CN115691165A

CN115691165A - Traffic signal lamp scheduling method, device and equipment and readable storage medium

Info

Publication number: CN115691165A
Application number: CN202211194018.0A
Authority: CN
Inventors: 周江涵; 顾龙; 王宗晖; 王广善
Original assignee: Beijing Dongtu Tuoming Technology Co ltd
Current assignee: Beijing Dongtu Tuoming Technology Co ltd
Priority date: 2022-09-28
Filing date: 2022-09-28
Publication date: 2023-02-03

Abstract

The application relates to the technical field of traffic flow prediction, in particular to the technical field of traffic signal lamp scheduling. The application discloses a traffic signal lamp scheduling method, a traffic signal lamp scheduling device, traffic signal lamp scheduling equipment and a readable storage medium, wherein the traffic signal lamp scheduling method comprises the following steps: acquiring traffic data of a target intersection; the traffic data comprises vehicle flow and at least one of time interval, intersection type, phase, weather information, vehicle flow in the previous time interval, holiday information, season and pedestrian flow; inputting the traffic data into a traffic prediction model to obtain predicted traffic flow; obtaining a traffic signal lamp scheduling strategy according to the predicted traffic flow and the traffic data; and scheduling the traffic signal lamp based on the traffic signal lamp scheduling strategy. The scheme provided by the invention can predict the traffic flow in advance, dynamically schedule the traffic signal lamp, realize self-adaptive adjustment, generate an intelligent scheduling strategy and improve the traffic efficiency.

Description

Traffic signal lamp scheduling method, device and equipment and readable storage medium

Technical Field

The invention relates to the technical field of traffic flow prediction, in particular to a traffic signal lamp scheduling method, a traffic signal lamp scheduling device, traffic signal lamp scheduling equipment and a readable storage medium.

Background

In an urban traffic network, road sections and intersections always encounter the problem of large traffic flow, and traffic light scheduling is one of effective methods for relieving traffic congestion. At present, most signalized intersections adopt a method for controlling signal lamps in a preset period or a fixed time interval, and the method is used for controlling by presetting a timing scheme according to the traffic requirements observed in the past. However, in the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art:

the above method relies on the periodicity that vehicle traffic has, so it is easier to summarize and predict in long period situations. In the prior art, under the sudden conditions of traffic accidents, special events and the like, the rapid change of the traffic flow is difficult to be rapidly coped with, the traffic flow cannot be predicted in advance, particularly, the delay degree of vehicles at the intersection is increased by adopting a fixed timing strategy, the problem of road congestion brings great inconvenience to the working and life of citizens, and the ordered development of traffic is seriously influenced.

Disclosure of Invention

In view of the above problems in the prior art, the present application provides a traffic signal light scheduling method, apparatus, device and readable storage medium, which can predict the traffic flow in advance and provide a more accurate and efficient signal light scheduling strategy.

In order to achieve the above object, a first aspect of the present application provides a traffic signal light scheduling method, including:

acquiring first data; the first data is related to the traffic flow of the target intersection;

inputting the first data into a flow prediction model to obtain a predicted traffic flow;

obtaining a traffic signal lamp scheduling strategy according to the predicted traffic flow and the working condition parameters of the target intersection;

and scheduling the traffic signal lamp based on the traffic signal lamp scheduling strategy.

As a possible implementation manner of the first aspect, the first data includes at least one of the following parameters:

time period, intersection type, signal lamp phase, traffic flow of at least one previous time period, date type and weather type.

As a possible implementation manner of the first aspect, the traffic prediction model is a classification model, and includes:

the system comprises a convolutional neural network layer, a long-term and short-term memory network layer and a self-attention layer which are sequentially cascaded.

As a possible implementation manner of the first aspect, the flow prediction model is a classification model, and when being trained, the classification model includes:

acquiring sample data, wherein an input label of the sample data comprises the first data, and an output label of the sample data comprises a traffic flow corresponding to the first data;

and training the flow prediction model by taking the input label of the sample data as the input of the flow prediction model and taking the output label as the expected input and output of the flow prediction model.

