to a user of the current entering vehicle so as to guide the current entering vehicle to the target parking space.

After the destination parking space is obtained, the destination parking space is transmitted to a user of the current entering vehicle, for example, an App which is transmitted to the user of the current entering vehicle through a network by using a parking lot management system, and is also transmitted together with the position, the reserved time period and the like of the destination parking space, and meanwhile, the reserved state and the expected free time of the corresponding parking space in the virtual parking space pool are updated by the parking lot management system.

As described above, the trained parking space occupation prediction model in the foregoing embodiment is obtained by training the parking space occupation prediction model, and specifically, the training method thereof includes steps S1 to S4:

step S1: and collecting parking space occupation historical data of the parking lot.

In the embodiment of the application, the collected historical parking space occupation data of the parking lot comprises information such as occupation state, time stamp, vehicle type, stay time length and the like of each parking space; these collected data may be divided into training sets, validation sets, and test sets. Further, the parking space occupation history data may be subjected to cleaning, denoising, normalization, and the like, and specifically, the parking space occupation state may be converted into a binary variable (i.e., 0 indicates idle, 1 indicates occupied), the timestamp may be converted into a cyclic characteristic (e.g., hours, weeks, etc.), the continuous variable (e.g., stay time period) may be subjected to normalization, and the like.

Step S2: and extracting key features of the parking space occupation from the parking space occupation historical data to obtain a parking space occupation feature vector.

The key features of parking space occupation, that is, key features affecting parking space occupation, such as time features (hours, weeks, etc.), parking space position features (distance from an entrance, floor, etc.), vehicle features (vehicle type, brand, etc.), environmental features (weather, temperature, etc.), etc. are extracted. And coding the extracted key features, such as single-heat coding, embedding and the like, to obtain the parking space occupation feature vector.

Step S3: the parking space occupation feature vector is input into the following parking space occupation prediction model for training:

wherein, Is a Logistic function (its prototype is) Y represents the probability of occupied parking space, x represents the feature vector of occupied parking space input to the parking space occupation prediction model,And b represents a weight vector and a bias term, respectively.

Step S4: after each training wheel is finished, evaluating the performance of the parking space occupation prediction model, and updating the weight vector of the parking space occupation prediction model according to the following formula until the parking space occupation prediction model converges or the performance reaches the expectation:

wherein, In order to update the weight vector before it is updated,In order for the weight vector to be updated,In order for the rate of learning to be high,And representing the gradient of the loss function of the parking space occupation prediction model to the weight.

In the weight vector updating formula of the parking space occupation prediction model, the learning rate is calculatedIs an important super-parameter of the machine learning model. The super-parameters are parameters set before model training, and need to be adjusted empirically or experimentally to obtain the best model performance. In addition to learning rate, other common super-parameters include number of hidden layers, number of hidden units, regularization coefficient, dropout rate and batch size, etc. The learning rate determines the weight updating amplitude in each iteration, a higher learning rate can accelerate the convergence speed, but the optimization process is possibly unstable and is difficult to converge to an optimal solution, and a lower learning rate can ensure stable convergence and the training time is longer. In the embodiment of the application, the learning rate of the weight vector update formula of the parking space occupation prediction model is calculatedThe selection may be by an attenuation scheme, which may be a piecewise constant attenuation or an exponential attenuation, or an adaptive tuning scheme, as described in more detail below.

Piecewise constant decay learning rateI.e. the training process is divided into several phases, each phase using a different learning rate, in other words the learning rate remains unchanged within each phase, but decays by a certain factor when crossing the phase, the specific process comprising:

1. Determining an attenuation phase and an attenuation factor (decay factor), e.g., attenuation once every 50 epochs, the attenuation factor being 0.1;

2. the learning rate remains unchanged during each phase, namely:

wherein, For the learning rate of the t-th epoch (round),For the initial rate of learning to be the same,

For the initial epoch of the ith stage,Is an attenuation factor.

