Traffic management method based on intelligent calculation

Abstract

The invention discloses a traffic management method based on intelligent computing, an edge computing node, a cloud server and a storage medium, and belongs to the technical field of traffic control. The method comprises the following steps: receiving road data collected by a sensor, and determining road recognition features and road scene types according to the road data; determining a traffic management decision mode according to the road scene type; when the traffic management decision mode is a cloud management decision, the road recognition features are sent to a cloud server, so that the cloud server determines first control data of the traffic management equipment according to the road recognition features and sends the first control data to the edge computing node; and receiving the first control data, and controlling the traffic management equipment according to the first control data, thereby realizing real-time dynamic traffic control and effectively reducing the cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of traffic control, in particular to a traffic management method based on intelligent computing, an edge computing node, a cloud server and a storage medium. BACKGROUND

[0002] With the development of intelligent transportation systems, the traditional technical architecture gradually exposes the theoretical limitations of multiple dimensions. In the past, the traffic control system was established on the basis of a static decision-making model, and its core operation logic depended on the periodic matching of the preset fixed timing scheme and the historical traffic pattern. This fixed decision-making logic is not suitable for the current traffic situation when the traffic flow changes. At present, real-time data is obtained and a traffic control system is constructed according to the real-time data, but this way requires a large amount of data to be transmitted to the traffic control system in actual use, thereby causing the problem of high application cost.

[0003] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide a traffic management method based on intelligent computing, an edge computing node, a cloud server and a storage medium, which aims to realize real-time dynamic traffic control and effectively reduce the cost. In order to achieve the above purpose, the present application provides a traffic management method based on intelligent computing, which is applied to an edge computing node, and the traffic management method based on intelligent computing comprises the following steps:

[0005] Receiving road data collected by a sensor, and determining road recognition features and road scene types according to the road data;

[0006] Determining a traffic management decision mode according to the road scene types;

[0007] When the traffic management decision mode is a cloud management decision, sending the road recognition features to a cloud server, so that the cloud server determines first control data of the traffic management device according to the road recognition features, and sends the first control data to the edge computing node;

[0008] Receiving the first control data and controlling the traffic management device according to the first control data. Optionally, after the step of determining the traffic management decision mode according to the road scene types, the method further comprises:

[0009] When the traffic management decision mode is an edge node management decision, determining a corresponding management mapping table according to the road scene types;

[0010] determining second control data according to the management mapping table and the road recognition feature;

[0011] controlling a traffic management device according to the second control data.

[0012] Optionally, the road data comprises: a vehicle motion image of a vehicle driving, point cloud data of a road, and obstacle echo data, and the step of determining a road recognition feature and a road scene type according to the road data comprises:

[0013] respectively performing data cleaning on the vehicle motion image, the point cloud data, and the obstacle echo data;

[0014] fusing the vehicle motion image, the point cloud data, and the obstacle echo data according to a data fusion algorithm to obtain target fusion data;

[0015] generating the road recognition feature according to the target fusion data, and determining a road scene type according to the road recognition feature.

[0016] Optionally, the road recognition feature comprises: a vehicle speed on the road, traffic flow data, and a moving direction of a vehicle, and the step of determining a road scene type according to the road recognition feature comprises:

[0017] when the vehicle speed is greater than or equal to a preset vehicle speed, the traffic flow data is less than a preset traffic flow data, and all the moving directions of the vehicles are the same as a preset vehicle driving direction, determining that the road scene is a low-flow scene;

[0018] when the vehicle speed is less than a preset vehicle speed, generating a congestion identifier corresponding to the road recognition feature;

[0019] when the traffic flow data is greater than or equal to the preset traffic flow data, generating a large-flow identifier corresponding to the road recognition feature;

[0020] when any of the moving directions is the same as the preset vehicle driving direction, generating an abnormal moving identifier corresponding to the road recognition feature;

[0021] when the road recognition feature has a corresponding identifier, determining the road scene type according to the identifier type.

[0022] Optionally, the first control data comprises: local lane closure, intersection flow control, and emergency vehicle priority passage.