As a possible implementation manner of the first aspect, the first data includes at least two parameters, and the training of the flow prediction model further includes:

calculating an importance coefficient of each parameter of the first data;

and coupling the importance coefficients of the parameters with the parameters to serve as the input of the flow prediction model, and using the output labels as the expected input and output of the flow prediction model, and training the flow prediction model.

As a possible implementation manner of the first aspect, new sample data is acquired after a preset time interval;

and training and updating the flow prediction model based on the new sample data.

As a possible implementation manner of the first aspect, obtaining the predicted traffic flow further includes: and carrying out rule constraint on the predicted traffic flow according to a data protocol strategy.

As a possible implementation manner of the first aspect, obtaining a traffic signal lamp scheduling policy according to the predicted traffic flow and the working condition parameter of the target intersection includes:

and adjusting the combination sequence of the traffic signal lights and the time length of the signal lights according to the length and the width of the intersection, the passing time length and the waiting time length of the vehicle under different traffic flow conditions.

The second aspect of the present application provides a traffic signal light dispatching device, including:

the data acquisition module is used for acquiring first data; the first data is related to the traffic flow of the target intersection;

the edge calculation module is used for inputting the first data into a flow prediction model to obtain predicted traffic flow;

the edge calculation module is also used for obtaining a traffic signal lamp scheduling strategy according to the predicted traffic flow and the working condition parameters of the target intersection;

and the traffic signal scheduling module is used for scheduling the traffic signal lamp based on the traffic signal lamp scheduling strategy.

A third aspect of the present application provides a computing device comprising:

a processor, and an interface circuit, wherein the processor accesses a memory through the interface circuit, the memory storing program instructions that, when executed by the processor, cause the processor to perform the traffic signal light scheduling method as described above.

The traffic signal lamp scheduling scheme can predict the vehicle flow of a target time period according to the flow prediction model, meanwhile, other factors influencing the traffic flow are comprehensively considered, an intelligent scheduling strategy is generated, the traffic signal lamp is scheduled and controlled according to the scheduling strategy, the traffic signal lamp can be dynamically scheduled accurately and effectively, the problem of crossing congestion is effectively solved, and the traffic efficiency is improved.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

Drawings

The various features and the connections between the various features of the present invention are further described below with reference to the attached figures. The figures are exemplary, some features are not shown to scale, and some of the figures may omit features that are conventional in the art to which the application relates and are not essential to the application, or show additional features that are not essential to the application, and the combination of features shown in the figures is not intended to limit the application. In addition, the same reference numerals are used throughout the specification to designate the same components. The specific drawings are illustrated as follows:

fig. 1 is a flowchart of a traffic signal light scheduling method provided in an embodiment of the present application;

FIG. 2 is a diagram of an example of intersection vehicle traffic statistics provided by an embodiment of the present application;

FIG. 3 is a graph illustrating exemplary significance of feature data provided by an embodiment of the present application;

FIG. 4 is a schematic diagram of traffic signal light scheduling provided by an embodiment of the present application;

fig. 5 is a schematic structural diagram of a traffic signal light dispatching device according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a computing device 300 provided in an embodiment of the present application.

Detailed Description

The technical solution provided by the present application is further described below by referring to the drawings and the embodiments. It should be understood that the system structure and the service scenario provided in the embodiments of the present application are mainly for illustrating possible implementation manners of the technical solutions of the present application, and should not be construed as the only limitations on the technical solutions of the present application. As can be known to those skilled in the art, with the evolution of the system structure and the appearance of new service scenarios, the technical solution provided in the present application is also applicable to similar technical problems.

It should be understood that the traffic signal lamp scheduling method, the traffic signal lamp scheduling device, the traffic signal lamp scheduling equipment, the readable storage medium and the like are provided in the embodiments of the present application. Since the principles of solving the problems of these solutions are the same or similar, some of the repeated parts may not be described again in the following description of the embodiments, but it should be understood that these embodiments have been referred to and combined with each other.

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. In the case of inconsistency, the meaning described in the present specification or the meaning derived from the content described in the present specification shall control. In addition, the terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the present application.