Exponential decay learning rateThe learning rate decays with time according to the form of an exponential function, the decay rate is controlled by super parameters, and the specific process comprises the following steps: determining initial learning rateAttenuation Rate(Usually 0.9 to 0.999); at each epoch, the learning rate is updated as follows:

wherein, For the learning rate of the t-th epoch,For the initial rate of learning to be the same,Is the decay rate.

The adaptive adjustment scheme of the learning rate refers to a method for automatically adjusting the learning rate of each parameter according to gradient information in a model training process, so as to accelerate convergence and improve training stability, and specifically, the adaptive adjustment scheme of the learning rate includes the following steps S1 and S2:

step S1: first moment estimation of initialization parameters Second moment estimationTime step t=0;

step S2: in each iteration, for each parameter The following procedure was performed:

calculating gradients Wherein, the method comprises the steps of, wherein,As a loss function;

updating the first moment estimate:

Updating the second moment estimate:

correcting the first moment estimation:

Correcting the second moment estimation:

Updating parameters:

wherein, AndFor the decay rate (usually taking=0.9，=0.999 ），For an initial learning rate (typically 0.001),To smooth items (usually takingE is a natural base), t is a time step,Representation pairThe square root is taken.

The self-adaptive adjustment scheme of the learning rate adaptively sets different learning rates for each parameter, the parameter learning rate with frequent updating is smaller, the parameter learning rate with infrequent updating is larger, the convergence can be accelerated, the vibration can be reduced, and meanwhile, the self-adaptive adjustment scheme of the learning rate is insensitive to the initial value of the learning rate and has stronger robustness.

Step S106: when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm.

Although the quick departure mode is turned on, a vehicle to be departed (i.e., a vehicle to be driven from the parking lot exit) can be quickly departed from the parking lot exit, objectively reducing congestion of the parking lot exit lane. However, if the vehicles to be discharged go around without destination and cannot reach the exit of the parking lot quickly, after the number of vehicles to be discharged reaches a certain number, the traffic jam of the traffic lane in the parking lot is caused with high probability, and even the traffic jam of the traffic lane of the exit of the parking lot is caused again in the reverse direction. Thus, in an embodiment of the present application, after the fast departure mode is turned on, the corresponding departure of the vehicle to be departed to the parking lot is guided based on the dynamic departure lane allocation scheme or the departure path guidance planning algorithm. As an embodiment of the present application, guiding the corresponding exit driving-off of the vehicle to be discharged to the parking lot based on the dynamic exit lane allocation scheme may be achieved through steps S1061 to S1064, as follows:

step S1061: the time of arrival at the exit of the parking lot is estimated based on the state of each vehicle in the parking lot.

Specifically, the time for each vehicle to reach the exit of the parking lot can be estimated by tracking and locating the state information such as the position, the speed and the like of each vehicle in the parking lot in real time through a vehicle positioning sensor, a vehicle speed measuring sensor and the likeThe following are provided:

wherein, Indicating the current time of day and,Indicating the distance of the vehicle to the exit of the parking lot,Indicating the average running speed of the vehicle.

Step S1062: and predicting the vehicle departure demand of each exit lane of the parking lot in a future period according to the time of each vehicle reaching the exit of the parking lot.

When the time for each vehicle to reach the exit of the parking lot is obtained, the vehicle departure demand of each exit lane of the parking lot in a future period of time can be predicted according to the time for each vehicle to reach the exit of the parking lot and comprehensively considering the data such as the lane occupation state, the time period, the weather condition and the likeThe predictive model may be expressed as:

wherein, Feature vector representing vehicle departure demand affecting ith exit lane of parking lot, including time of arrival of each vehicle at parking lot exitLane occupancy status, time period, weather conditions, etc.

Step S1063: the exit lanes are dynamically allocated according to the predicted vehicle departure demand of each exit lane and the arrival time of each vehicle at the exit of the parking lot, so that the waiting time of each exit lane is minimized.

