Computing power scheduling methods, devices, electronic equipment and storage media

Abstract

This application relates to the field of Internet of Things (IoT) technology, and provides a computing power scheduling method, apparatus, electronic device, and storage medium. The computing power scheduling method includes: acquiring the functional status and redundant computing power of each edge computing device in each sub-node; receiving the traffic status of a corresponding road intersection monitored by each edge computing device in each sub-node, wherein the traffic status of the road intersection is determined by the edge computing device at the road intersection after calculating all received roadside sensing data based on its own computing power, and the roadside sensing data is generated by the roadside sensing devices at the road intersection; if the traffic status of the target intersection is abnormal, then scheduling the computing power of multiple edge computing devices at the target intersection according to the functional status and redundant computing power of all edge computing devices. The above scheme can achieve balanced allocation of computing power among edge computing devices in the system.

Description

Technical Field

[0001] This application belongs to the field of Internet of Things (IoT) technology, and in particular relates to a computing power scheduling method, device, electronic device, and storage medium. Background Technology

[0002] Edge computing refers to an open platform that integrates network, computing, storage, and application capabilities on the side closest to the data source, providing services to the nearest end. Mobile edge computing (MEC) devices, compared to terminal devices, have higher hardware specifications such as central processing units (CPUs) and memory, possessing considerable computing power resources, and are typically used in fields such as intelligent transportation. These edge computing devices handle CPU-intensive tasks most of the time, such as traffic condition fault detection.

[0003] Edge computing devices typically have redundant computing power, which can be used to handle other tasks temporarily assigned by the CPU. However, because the computing power among edge computing devices is usually unbalanced, and it is difficult to balance the computing power among edge computing devices when individual edge computing devices fail, the CPU's resource tasks cannot be completed, and the system cannot maintain normal operation. Summary of the Invention

[0004] This application provides a computing power scheduling method, apparatus, electronic device, and storage medium, which can achieve balanced allocation of computing power among edge computing devices in the system.

[0005] A first aspect of this application provides a computing power scheduling method, the computing power scheduling method comprising:

[0006] Obtain the functional status and redundant computing power of each edge computing device in each child node;

[0007] The system receives traffic status data from each edge computing device in each sub-node, which is monitored by the corresponding road intersection. The traffic status of the road intersection is determined by the edge computing device at the road intersection after calculating all the received road test data based on its own computing power. The road test data is generated by the road test device at the road intersection.

[0008] If the traffic status of the target intersection is abnormal, the computing power of multiple edge computing devices at the target intersection will be scheduled according to the functional status and redundant computing power of all edge computing devices.

[0009] A second aspect of this application provides a computing power scheduling device, the computing power scheduling device comprising:

[0010] The acquisition module is used to acquire the functional status and redundant computing power of each edge computing device in each sub-node;

[0011] The receiving module is used to receive the traffic status of the corresponding road intersection monitored by each edge computing device in each sub-node. The traffic status of the road intersection is determined by the edge computing device of the road intersection after calculating all the received road test perception data according to its own computing power. The road test perception data is generated by the road test device of the road intersection.

[0012] The scheduling module is used to schedule the computing power of multiple edge computing devices at the target intersection based on the functional status and redundant computing power of all edge computing devices if the traffic status of the target intersection is abnormal.

[0013] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the computing power scheduling method described in the first aspect above.

[0014] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the computing power scheduling method described in the first aspect.

[0015] The fifth aspect of this application provides a computer program product that, when run on an electronic device, causes the electronic device to execute the computing power scheduling method described in the first aspect.

[0016] The beneficial effects of the embodiments in this application compared with the prior art are:

[0017] In this embodiment, where sub-nodes are configured at each road intersection, the functional status and redundant computing power of each edge computing device in each sub-node are first acquired to determine the function and computing capability of each edge computing device. Secondly, the traffic status of the corresponding road intersection detected by each edge computing device in each sub-node is received. The traffic status of the road intersection is determined by the edge computing device at the road intersection after calculating all received roadside sensing data based on its own computing power. The roadside sensing data is generated by the roadside sensing devices at the road intersection. Finally, the acquired and received data are summarized. If the traffic status of the target intersection is abnormal, the computing power of multiple edge computing devices at the target intersection is scheduled according to the functional status and redundant computing power of all edge computing devices. This scheme, which schedules computing power among edge computing devices based on their functional status and redundant computing power, can achieve a balanced allocation of computing power among the edge computing devices in the system. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the structure of a child node provided in Embodiment 1 of this application;

[0020] Figure 2 This is a flowchart illustrating a computing power scheduling method provided in Embodiment 2 of this application;

[0021] Figure 3 This is a diagram illustrating the computing power scheduling of edge computing devices.