There are three main traffic light schedules at present: (1) The fixed time scheduling method is used for setting a fixed time according to the statistics of the number of vehicles in each direction of a road junction; (2) The time-sharing fixed-time scheduling method is that on the basis of the first method, fixed time is respectively set for different time periods (such as rush hour, daytime and night) every day; (3) The traffic light dispatching method is adjusted manually or systematically according to traffic conditions in different seasons or in past days, and is relatively fixed in a future period of time. The existing traffic signal control basically adjusts the traffic signal lamp by calculating and detecting the instant traffic flow and performing a static strategy, so that the traffic flow cannot be pre-judged in advance, signal timing can be performed more accurately and efficiently, and particularly, the delay degree of vehicles at the intersection can be increased by adopting a fixed timing strategy.

The embodiment provided by the invention is mainly applied to the prediction of traffic flow and the scheduling and control of traffic lights, the traffic flow data and the characteristic data of the influence factors (influencing the flow) are input into the flow prediction model, the characteristic data of the influence factors can be introduced from the input layer of the flow prediction model, so that the characteristic data of the influence factors are introduced in the whole process of the flow prediction model, a plurality of characteristics and characteristic data including historical flow and the like are fused, and a traffic light scheduling strategy is given according to the result of the flow prediction, so that the traffic flow can be predicted in advance, an accurate and effective traffic light scheduling strategy can be given, and the emergency condition can be effectively coped with.

Referring to fig. 1 and fig. 4, fig. 1 is a flowchart of a traffic signal light scheduling method provided in an embodiment of the present application, and fig. 4 is a schematic diagram of traffic signal light scheduling provided in the embodiment of the present application based on the defects in the prior art.

As shown in fig. 1, an embodiment of the present application provides a traffic signal light scheduling method, including:

s101: acquiring first data; the first data is related to traffic flow at the target intersection.

In some examples, the first data related to intersection traffic flow includes traffic flow monitoring data of a target intersection, traffic flow data recording a preset time period and a preset time interval, such as vehicle flow data and people flow data of each hour in a day; traffic data at different times of the day, such as early peak and late peak traffic data; and traffic flow data under different weather conditions in each time period, such as traffic flow data in sunny weather of the peak in the morning and at night, traffic flow data in rainy and snowy weather of the peak in the morning and at night, and the like.

The monitoring record of the first data can adopt equipment or units such as a camera erected at the intersection, a radar arranged around the road or underground, and the like to shoot and collect traffic data and information of the intersection. The traffic flow information collected by a camera shooting video or a radar device is identified by inputting the data subjected to the operations such as preprocessing and the like into an image identification device (a flow prediction model). The system can also be assisted with network equipment, mobile equipment and the like for collecting traffic flow data, such as vehicle-mounted networking equipment, mobile phones and the like.

In addition, other traffic data influencing the traffic flow in each unit time interval or preset time interval can be obtained, wherein the other traffic data comprise the type of a crossing, the phase, weather information, the traffic flow in the previous time period, holiday information, seasons, pedestrian flow and the like. Wherein, the intersection type can include: the intersection type control system is characterized in that the intersection type control system comprises a cross shape, an X shape, a T shape, a Y shape, a staggered intersection, a composite intersection and the like according to the form of the intersection, a simple intersection, a widened intersection type intersection, a channelized intersection and the like according to the degree of channelized traffic, and a signal control intersection, a signal control intersection (point control, line control and surface control) (fixed period and non-fixed period, manual control and automatic control) and the like according to the traffic control classification.

The phases are different signal lamp control types required by a certain intersection based on a vehicle motion mode, corresponding releasing time is given for traffic flows in different directions, and different phases are independent in batches and do not interfere with each other. For example, a standard intersection has twelve vehicle movement patterns, namely straight (east-west, west-east, south-north, north-south), small turn (east-north, west-south, north-west, south-east), and large turn (east-south, west-north, north-east, south-west). These twelve movements can be divided into four groups:

1) The east-west goes straight: east-west, west-east, east-north, west-south;

2) The south and north move straight: north-south, south-east, north-west;

3) Southeast, northwest: east-south, west-north;

4) North, south and east: north-east, south-west;

the 4 groups of signal lamps described above require different signal control, i.e. four different phases.