Specifically, the exit lanes are dynamically allocated according to the predicted vehicle departure demand for each exit lane and the time each vehicle arrives at the exit of the parking lot, such that minimizing the waiting time for each exit lane can be achieved based on the following minimum cost maximum flow model, wherein:

the lane allocation objective function of the model is as follows:

The constraint conditions are as follows:

wherein, In order to be a function of the sum,Indicating the number of vehicles allocated to the i-th exit lane,Is a trade-off factor balancing maximum flow and minimum cost,A weight (e.g., degree of congestion) representing the i-th exit lane, which may be based on exit lane length, exit lane vehicle departure demandAnd other relevant factors and the like,Indicating the number of vehicles entering the parking lot,Indicating the number of vehicles driving off the parking lot,Indicating the upper limit of the capacity of the i-th exit lane,Indicating the total departure demand of the parking lot.

Step S1064: based on the dynamic allocation result of the exit lanes, the number of waiting vehicles and the estimated waiting time of each exit lane are issued to the user of the vehicle to be discharged in real time so as to guide the vehicle to be discharged to the corresponding exit of the parking lot to drive away.

Specifically, according to the dynamic allocation result of the exit lanes, information such as the number of waiting vehicles and estimated waiting time of each exit lane can be issued to users of the vehicles to be discharged in real time through an electronic display screen, a mobile phone application program and the like, and the vehicles to be discharged are guided to drive away from corresponding exits of the parking lot. Furthermore, an elastic lane management mechanism can be introduced, namely, in the peak period of the exit of the parking lot, part of the entrance lanes are dynamically adjusted to be exit lanes, so that the exit traffic capacity is increased; and in the off-valley period of the parking lot, the elastic lanes are recovered to be entrance lanes, and the lane function switching can be realized through measures such as variable traffic signs, signal lamps and different color delay switching thereof.

In the embodiment of the present application, the above-mentioned scheme may be appropriately adjusted based on some groups or priorities of vehicles, and in particular, the above-mentioned method for guiding the vehicles to be discharged to the corresponding exit of the parking lot based on the dynamic exit lane allocation scheme may further include steps Sa1061 to Sa1063, which are described in detail as follows:

Step Sa1061: the priority weight of the vehicle to be outbound is determined based on the priority of the vehicle to be outbound.

Specifically, the priority level may be defined in advance according to the type of vehicle and the special requirement, for example, a general vehicle defines its priority=1; disabled vehicles define their priority = 2; emergency rescue vehicles define their priority = 3, etc.; the higher the priority value, the higher the priority release right of the vehicle; then, for the ith vehicle, a priority weight is introduced：

Wherein, A priority value representing the ith vehicle,Is a positive scaling factor for adjusting priority weightsIs a function of the degree of influence of (a). By adjusting the scaling factorCan flexibly control the priority weightIs a function of the degree of influence of (a). Larger sizeThe value will amplify the priority difference, making the high priority vehicle more prominent; smaller and smallerThe value reduces the priority difference, so that the distribution of the exit lanes is more balanced; can select proper according to the actual demand and the management strategyValues.

Step Sa1062: and applying the priority weight of the outgoing vehicle to the lane allocation objective function of the dynamic allocation outgoing lane scheme to obtain a modified lane allocation objective function.

Specifically, the modified lane allocation objective function is as follows:

wherein, Indicating the expected exit passage length of the ith vehicle using the assigned lane,In order to be a function of the sum,Indicating the number of vehicles allocated to the i-th exit lane,AndRespectively a function of the maximum and minimum values,Is the priority weight described above. The modified lane allocation objective function described above shows that: by dividing by priority weightsVehicles with higher priorities are given a smaller contribution value in the objective function and are therefore preferentially allocated to a better exit lane.