[0023] In addition, to achieve the above-mentioned purpose, the intelligent computing-based traffic management method can be applied to a cloud server, and the intelligent computing-based traffic management method comprises the following steps:

[0024] receiving the road recognition feature sent by the edge computing node;

[0025] determining first control data of the traffic management device according to the road recognition feature;

[0026] sending the first control data to the edge computing node, so that the edge computing node receives the first control data and controls the traffic management device according to the first control data.

[0027] Optionally, the step of determining the first control data of the traffic management device according to the road recognition feature comprises:

[0028] generating a digital twin simulation model according to the road recognition feature;

[0029] determining a plurality of alternative control strategies according to the digital twin model;

[0030] determining the first control data according to the plurality of alternative control strategies.

[0031] In addition, to achieve the above object, the application further provides an edge computing node, characterized in that the edge computing node comprises a memory, a processor and an intelligent computing-based traffic management program stored in the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to implement the steps of the intelligent computing-based traffic management method applied to the edge computing node according to any one of the above.

[0032] In addition, to achieve the above object, the application further provides a cloud server, characterized in that the cloud server comprises a memory, a processor and an intelligent computing-based traffic management program stored in the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to implement the steps of the intelligent computing-based traffic management method applied to the cloud server according to any one of the above.

[0033] In addition, to achieve the above object, the application further provides a storage medium, characterized in that the storage medium stores an intelligent computing-based traffic management program, and the intelligent computing-based traffic management program implements the steps of the intelligent computing-based traffic management method according to any one of the above when executed by a processor.

[0034] The application provides a traffic management method based on intelligent computing, which receives road data collected by a sensor, determines road identification features and a road scene type according to the road data, determines a traffic management decision mode according to the road scene type, and identifies the road scene type, so that cloud management decision can be realized in some modes, that is, when the traffic management decision mode is cloud management decision, the road identification features are sent to a cloud server, and the first control data is sent to the edge computing node; the first control data is received, and traffic management equipment is controlled according to the first control data, so that the cloud server can be used to control traffic only in some scenes, real-time dynamic traffic control can be realized, and the cost can be effectively reduced. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is a structural schematic diagram of an edge computing node of a hardware running environment related to an embodiment scheme of the application.

[0036] Figure 2 is a flow schematic diagram of a first embodiment of the traffic management method based on intelligent computing of the application.

[0037] Figure 3 is a flow schematic diagram of a third embodiment of the traffic management method based on intelligent computing of the application.

[0038] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0039] It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

[0040] Reference Figure 1 , Figure 1 is a structural schematic diagram of an edge computing node of a hardware running environment related to an embodiment scheme of the application.

[0041] As Figure 1As shown, the edge computing node can include a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interaction device 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between the components. The interaction device 1003 can include a display, an input unit such as a keyboard, and can also be connected with the communication bus through a standard wired interface or a wireless interface. The network interface 1004 can optionally include a standard wired interface or a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 1005 can be a high-speed random access memory (RAM) memory or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 can also be a storage device independent of the aforementioned processor 1001.

[0042] Those skilled in the art can understand that Figure 1 The structure shown in the figure does not constitute a limitation on the edge computing node, and can include more or fewer components than the figure, or combine certain components, or different component arrangements.

[0043] As Figure 1 As shown, the memory 1005 as a storage medium can include an operating system, a data storage module, a network communication module, a user interface module, and a smart computing-based traffic management program.

[0044] In Figure 1 In the edge computing node shown, the network interface 1004 is mainly used for data communication with other devices; the interaction device 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the edge computing node of the application can be arranged in the edge computing node, and the edge computing node calls the smart computing-based traffic management program stored in the memory 1005 through the processor 1001, and executes the smart computing-based traffic management method provided by the embodiment of the application.

[0045] The embodiment of the application provides a smart computing-based traffic management method, which is described with reference to Figure 2 , Figure 2 The flowchart of the first embodiment of the smart computing-based traffic management method of the application is shown.