Meanwhile, the influence of different weather on the road traffic flow is different in different seasons, the traffic flow of different phases at the same intersection and the traffic flow at different time periods are possibly greatly different, such as early peak and late peak; for example, on holidays such as saturday, traffic flows at the same intersection in the same time period may be greatly different. For example, the traffic flow at the early peak of the working day in rainy or snowy weather may be greatly different from the traffic flow at the early peak of the rest day in rainy or snowy weather.

In the present embodiment, the first data includes the traffic flow data and a plurality of factors affecting the traffic flow in many aspects, and more accurate results can be obtained when the traffic flow is analyzed, predicted, or the like.

S102: and inputting the first data into a flow prediction model to obtain predicted traffic flow.

The vehicle flow in the traffic data and a plurality of influence factors (data) contained in the traffic data are input into the flow prediction model for processing, so that the predicted traffic flow of the target intersection in the target time period needing to be predicted can be obtained.

In the embodiment, the traffic flow data and the feature data of the influence factors influencing the traffic flow in various aspects are input into the flow prediction model, and the feature data of the influence factors can be introduced from the input layer of the flow prediction model, so that the feature data of the influence factors are introduced in the whole process of the flow prediction model, and a plurality of features including historical flow and the like and feature data are fused, the problem that the fusion of the feature data of the influence factors and the historical flow features in the related technology is delayed can be effectively solved, the influence of the feature data of the influence factors on the predicted flow can be accurately reflected, and the performance of the flow prediction model can be improved.

S103: and obtaining a traffic signal lamp dispatching strategy according to the predicted traffic flow and the working condition parameters of the target intersection. The working condition parameters of the target intersection comprise: the grade of the road, the linear characteristics of the road (such as road length, lane width, lane number, tortuosity and the like), the construction level of the intersection (such as cross section, flat longitudinal section, curve and the like), the intersection environment (such as urban road intersection, plain road intersection, mountain road intersection and the like) and other road traffic facilities (such as pedestrian crosswalk, zebra crossing and underground passage) and the like.

After the traffic flow is obtained, dynamic traffic light regulation and control can be carried out to obtain a traffic light scheduling strategy, including but not limited to traffic light combination sequence optimization, signal light duration optimization and the like, and on the basis of the combination of the predicted traffic flow and a plurality of traffic data edge calculation units, traffic light regulation and control are carried out based on factors such as different time periods, weather and the like. For example, according to the historical traffic flow of the day before a certain intersection and the result of predicting the traffic flow in the target time period, most of the time in the intersection is more vehicles coming from east and west than vehicles coming from south and north, an ideal traffic light control strategy should provide longer green light time for east and west directions, and the proportion of the green light time should be higher in the time periods such as the morning rush hour and the evening rush hour.

In the regulation and control optimization of the traffic signal lamp, according to the optimization content, the optimization can be divided into signal lamp phase and phase sequence optimization, signal lamp phase-to-green ratio (effective green time/signal lamp period) optimization, signal lamp period optimization and phase difference optimization, and influence factors which can possibly determine to obtain a traffic signal combination, including traffic flow, waiting time, idle interval time and other traffic data, can be comprehensively considered.

In steps S101 to S103, control delay due to acceleration/deceleration of the vehicle, travel time delay due to parking, and the like, which are mainly caused by signal control, may be considered. In determining the control delay, a weibert timed intersection delay model based on an approximate delay model-steady state model, for example, may be used:

where C is the cycle time(s), λ is the green signal ratio (effective green time/signal period),

is the flow ratio of the motor vehicle lane, and q is the actual flow (veh/h) of the inlet motor vehicle lane; s is the inlet channel saturation flow. Wherein the first term is the equilibrium phase delay, calculated under the assumption that the long-time arrival rate is taken as a constant value; the second term is random phase delay, which is calculated under the assumption that arrival in a short time is considered to obey poisson distribution; the third term is a random correction term and is obtained through a traffic flow simulation experiment.