Step Sa1063: and constructing a lane allocation optimization problem by using the modified lane allocation objective function and solving the lane allocation optimization problem to obtain an exit lane allocation scheme containing vehicle priority factors.

And constructing a lane allocation optimization problem by using the modified lane allocation objective function and combining other constraint conditions (such as lane capacity limit and the like), and then solving the lane allocation optimization problem by using integer programming, heuristic algorithm and other optimization solving technologies to obtain an exit lane allocation scheme containing vehicle priority factors.

As an embodiment of the present application, guiding the corresponding exit drive-off of the vehicle to be discharged to the parking lot based on the departure path guidance planning algorithm may be achieved through steps S '1061 to S'1063, which are described in detail as follows:

Step S'1061: and collecting real-time traffic state data of an exit lane of the parking lot.

Specifically, the real-time traffic state data of the exit lane of the parking lot can be collected in real time through equipment such as a parking space occupation sensor, a camera and the like. For the ith parking lot exit lane, real-time traffic state data may be used for average speedFlow rate and flow rateAnd the like.

Step S'1062: and inputting the real-time traffic state data of the parking lot exit lane into a trained parking lot road network congestion prediction model to predict the congestion degree of the parking lot exit lane.

Similar to the aforementioned trained parking space occupation prediction model, in the embodiment of the present application, the trained parking space network congestion prediction model is obtained by training the parking space network congestion prediction model using a machine learning algorithm, such as Support Vector Regression (SVR), long-term memory network (LSTM), and the like. The data used for training the parking lot road network congestion prediction model can be historical traffic state data or real-time traffic state data. In the embodiment of the application, the congestion degree of the ith exit lane of the parking lot can be expressed as follows by using a journey time index:

wherein, Indicating a predicted travel time of an i-th exit lane of the parking lot,And (5) representing the travel time of the ith exit lane of the parking lot in the free flow state.

Step S'1063: and generating real-time departure path guidance information according to the congestion degree of the exit lane of the parking lot, and guiding the vehicles to be departed to select a clear path to travel to the corresponding exit of the parking lot.

According to the predicted congestion degree of the ith exit lane of the parking lotReal-time departure path guidance information is generated by identifying areas with severe congestion and exit lanes in a parking lot, including path recommendation, exit information, and the like, wherein the path recommendation is based on real-time congestion conditions and prediction results, and the recommended relatively clear paths to exits, and the exit information includes real-time states of various exits of the parking lot, such as open lane numbers, predicted transit times, and the like. The real-time departure path guidance information can be issued to users of vehicles to be departed through channels such as electronic display screens, application programs of smart phones and the like so as to guide the vehicles to be departed to travel to corresponding exits of the parking lot by selecting a clear path. In guiding, for each vehicle to be discharged, the real-time traffic state data of the parking lot exit lane and the predicted congestion degree of the parking lot exit lane can be utilized in combination with the current position, the expected departure time and other factorsAn optimal path from the current location to the parking lot exit is planned. The path planning may be based on a shortest path algorithm, such as Dijkstra's algorithm or a-x algorithm, and the objective function may be expressed as:

wherein, In order to be a function of the sum,In order to find the function of the minimum value,Indicating the length of the ith exit lane of the parking lot.

As can be seen from the above-mentioned parking lot vehicle traffic control method illustrated in fig. 1, if it is determined that congestion occurs in the necessary lanes at the entrance and exit of the parking lot, the fast entrance mode or the fast exit mode is started; further, after the rapid entry mode is started, guiding the current entry vehicle to the target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm. On the one hand, when judging that the traffic lane at the entrance and exit of the parking lot is jammed, the rapid entrance mode or the rapid exit mode is started, so that the current vehicle can rapidly enter and exit the entrance and exit of the parking lot; on the other hand, based on the prediction of the parking space occupation condition of the parking lot, the current incoming vehicles are guided to the target parking space, so that long-time stay of the incoming vehicles in the parking lot is effectively avoided, traffic jam of lanes in the parking lot can be relieved objectively, and based on a dynamic exit lane distribution scheme or an exit path guidance planning algorithm, corresponding exit of the vehicles to be exited to the parking lot is guided, long-time stay of the exit vehicles in the parking lot is effectively avoided, and traffic jam of exit lanes of the parking lot can be relieved. In conclusion, the technical scheme of the application effectively improves the vehicle passing efficiency of the parking lot.