[0046] In this embodiment, the smart computing-based traffic management method applied to the edge computing node includes:

[0047] Step S1, receiving road data collected by a sensor, and determining road recognition features and a road scene type according to the road data;

[0048] Specifically, the position of the sensor is not limited and can be arranged on both sides of the road or above the road. The sensor can include at least one of a camera, a laser radar, and a millimeter wave radar. The edge computing node can be used for data transmission, for example, receiving road data collected by a sensor and saving the data. In addition, the edge computing node can be used for data processing and calculation. Specifically, the road data is preprocessed, the feature values in the road data are extracted, and it is controlled whether the data needs to be sent to the cloud server. Specifically, the edge computing node can receive one or more road data and perform data cleaning to remove noise in the road data. Optionally, the image distortion in the road data can also be adjusted. The corresponding road recognition features are extracted from the road data. The road recognition features include the features of the driving process of each vehicle on the road and the features of the motion of pedestrians. The current road scene type is analyzed according to the road recognition features.

[0049] Step S2, determining a traffic management decision mode according to the road scene type;

[0050] In this embodiment, different road scene types correspond to different traffic management decision modes. Generally, in a road scene in a normal operating state, complex traffic management is not actually needed, so traffic management decision can be made at the edge computing node, that is, the traffic management decision mode is edge node management decision. When there is a complex traffic accident or a road needs to be closed for performance, etc., the traffic management decision mode is determined to be cloud management decision.

[0051] Step S3, when the traffic management decision mode is cloud management decision, sending the road recognition features to the cloud server, so that the cloud server determines first control data of the traffic management device according to the road recognition features, and sends the first control data to the edge computing node;

[0052] Specifically, when the traffic management decision mode is cloud management decision, all road recognition features stored by the edge computing node can be sent to the cloud server, or only a part of the relevant road recognition features can be sent to the cloud server. It should be noted that when the cloud management decision mode is determined, a corresponding sending interval can be set, and the road recognition features are sent to the cloud server according to the sending interval. During the sending process, only the road recognition features obtained at the current time are sent. Thus, the amount of data sent can be reduced, thereby reducing the cost of communication. Preferably, the first control data herein is only the first control data generated according to the road recognition features sent by the edge computing node currently in the cloud management decision mode.

[0053] Step S4, receiving the first control data, and controlling the traffic management equipment according to the first control data.

[0054] Specifically, the control instructions for each traffic management equipment in the road are generated according to the first control data, and the traffic management equipment is controlled according to the control instructions. In this embodiment, the traffic management equipment can be a traffic signal lamp or the like. Alternatively, in some embodiments, the traffic management personnel is instructed by the control instructions to manage the traffic flow on the road according to the instructions, for example, adjusting or setting the position of the water horse barrier. In addition, alternatively, for vehicles that can execute the VX2 protocol, corresponding prompt information or instructions are generated according to the first control data, and are broadcast to the relevant vehicles through V2X.

[0055] In this embodiment, the road data collected by the sensor is received, the road recognition features and the road scene type are determined according to the road data, the traffic management decision mode is determined according to the road scene type, and the traffic management decision mode is determined according to the road scene type. By identifying the road scene type, it can be realized that the cloud management decision is used in some modes, that is, when the traffic management decision mode is cloud management decision, the road recognition features are sent to the cloud server, and the first control data is sent to the edge computing node; the first control data is received, and the traffic management equipment is controlled according to the first control data, thereby realizing the use of the cloud server to control the traffic only in some scenes, thereby realizing real-time dynamic traffic control, and effectively reducing the cost.

[0056] Further, based on the first embodiment, a second embodiment of the traffic management method based on intelligent computing is proposed. In this embodiment, after the step of determining the traffic management decision mode according to the road scene type, the following steps are further included:

[0057] When the traffic management decision mode is edge node management decision, a corresponding management mapping table is determined according to the road scene type;

[0058] Preferably, the management mapping table herein is fixed, for example: in the scene of low traffic, increase the green light time of the main road, and start the speed measuring device, optionally, in the state of general traffic, the traffic signal lamp can be controlled according to the pre-set traffic signal lamp time, the safety belt shooting device starts and so on. It should be noted that the corresponding management mapping table is determined according to the road scene type, since the corresponding management mapping table is pre-stored data, and no calculation is needed, therefore, the use of computing resources can be effectively reduced in practice.

[0059] According to the management mapping table and the road recognition feature, the second control data is determined.