Furthermore, the delay model-transient model can also be fixed: and describing the number of the vehicles arriving in a specific time period by using the integral, describing the number of the vehicles leaving in the specific time period by using the integral, modeling the queuing process, and further obtaining the delay time.

Based on the control delay, traffic light control can be optimized, for example, for single-point signal light optimization:

(1) Setting the shortest green time of each phase to ensure that the pedestrian can normally pass (the estimated speed of the pedestrian crossing the street is 1.2/s), and further obtaining the shortest signal lamp period by superposing the green time loss;

(2) Calculating the shortest green light interval or crossing clearing time according to the conflict point so as to avoid the possibility of collision of vehicles in two directions;

(3) And setting the longest signal lamp period according to the actual situation so as to avoid the problems of overlong waiting and the like.

According to the embodiment, the combination calculation is carried out according to the predicted traffic flow and the working condition parameters of the target intersection, conditions and influence factors in signal lamp regulation and optimization are fully considered, an accurate and effective signal lamp scheduling strategy is obtained, the signal lamp scheduling efficiency of the intersection is improved, and the problem of congestion is effectively solved.

S104: and scheduling the traffic signal lamp based on the traffic signal lamp scheduling strategy.

According to the embodiment, the traffic light dispatching strategy is obtained by predicting the vehicle flow of the target time period according to the flow prediction model and combining the vehicle flow of the target time period with other influence factors (such as time periods, weather, holidays and the like), and the traffic light is dispatched and controlled according to the dispatching strategy, so that the traffic light can be accurately and effectively dispatched dynamically.

In the embodiment, a video image shot by a video device arranged at the target intersection can be collected; and the video (image) information of the existing monitoring camera at the intersection can be utilized and collected, so that the upgrading and reconstruction cost and resources can be saved.

As shown in fig. 2, fig. 2 is a schematic diagram of intersection vehicle flow statistics provided by the embodiment of the present application. In the embodiment, information such as vehicle traffic is recognized based on recognition of a video image captured by a camera.

When the video image is identified, the yolov series algorithm can be adopted to carry out image identification processing on the video image, such as the yolov5 algorithm, and the method has the advantages of high detection and identification precision and high calculation speed. It should be noted that the algorithm illustrated in the present embodiment is only a preferred embodiment of the present application, and other algorithms or devices capable of achieving the same purpose may be adopted in practical applications according to practical situations, and the present application is not limited to this.

In other embodiments, the traffic prediction model may be constructed based on a combination of a convolutional neural network and a long-short term memory network and an attention mechanism, and the model may be a convolutional neural network layer, a long-short term memory network layer and a self-attention layer which are sequentially cascaded.

When the traffic prediction model is constructed, the data source may be traffic data of the intersection collected at regular time granularity, for example, vehicle traffic data with time granularity of 15 minutes, and work parameter data, weather data and weather information of the intersection. The available features (data) in the data source can be screened out, effective features such as an XgBoost algorithm and the like are selected through a correlation coefficient method, the importance of the features is ranked, and main features are selected for training. Wherein the main features may include: 1. time interval 2, whether holidays are 3, weather 4, intersection type 5, phase 6, traffic flow in the previous time interval and the like, please refer to the example graph of the importance of the characteristic data shown in fig. 3.

In addition, when the embodiment of the application constructs a model, several main data network algorithms are tried and counted, and the following table is shown:

model name	Training sample	Percent mean flow deviation	Mean value of flow deviation
				DNN	100000	18％	19
LSTM	100000	18.5％	19
				CNN	100000	16.2％	18

It can be found that the CNN network has better effect than the LSTM network and the DNN network, and therefore, the embodiment of the present application preferably trains the CNN convolutional neural network when constructing the model.