Referring to fig. 2, the parking lot vehicle traffic control device provided in the embodiment of the present application may include an acquisition module 201, a determination module 202, a first opening module 203, a second opening module 204, a first guiding module 205, and a second guiding module 206, which are described in detail as follows:

the acquisition module 201 is used for acquiring real-time traffic videos entering and exiting the parking lot through a camera;

the judging module 202 is used for analyzing the real-time traffic flow video and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video;

the first opening module 203 is configured to open a normal entrance/exit mode if congestion does not occur in the necessary lanes at the entrance/exit of the parking lot;

The second opening module 204 is configured to open a fast entrance mode or a fast exit mode if congestion occurs in the traffic lane of the entrance/exit of the parking lot, so that the vehicles that are currently in the entrance/exit of the parking lot can quickly enter/exit from the entrance/exit of the parking lot;

the first guiding module 205 is configured to guide the current entering vehicle to the target parking space based on the prediction of the parking space occupation situation of the parking lot after the fast entering mode is started;

The second guiding module 206 is configured to guide the corresponding exit of the vehicle to be outbound to the parking lot based on the dynamic exit lane allocation scheme or the departure path guidance planning algorithm after the fast departure mode is started.

As can be seen from the above-mentioned parking lot vehicle traffic control device illustrated in fig. 2, if it is determined that congestion occurs in the necessary lanes at the entrance/exit of the parking lot, the fast entrance mode or the fast exit mode is started; further, after the rapid entry mode is started, guiding the current entry vehicle to the target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm. On the one hand, when judging that the traffic lane at the entrance and exit of the parking lot is jammed, the rapid entrance mode or the rapid exit mode is started, so that the current vehicle can rapidly enter and exit the entrance and exit of the parking lot; on the other hand, based on the prediction of the parking space occupation condition of the parking lot, the current incoming vehicles are guided to the target parking space, so that long-time stay of the incoming vehicles in the parking lot is effectively avoided, traffic jam of lanes in the parking lot can be relieved objectively, and based on a dynamic exit lane distribution scheme or an exit path guidance planning algorithm, corresponding exit of the vehicles to be exited to the parking lot is guided, long-time stay of the exit vehicles in the parking lot is effectively avoided, and traffic jam of exit lanes of the parking lot can be relieved. In conclusion, the technical scheme of the application effectively improves the vehicle passing efficiency of the parking lot.

Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 3, the electronic apparatus 3 of this embodiment mainly includes: a processor 30, a memory 31 and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a program of a parking lot vehicle traffic control method. The steps in the above-described embodiments of the parking lot vehicle passage control method, such as steps S101 to S106 shown in fig. 1, are implemented when the processor 30 executes the computer program 32. Or the processor 30 may perform the functions of the modules/units in the above-described apparatus embodiments when executing the computer program 32, for example, the functions of the acquisition module 201, the determination module 202, the first opening module 203, the second opening module 204, the first guiding module 205, and the second guiding module 206 shown in fig. 2.