[0060] Specifically, according to the traffic flow data on the road and the management mapping table, the second control data for controlling the traffic management device is generated. In the management mapping table, the parameters corresponding to the traffic flow data can also include vehicle speed and vehicle moving direction. For example: the statistical value of the moving direction of the vehicle at the intersection and the management mapping table determine the time of each direction of the traffic signal lamp.

[0061] According to the second control data, the traffic management device is controlled.

[0062] The second control data generates corresponding control instructions, and the traffic management device is controlled according to the control instructions. Optionally, the control instructions herein can be instructions for setting the operating parameters of the traffic management device, and the control of the traffic management device is realized by adjusting the operating parameters of the traffic management device

[0063] In the embodiment, when the traffic management decision mode is the edge node management decision, the corresponding management mapping table is determined according to the road scene type, so that a control mode suitable for the edge node can be selected, specifically, the second control data is determined according to the management mapping table and the road recognition feature, thereby realizing the low-cost control of the traffic management device.

[0064] Further, based on the first embodiment or the second embodiment, the third embodiment of the traffic management method based on intelligent calculation of the present application is proposed, in which Figure 3 , the road data includes: vehicle motion image of the vehicle driving, point cloud data of the road and obstacle echo data, and the step of determining the road recognition feature and the road scene type according to the road data includes:

[0065] Step S11, respectively, the vehicle motion image, the point cloud data and the obstacle echo data are data cleaned;

[0066] Specifically, the noise and abnormal data values collected in the road data are removed by data cleaning. Commonly

[0067] Step S12, according to the data fusion algorithm, the vehicle motion image, the point cloud data and the obstacle echo data are fused to obtain target fusion data;

[0068] Generally, by mapping the vehicle motion image and the echo data into the point cloud data space, specifically by matching the feature points, for example, by mapping the vehicle motion image and the echo data into the point cloud of the road divider in the point cloud data through the corner points of the road divider. After mapping, a part of the points in the point cloud data are set by the corresponding three-channel storage space for storing the color information obtained by image mapping.

[0069] Step S13, according to the target fusion data, the road recognition feature is generated, and the road scene type is determined according to the road recognition feature.

[0070] In this embodiment, the type of the road recognition feature is generally set in advance, and the corresponding feature extraction algorithm is set, for example, the point cloud data space is segmented according to the lanes, and the number of vehicles on each lane and the driving direction of the vehicles are identified according to the image data. Common image processing methods can realize the identification of vehicles on the lane, for example, template matching and deep learning model.

[0071] In this embodiment, the vehicle motion image, the point cloud data and the obstacle echo data are respectively cleaned, and the vehicle motion image, the point cloud data and the obstacle echo data are fused according to the data fusion algorithm to obtain target fusion data, and the road recognition feature is generated according to the target fusion data, and the road scene type is determined according to the road recognition feature, so as to improve the accuracy of the road scene recognition.

[0072] Further, based on any of the above embodiments, the fourth embodiment of the intelligent computing-based traffic management method of the present application is proposed. In this embodiment, the road recognition feature includes vehicle speed, traffic flow data and vehicle moving direction on the road, and the step of determining the road scene type according to the road recognition feature includes:

[0073] When the vehicle speed is greater than or equal to the preset vehicle speed, the traffic flow data is less than the preset traffic flow data, and the moving direction of all vehicles is the same as the preset vehicle driving direction, the road scene is determined as a low-flow scene;

[0074] Among them, the low-flow scene is a simple type scene.

[0075] When the vehicle speed is less than the preset vehicle speed, the congestion identifier corresponding to the road recognition feature is generated;

[0076] When the traffic volume data is greater than or equal to the preset traffic volume data, a large-flow identifier corresponding to the road recognition feature is generated.

[0077] When any of the moving directions is the same as the preset vehicle driving direction, an abnormal moving identifier corresponding to the road recognition feature is generated.

[0078] When the road recognition feature has a corresponding identifier, the road scene type is determined according to the identifier type.