Extracting a first data characteristic corresponding to the source data through a convolutional neural network based on the data source; inputting the first data characteristic into a long-term and short-term memory network for processing to obtain a second data characteristic; and processing the second data characteristic through an attention mechanism to obtain characteristic data with large influence on the intersection flow. The feature data with large influence on the intersection flow obtained in the previous step can be used for setting the feature data with preset types and quantities according to conditions such as precision requirements and the like, and the attention mechanism can well judge the attention requirement of the current step according to the output of the previous step, so that important feature data can be emphasized or (relatively) unimportant feature data can be restrained.

In addition, the first data features can be extracted through multilayer convolutional networks, the convolutional structures are stacked into a plurality of layers of regions (receptive fields) where the convolutional neural network features of the model can be increased, and the process of mapping the bottom layer features to the high layer features can realize the identification and extraction of key information and features which have large influence on intersection flow; in the steps, the important characteristic information of the preset type and quantity is used as input according to the requirement, compared with the input of the original input source electrode, the parameters of the model are greatly reduced, and the efficiency of the model is improved.

In the embodiment, the convolutional network can better summarize the plane information of the intersection, such as the information of the intersection working parameters and the like; the long-short term memory network can memorize the change rule of the traffic flow at the intersection along with the time; by combining with the attention mechanism, the characteristic data with large influence on the intersection flow can be better summarized. The interference among different types of events is reduced, the calculation amount of a subsequent convolutional neural network is reduced, and the flow prediction speed can be improved. The characteristics of intersection traffic data are extracted through the convolutional neural network and the long-short term memory network, the time characteristics among all influencing factors are also extracted, and the extracted characteristics are more accurately represented by means of attention due to the fact that the time characteristics among events are considered. Therefore, when the model is designed and constructed, a mode of combining the convolutional neural network, the long-term and short-term memory network and the attention mechanism can be adopted, and the vehicle flow can be accurately and effectively predicted.

In other embodiments, the traffic prediction model is trained and updated based on the traffic data over a preset time interval.

In this embodiment, after the traffic prediction model is constructed, the traffic prediction model is deployed, for example, the traffic prediction model is deployed on an edge computer near a traffic signal machine, the latest intersection traffic data (traffic data) can be acquired at intervals of preset time by setting a timing task or a timing frame, and the like, so that the traffic prediction model is trained, and meanwhile, the trained latest model is thermally deployed, so that the model can be kept to learn the latest change trend of intersection traffic, and inaccurate prediction caused by model solidification is prevented.

For example, every seven days is a period, the obtained latest traffic flow data is input into the flow prediction model, the weights of all cascaded networks in the model are optimized and updated, the latest flow prediction model is correspondingly obtained, and meanwhile, the latest flow prediction model is deployed on the edge computer in real time. The edge computer can be integrated with a traffic signal lamp controller, and can also be a computer, a cloud computing server and the like which are connected in a wired or wireless mode.

The predicted traffic flow can also be subjected to rule constraint according to a data reduction strategy, so that the predicted value is prevented from deviating from the baseline too much.

In this embodiment, the data reduction strategy may adopt a parametric method such as a regression (linear regression and multiple regression) model and a log-linear model, and may also adopt a non-parametric method such as histogram, clustering, sampling, and the like (sampling). After prediction is carried out, actual data of a time period or a time node corresponding to the predicted traffic flow can be collected, parameters such as the weight of the model are adjusted after the actual data are compared with the data of the actual traffic flow, the predicted flow data result can be prevented from deviating too much from the actual result, in addition, the influence of invalid and wrong data on modeling can be reduced by combining an extrapolation model on the flow prediction model, and the modeling accuracy is improved.

According to the embodiment of the application, the flow prediction model can be obtained and flow prediction can be carried out according to data information of sensors such as a camera and a radar and information such as crossing work parameters, meanwhile, the model can be continuously trained and perfected to realize unattended autonomous learning, self-adaptive adjustment is realized, an intelligent scheduling strategy is generated, and a better scheduling effect can be obtained according to a strategy control field traffic signal machine.

Based on an inventive concept, the present application further provides a traffic signal light dispatching device 200, as shown in fig. 5, fig. 5 is a schematic structural diagram of the traffic signal light dispatching device provided in the embodiment of the present application, and the traffic signal light dispatching device 200 is specifically configured to perform the foregoing step S101-step S104 and any optional example thereof.