Illustratively, the computer program 32 of the parking lot vehicle traffic control method basically includes: collecting real-time traffic flow videos entering and exiting a parking lot through a camera; analyzing the real-time traffic flow video, and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video; if the necessary lanes at the entrance and the exit of the parking lot are not congested, a common entrance and exit mode is started; if the necessary lanes at the entrance and the exit of the parking lot are congested, a rapid entrance mode or a rapid exit mode is started so that vehicles which enter and exit from the entrance and the exit of the parking lot rapidly enter and exit from the entrance and the exit of the parking lot; after the rapid entry mode is started, guiding the current entry vehicle to a target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm. The computer program 32 may be divided into one or more modules/units, which are stored in the memory 31 and executed by the processor 30 to complete the present application. One or more of the modules/units may be a series of computer program instruction segments capable of performing a specific function for describing the execution of the computer program 32 in the electronic device 3. For example, the computer program 32 may be divided into functions of the acquisition module 201, the judgment module 202, the first opening module 203, the second opening module 204, the first guidance module 205, and the second guidance module 206 (modules in the virtual device), and specific functions of each module are as follows: the acquisition module 201 is used for acquiring real-time traffic videos entering and exiting the parking lot through a camera; the judging module 202 is used for analyzing the real-time traffic flow video and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video; the first opening module 203 is configured to open a normal entrance/exit mode if congestion does not occur in the necessary lanes at the entrance/exit of the parking lot; the second opening module 204 is configured to open a fast entrance mode or a fast exit mode if congestion occurs in the traffic lane of the entrance/exit of the parking lot, so that the vehicles that are currently in the entrance/exit of the parking lot can quickly enter/exit from the entrance/exit of the parking lot; the first guiding module 205 is configured to guide the current entering vehicle to the target parking space based on the prediction of the parking space occupation situation of the parking lot after the fast entering mode is started; the second guiding module 206 is configured to guide the corresponding exit of the vehicle to be outbound to the parking lot based on the dynamic exit lane allocation scheme or the departure path guidance planning algorithm after the fast departure mode is started.

The electronic device 3 may include, but is not limited to, a processor 30, a memory 31. It will be appreciated by those skilled in the art that fig. 3 is merely an example of the electronic device 3 and is not meant to be limiting of the electronic device 3, and may include more or fewer components than shown, or may combine certain components, or different components, e.g., a computing electronic device may also include an input-output device, a network access device, a bus, etc.

The Processor 30 may be a central processing unit (Central Processing Unit, CPU), other general purpose Processor, digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), off-the-shelf Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 31 may be an internal storage unit of the electronic device 3, such as a hard disk or a memory of the electronic device 3. The memory 31 may also be an external storage electronic device of the electronic device 3, such as a plug-in hard disk provided on the electronic device 3, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), or the like. Further, the memory 31 may also include both an internal storage unit of the electronic device 3 and an external storage electronic device. The memory 31 is used to store computer programs and other programs and data required by the electronic device. The memory 31 may also be used to temporarily store data that has been output or is to be output.

It will be apparent to those skilled in the art that the above-described functional units and modules are merely illustrated for convenience and brevity of description, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above device may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of modules or units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another apparatus, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated modules/units may be stored in a storage medium if implemented in the form of software functional units and sold or used as stand-alone products. Based on such understanding, the present application may implement all or part of the procedures in the methods of the above embodiments, or may be implemented by instructing related hardware by a computer program, where the computer program of the parking lot vehicle traffic control method may be stored in a storage medium, and the computer program, when executed by a processor, may implement the steps of each method embodiment described above, that is, collecting real-time traffic video from and to the parking lot by using a camera; analyzing the real-time traffic flow video, and judging whether the necessary lanes at the entrance and the exit of the parking lot are congested according to the analysis result of the real-time traffic flow video; if the necessary lanes at the entrance and the exit of the parking lot are not congested, a common entrance and exit mode is started; if the necessary lanes at the entrance and the exit of the parking lot are congested, a rapid entrance mode or a rapid exit mode is started so that vehicles which enter and exit from the entrance and the exit of the parking lot rapidly enter and exit from the entrance and the exit of the parking lot; after the rapid entry mode is started, guiding the current entry vehicle to a target parking space based on the prediction of the parking space occupation condition of the parking lot; when the fast departure mode is started, guiding the vehicles to be departed to the corresponding exits of the parking lot based on a dynamic exit lane allocation scheme or an exit path guidance planning algorithm. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, executable files or in some intermediate form, etc. The storage medium may include: any entity or device capable of carrying computer program code, a recording medium, a USB flash disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the content of the storage medium may be appropriately increased or decreased according to the requirements of jurisdictions in which the legislation and the patent practice, such as in some jurisdictions, the storage medium does not include electrical carrier signals and telecommunication signals according to the legislation and the patent practice.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.