[0079] In this embodiment, the number of identifier types is not limited, and is not limited to the above identifier types. For each road recognition feature, one or more corresponding identifier types can be set, for example, the change of the moving direction of each vehicle is identified. When the moving direction of at least two vehicles frequently changes, or the number of lane changes is greater than a preset number of lane changes, an aggressive driving identifier can be generated. Of course, the road recognition feature can also include whether there is a special vehicle. When there is a special vehicle, a corresponding identifier is set. The number of marker types corresponding to the road recognition feature of the current road data is counted. Optionally, the road scene type is determined according to the number of marker types. Specifically, when the number of marker types is greater than or equal to a preset number of marker types, the road scene type is determined to be a complex type scene, and when the number of marker types is less than the preset number of marker types, the road scene type is determined to be a simple type scene. In other embodiments, different marker types are set to have different complexity scores. The total complexity score of the marker types corresponding to the road recognition feature of the current road data is calculated, and the road scene type is determined according to the total complexity score, specifically, a complex type scene or a simple type scene. In addition, the traffic management decision mode determined for the complex type scene is cloud management decision, and the traffic management decision mode determined for the simple type scene is edge node management decision.

[0080] In this embodiment, the data of multiple road recognition features is used to generate corresponding identifier types, and then the statistical results or statistical complexity scores of the identifier types are used, so that the standardized classification of road scenes under multiple different data can be accurately measured, and the determination of different types of traffic management decision modes for subsequent complex type scenes and simple type scenes plays an important role.

[0081] Further, the first control data includes local lane closure, intersection flow regulation, and emergency vehicle priority passage.

[0082] When a local lane is closed, the traffic signal of the corresponding lane is adjusted to red, and information is sent to the management personnel or traffic director, so that the management personnel or traffic director can effectively close the local lane. For intersection flow control and emergency vehicle priority passage, corresponding control instructions can also be generated to realize the first control data.

[0083] The embodiment of the application provides a fifth embodiment of a traffic management method based on intelligent computing, and in the embodiment, the traffic management method based on intelligent computing is applied to a cloud server and includes the following steps.

[0084] Receiving the road recognition feature sent by the edge computing node;

[0085] Determining first control data of the traffic management device according to the road recognition feature;

[0086] Sending the first control data to the edge computing node, so that the edge computing node receives the first control data and controls the traffic management device according to the first control data.

[0087] In the embodiment, the number of cloud servers is multiple, that is, there are multiple cloud server rooms, and specifically, in the embodiment, when it is detected that the edge computing node approaches the computing power threshold, a load migration protocol is started, and the computing process of some edge computing nodes can also be migrated to the cloud server. In other embodiments, the computing task is distributed to the corresponding cloud server and edge computing node through a chaotic load evaluation model. In addition, it should be noted that the cloud server can update the running data of the edge computing node to optimize the edge computing node.

[0088] Further, the step of determining the first control data of the traffic management device according to the road recognition feature includes:

[0089] Generating a digital twin simulation model according to the road recognition feature;

[0090] Determining a plurality of alternative control strategies according to the digital twin model;

[0091] Determining the first control data according to the plurality of alternative control strategies.

[0092] In the embodiment, the digital twin simulation model is generated according to the road recognition features, a plurality of alternative control strategies are determined according to the digital twin simulation model, so that the simulation of the digital twin simulation model is realized through the cloud server with high computing power, a plurality of prediction results are generated according to the plurality of alternative control strategies, the first control data is determined according to the data indexes of the prediction results, and the accuracy of real-time dynamic traffic control is realized.

[0093] In the embodiment, the digital twin simulation model is generated according to the road recognition features, a plurality of alternative control strategies are determined according to the digital twin simulation model, so that the simulation of the digital twin simulation model is realized through the cloud server with high computing power, a plurality of prediction results are generated according to the plurality of alternative control strategies, the first control data is determined according to the data indexes of the prediction results, and the accuracy of real-time dynamic traffic control is realized.

[0094] In addition, the embodiment of the present application further provides an edge computing node, characterized in that the edge computing node comprises a memory, a processor and an intelligent computing-based traffic management program stored in the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to realize the steps of the intelligent computing-based traffic management method in any one of the first embodiment to the fourth embodiment.

[0095] In addition, the embodiment of the present application further provides a cloud server, characterized in that the cloud server comprises a memory, a processor and an intelligent computing-based traffic management program stored in the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to realize the steps of the fifth embodiment of the intelligent computing-based traffic management method.