The traffic signal light dispatching device 200 includes:

a data obtaining module 210, configured to obtain first data; the first data is related to the traffic flow of the target intersection; the first data includes at least one of the following parameters: time period, intersection type, signal lamp phase, traffic flow of at least one previous time period, date type and weather type.

The data acquisition module 210 captures and collects traffic data and information at the intersection. The data after some necessary preprocessing is input into an image recognition device (flow prediction model) to recognize the vehicle flow in the video, and other traffic data influencing the vehicle flow in each unit time interval can be acquired by a network and the like.

The edge calculation module 220 is configured to input the first data into a traffic prediction model to obtain a predicted traffic flow; the edge calculation module 220 inputs the vehicle flow in the traffic data and a plurality of data contained in the traffic data into the flow prediction model for processing, so as to obtain the predicted traffic flow of the target time period needing to be predicted.

The edge calculation module 220 is further configured to obtain a traffic signal light scheduling policy according to the predicted traffic flow and the working condition parameters of the target intersection. After the predicted traffic flow is obtained, a traffic signal lamp scheduling strategy is generated according to other factors (traffic flow data) influencing the vehicle flow.

A traffic signal scheduling module 230, configured to schedule the traffic signal based on the traffic signal scheduling policy. The traffic signal scheduling module 230 may be a traffic signal machine at the intersection, and the edge calculating module 220 may send the scheduling policy to the traffic signal machine, so that the traffic signal performs scheduling control on traffic signals such as traffic lights.

In some other embodiments, the data acquisition module 210 further includes a video acquisition unit 211, configured to acquire a video image of the target intersection, where the video acquisition unit 211 includes, but is not limited to, a camera, a radar, and the like.

The edge calculation module 220 further includes an image recognition unit 221, configured to perform image recognition processing on the video image to obtain the vehicle flow.

In some other embodiments, the edge calculation module 220 further comprises:

and the model training unit 222 is used for constructing the flow prediction model based on the combination of the convolutional neural network, the long-term and short-term memory network and the attention mechanism.

In some embodiments, the traffic prediction model constructed by the model training unit 222 includes a convolutional neural network layer, a long-short term memory network layer, and a self-attention layer, which are sequentially cascaded. The flow prediction model constructed by the model training unit 222 is a classification model, and when trained, includes:

Wherein the first data includes at least two of the parameters, and the training of the flow prediction model further includes:

calculating importance coefficients of all parameters of the first data;

In some embodiments of the present application, the model training unit 222 is further configured to obtain new sample data, train and update the flow prediction model after a preset time interval.

In some embodiments of the present application, the obtaining, by the edge calculation module 220, a traffic signal light scheduling policy according to the predicted traffic flow and the operating condition parameter of the target intersection includes:

With respect to the traffic signal light dispatching system 200 in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

The traffic signal lamp scheduling system 200 provided by this embodiment can predict the traffic flow in advance, dynamically schedule traffic signal lamps, and meanwhile can also realize adaptive adjustment, generate an intelligent scheduling strategy, and improve the traffic efficiency.

Fig. 5 is a schematic structural diagram of a computing device 300 provided in an embodiment of the present application. The device can be used as a traffic light control device to execute various optional embodiments of the traffic light scheduling method, and the computing device can be a terminal and also can be a chip or a chip system in the terminal. As shown in fig. 5, the computing device 300 includes: a processor 310, a memory 320, and a communication interface 330.

It should be understood that the communication interface 330 in the computing device 300 shown in fig. 6 may be used for communicating with other devices, and may specifically include one or more transceiver circuits or interface circuits.

The processor 310 may be connected to the memory 320. The memory 320 may be used to store the program codes and data. Therefore, the memory 320 may be a storage unit inside the processor 310, an external storage unit independent of the processor 310, or a component including a storage unit inside the processor 310 and an external storage unit independent of the processor 310.