Claims

1. A vehicle traffic control method for a parking lot, the method comprising:

collecting real-time traffic videos entering and exiting a parking lot through a camera, wherein the real-time traffic videos comprise the number of vehicles, the running speed and the vehicle density of lanes at the entrance and the exit of the parking lot;

Analyzing the real-time traffic flow video, judging whether the necessary lanes of the parking lot access are congested according to the analysis result of the real-time traffic flow video, wherein the judging whether the necessary lanes of the parking lot access are congested comprises judging whether the necessary lanes of the parking lot access are congested from the traffic flow, the speed, the occupancy, the queuing length and the delay time on the necessary lanes of the parking lot access;

When the fast entry mode is started, guiding the current entry vehicle to a target parking space based on the prediction of the parking space occupation condition of the parking lot, and guiding the current entry vehicle to the target parking space based on the prediction of the parking space occupation condition of the parking lot, wherein the fast entry mode comprises the following steps: acquiring a parking space reservation request of the current entering vehicle; acquiring a reserved state of a virtual parking space pool from a management system of the parking lot; determining a candidate parking space set according to the parking space reservation request of the current entering vehicle and the reservation state of the virtual parking space pool; selecting the target parking space from the candidate parking space set, wherein the target parking space is a parking space with the smallest occupied probability at the expected arrival time of the current entering vehicle, which is output by a trained parking space occupation prediction model; transmitting the target parking space to a user of the current entering vehicle so as to guide the current entering vehicle to the target parking space;

When the rapid departure mode is started, guiding a corresponding exit of a vehicle to be departed to the parking lot to leave based on a dynamic exit lane allocation scheme or a departure path guidance planning algorithm;

the determining the candidate parking space set according to the parking space reservation request of the current entering vehicle and the reserved state of the virtual parking space pool comprises the following steps: according to the parking space reservation request of the current entering vehicle and the reservation state of the virtual parking space pool, k parking spaces with the expected idle time earlier than the expected arrival time of the current entering vehicle and closest to the target position are used as a candidate parking space set C:

wherein, Representing candidate parking spacesIs used to determine the reservation state of the (c),Representing candidate parking spacesThe estimated probability of idleness at the estimated arrival time T of the current oncoming vehicle,Is a preset probability threshold value, and the probability threshold value is set,Representing free parking spacesDistance to destination D, andShould also satisfyAndMin () is a minimum function, representing the minimum value of the expression in brackets,Representing the representation candidate parking spaceThere is no reservation of the reservation,

Representing the candidate parking spaceThe distance from the destination position D is the shortest distance from all the idle parking spaces in the candidate parking space set C to the destination position D.

2. The vehicle traffic control method for the parking lot according to claim 1, wherein the trained parking space occupation prediction model is trained by:

Collecting parking space occupation historical data of the parking lot;

extracting key features of parking space occupation from the parking space occupation history data to obtain parking space occupation feature vectors;

Inputting the parking space occupation feature vector into the following parking space occupation prediction model for training:

the y represents the occupied probability of the parking space, the x represents the feature vector of the occupied parking space input to the parking space occupation prediction model, and the And b represents a weight vector and a bias term, respectively;

After each wheel training is finished, evaluating the performance of the parking space occupation prediction model, and updating the weight vector of the parking space occupation prediction model according to the following formula until the parking space occupation prediction model converges or the performance reaches the expectation:

The said For the weight vector before updating, theFor the updated weight vector, theFor learning rate, theAnd representing the gradient of the loss function of the parking space occupation prediction model to the weight.