[0096] In addition, the embodiment of the present application further provides a storage medium, wherein the storage medium stores an intelligent computing-based traffic management program, and the intelligent computing-based traffic management program realizes the steps of the intelligent computing-based traffic management method in any one of the above embodiments when executed by a processor.

[0097] It should be noted that, in this document, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0098] The above-mentioned embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0099] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and the necessary general hardware platform, of course, they can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) as described above, and includes a number of instructions for making a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) execute the methods described in the various embodiments of the present application.

[0100] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A traffic management method based on intelligent computing, characterized in that, The intelligent computing-based traffic management method applied to an edge computing node comprises the following steps: receiving road data collected by sensors and determining road identification features and road scene types according to the road data; determining a traffic management decision mode according to the road scene types; when the traffic management decision mode is a cloud management decision, sending the road identification features to a cloud server to enable the cloud server to determine first control data of the traffic management device according to the road identification features and send the first control data to the edge computing node; receiving the first control data and controlling the traffic management device according to the first control data.

2. The intelligent computing-based traffic management method of claim 1, wherein, After the step of determining the traffic management decision mode according to the road scene types, the method further comprises the following steps: when the traffic management decision mode is an edge node management decision, determining a corresponding management mapping table according to the road scene types; determining second control data according to the management mapping table and the road identification features; controlling the traffic management device according to the second control data.

3. The intelligent computing-based traffic management method of claim 1, wherein, The road data comprises vehicle motion images of vehicles traveling, point cloud data of roads and obstacle echo data, and the step of determining road identification features and road scene types according to the road data comprises the following steps: respectively performing data cleaning on the vehicle motion images, the point cloud data and the obstacle echo data; performing fusion on the vehicle motion images, the point cloud data and the obstacle echo data according to a data fusion algorithm to obtain target fusion data; generating the road identification features according to the target fusion data and determining road scene types according to the road identification features.

4. The intelligent computing-based traffic management method of claim 1, wherein, The road identification features comprise vehicle speeds on roads, vehicle flow data and moving directions of vehicles, and the step of determining road scene types according to the road identification features comprises the following steps: when the vehicle speed is greater than or equal to a preset vehicle speed, the vehicle flow data is less than a preset vehicle flow data, and the moving directions of all vehicles are the same as a preset vehicle driving direction, determining that the road scene is a low-flow scene; when the vehicle speed is less than a preset vehicle speed, generating a congestion identifier corresponding to the road identification features; when the vehicle flow data is greater than or equal to the preset vehicle flow data, generating a large-flow identifier corresponding to the road identification features; when any of the moving directions is the same as the preset vehicle driving direction, generating an abnormal moving identifier corresponding to the road identification features; when the road identification features have corresponding identifiers, determining the road scene types according to the identifier types.

5. The intelligent computing-based traffic management method according to any one of claims 1 to 4, wherein, The first control data comprises local lane closure, intersection flow regulation and emergency vehicle priority passage.

6. A traffic management method based on intelligent computing, characterized by, The intelligent computing-based traffic management method applied to a cloud server comprises the following steps: receiving the road identification features sent by an edge computing node; determining first control data of the traffic management device according to the road identification features; The first control data is sent to the edge computing node, so that the edge computing node receives the first control data and controls the traffic management device according to the first control data.

7. The intelligent computing based traffic management method as claimed in claim 6, wherein, The step of determining the first control data of the traffic management device according to the road identification feature comprises: generating a digital twin simulation model according to the road identification feature; determining a plurality of alternative control strategies according to the digital twin model; determining the first control data according to the plurality of alternative control strategies.

8. An edge computing node, characterized by, The edge computing node comprises a memory, a processor, and an intelligent computing-based traffic management program stored on the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to implement the steps of the intelligent computing-based traffic management method according to any one of claims 1 to 5.

9. A cloud server, characterized by The cloud server comprises a memory, a processor, and an intelligent computing-based traffic management program stored on the memory and executable on the processor, and the intelligent computing-based traffic management program is configured to implement the steps of the intelligent computing-based traffic management method according to any one of claims 6 to 7.

10. A storage medium, characterized by The storage medium stores an intelligent computing-based traffic management program, and the intelligent computing-based traffic management program, when executed by a processor, implements the steps of the intelligent computing-based traffic management method according to any one of claims 1 to 7.

Images