Optionally, the computing device 300 may also include a bus. The memory 320 and the communication interface 330 may be connected to the processor 310 through a bus. The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, a line without arrows is used in FIG. 3, but does not indicate that there is only one bus or one type of bus.

It should be understood that, in the embodiment of the present application, the processor 310 may adopt a Central Processing Unit (CPU). The processor may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, 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. Or the processor 310 may employ one or more integrated circuits for executing related programs to implement the technical solutions provided in the embodiments of the present application.

The memory 320 may include both read-only memory and random access memory and provides instructions and data to the processor 310. A portion of the processor 310 may also include non-volatile random access memory. For example, the processor 310 may also store information of the device type.

When the computing device 300 is running, the processor 310 executes the computer-executable instructions in the memory 320 to perform any of the operational steps of the methods described above and any optional embodiment thereof.

It should be understood that the computing device 300 according to the embodiment of the present application may correspond to a corresponding main body for executing the method according to the embodiments of the present application, and the above and other operations and/or functions of each module in the computing device 300 are respectively for implementing corresponding flows of each method of the embodiment, and are not described herein again for brevity.

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 implementation. 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.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

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

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

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The present embodiments also provide a computer-readable storage medium, on which a computer program is stored, where the computer program is used to execute the method described above when executed by a processor, and the method includes at least one of the solutions described in the above embodiments.

The computer storage media of the embodiments of the present application may take any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). Additionally, the terms first, second, third and the like in the description and in the claims, or module A, module B, module C and the like are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order, it being understood that specific orders or sequences may be interchanged where permissible to effect embodiments of the application described herein in other sequences than illustrated or described herein.

In the above description, reference numbers indicating steps, such as S110, S120 … …, etc., do not necessarily indicate that the steps are executed in this order, and the order of the preceding and following steps may be interchanged or executed simultaneously, if permitted.

The term "comprising" as used in the specification and claims should not be construed as being limited to the contents listed thereafter; it does not exclude other elements or steps. It should therefore be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, and groups thereof. Thus, the expression "a device comprising means a and B" should not be limited to a device consisting of only components a and B.

Reference in the specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments, as would be apparent to one of ordinary skill in the art from this disclosure.

It is to be noted that the foregoing is only illustrative of the presently preferred embodiments and application of the principles of the present invention. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application.

Claims

1. A traffic signal light scheduling method is characterized by comprising the following steps:

inputting the first data into a flow prediction model to obtain predicted traffic flow;

2. The method of claim 1, wherein the first data comprises at least one of:

time period, intersection type, signal lamp phase, vehicle flow of at least previous time period, date type and weather type.

3. The method of claim 2, wherein the flow prediction model is a classification model comprising:

4. The method of claim 2 or 3, wherein the flow prediction model is a classification model, which when trained, comprises:

and taking the input label of the sample data as the input of the flow prediction model, taking the output label as the expected input and output of the flow prediction model, and training the flow prediction model.

5. The method of claim 4, wherein the first data includes at least two of the parameters, and wherein training the flow prediction model further comprises:

calculating an importance coefficient of each parameter of the first data;

and coupling the importance coefficients of the parameters as the input of the flow prediction model and the output labels as the expected input and output of the flow prediction model, and training the flow prediction model.

6. The method according to claim 4 or 5, comprising:

acquiring new sample data after a preset time interval;

7. The method of claim 1, wherein said deriving a predicted traffic flow further comprises: and carrying out rule constraint on the predicted traffic flow according to a data protocol strategy.

8. The method according to claim 1, wherein obtaining a traffic signal light scheduling strategy according to the predicted traffic flow and the condition parameters of the target intersection comprises:

and adjusting the combination sequence of the traffic signal lights and the time length of the signal lights according to the length and the width of the intersection, the passing time length and the waiting time length of the vehicle under different vehicle flow conditions.

9. A traffic signal light dispatching device is characterized by comprising:

10. A computing device, comprising:

a processor, and an interface circuit, wherein the processor accesses a memory through the interface circuit, the memory storing program instructions that, when executed by the processor, cause the processor to perform the traffic signal light scheduling method of any of claims 1 to 8.