3. The parking lot vehicle traffic control method according to claim 1, wherein the guiding the corresponding exit drive-off of the vehicle to be discharged to the parking lot based on the dynamic exit lane allocation scheme includes:

Estimating the time of each vehicle reaching the exit of the parking lot according to the state of each vehicle of the parking lot;

Predicting the vehicle departure demand of each exit lane of the parking lot in a future period according to the time of each vehicle reaching the exit of the parking lot;

dynamically distributing the exit lanes according to the predicted vehicle departure demand of each exit lane and the time of each vehicle reaching the exit of the parking lot, so that the waiting time of each exit lane is minimized;

And based on the dynamic allocation result of the exit lanes, issuing the number of waiting vehicles and the estimated waiting time of each exit lane to the user of the vehicle to be discharged in real time so as to guide the vehicle to be discharged to the corresponding exit of the parking lot to drive away.

4. A parking lot vehicle traffic control method as recited in claim 3, wherein said method further comprises:

Determining the priority weight of the vehicles to be discharged according to the priority of the vehicles to be discharged;

Applying the priority weight of the vehicles to be discharged to a lane allocation objective function of a dynamic allocation outlet lane scheme to obtain a modified lane allocation objective function;

And constructing a lane allocation optimization problem by using the modified lane allocation objective function, solving the lane allocation optimization problem, and obtaining an exit lane allocation scheme containing vehicle priority factors.

5. The parking lot vehicle traffic control method according to claim 1, wherein the guiding the corresponding exit drive-off of the vehicle to be discharged to the parking lot based on the discharge path guidance planning algorithm includes:

Collecting real-time traffic state data of an exit lane of the parking lot;

Inputting the real-time traffic state data of the parking lot exit lane into a trained parking lot road network congestion prediction model, and predicting the congestion degree of the parking lot exit lane;

And generating real-time departure path induction information according to the congestion degree of the exit lane of the parking lot, and guiding the vehicles to be departed to select a clear path to travel to the corresponding exit of the parking lot.

6. A parking lot vehicle passage control apparatus, characterized by comprising:

The acquisition module is used for acquiring real-time traffic flow videos entering and exiting the parking lot through the camera, the real-time traffic video comprises the number of vehicles in the entrance lane and the exit lane of the parking lot, the running speed and the vehicle density;

The judging module is used for analyzing the real-time traffic flow video, judging whether the necessary lanes of the parking lot access are congested according to the analysis result of the real-time traffic flow video, wherein the judging whether the necessary lanes of the parking lot access are congested comprises judging whether the necessary lanes of the parking lot access are congested according to the traffic flow, the speed, the occupancy, the queuing length and the delay time of the necessary lanes of the parking lot access;

The first guiding module is configured to guide a current entering vehicle to a target parking space based on a prediction of a parking space occupation situation of the parking lot after the fast entering mode is started, and guide the current entering vehicle to the target parking space based on the prediction of the parking space occupation situation of the parking lot, including: acquiring a parking space reservation request of the current entering vehicle; acquiring a reserved state of a virtual parking space pool from a management system of the parking lot; determining a candidate parking space set according to the parking space reservation request of the current entering vehicle and the reservation state of the virtual parking space pool; selecting the target parking space from the candidate parking space set, wherein the target parking space is a parking space with the smallest occupied probability at the expected arrival time of the current entering vehicle, which is output by a trained parking space occupation prediction model; transmitting the target parking space to a user of the current entering vehicle so as to guide the current entering vehicle to the target parking space;

The second guiding module is used for guiding the vehicles to be discharged to the corresponding outlet of the parking lot to be discharged based on a dynamic outlet lane allocation scheme or a departure path guidance planning algorithm after the rapid departure mode is started;

7. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 5 when the computer program is executed by the processor.

8. A storage medium storing a computer program which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.