[0022] Figure 4 This is a flowchart of the computing power scheduling process when the traffic status of the target intersection is abnormal.

[0023] Figure 5 This is a diagram illustrating the location coordinates of the abnormal event.

[0024] Figure 6 This is another schematic diagram illustrating the acquisition of the location coordinates of the abnormal event;

[0025] Figure 7 This is a schematic diagram of the target road test equipment;

[0026] Figure 8 This is a schematic diagram of the edge computing device structure for selecting targets from the corresponding sub-nodes of other intersections;

[0027] Figure 9 It is a schematic diagram of the vehicle's trajectory;

[0028] Figure 10 This is a schematic diagram of the computing power scheduling device provided in Embodiment 3 of this application;

[0029] Figure 11 This is a schematic diagram of the structure of the electronic device provided in Embodiment 4 of this application. Detailed Implementation

[0030] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0031] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0032] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0033] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0034] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0035] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0036] With the continuous development of technology, the concept of empowering high-quality development of urban clusters through new infrastructure has gradually matured. New infrastructure, also known as new type of infrastructure construction, mainly includes seven major areas: 5G base station construction, ultra-high voltage power transmission, intercity high-speed railways and urban rail transit, new energy vehicle charging piles, big data centers, artificial intelligence, and industrial internet. It involves many industrial chains and is an infrastructure system based on the new development concept, driven by technological innovation, and grounded in information networks. It is oriented towards the needs of high-quality development and provides services such as digital transformation, intelligent upgrading, and integrated innovation.

[0037] In practical applications of urban rail transit, according to the concept of new infrastructure, the construction of urban rail transit requires a certain level of computing power at the roadside. This means that the construction of new urban rail transit systems requires a large number of edge computing devices to be installed at the roadside. Edge computing refers to an open platform that integrates network, computing, storage, and application core capabilities on the side closest to the data source, providing services to the nearest point. Compared to terminal devices, mobile edge computing devices have higher hardware specifications, such as central processing units and memory, and possess considerable computing power resources. In the construction of new urban rail transit systems, such edge computing devices can detect traffic faults. Furthermore, in addition to detecting traffic faults, edge computing devices typically have redundant computing power to handle other tasks temporarily assigned by the CPU. Due to the uneven distribution of these temporary CPU tasks, the computing power among edge computing devices is often unbalanced. Moreover, when individual edge computing devices fail, it is difficult to evenly allocate computing power among them, resulting in the inability to complete CPU tasks and the system failing to maintain normal operation.

[0038] To address the problems in the prior art, this application proposes a computing power scheduling method. In this embodiment, where sub-nodes are configured at each road intersection, the method first acquires the functional status and redundant computing power of each edge computing device in each sub-node to determine the function and computing capability of each device. Secondly, it receives the traffic status of the corresponding road intersection detected by each edge computing device in each sub-node. The traffic status of the road intersection is determined by the edge computing device at the intersection after calculating all received roadside sensing data based on its own computing power. The roadside sensing data is generated by the roadside sensing devices at the intersection. Finally, the acquired and received data are summarized. If the traffic status of the target intersection is abnormal, the computing power of multiple edge computing devices at the target intersection is scheduled based on the functional status and redundant computing power of all edge computing devices. This scheme, by scheduling computing power among edge computing devices based on their functional status and redundant computing power, can achieve a balanced allocation of computing power among the edge computing devices in the system.

[0039] To illustrate the technical solution of this application, specific embodiments are described below.

[0040] It should be understood that the sequence number of each step in this embodiment does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this application embodiment.

[0041] Reference Figure 1 This illustrates a schematic diagram of the structure of a child node provided in Embodiment 1 of this application. Figure 1As shown, the sub-nodes configured at any road intersection can include two edge computing devices (MECs) and eight road test devices surrounding the road intersection. The eight road test devices are located in four different directions, with two road test devices placed in each direction. In the original configuration (i.e., the original configuration of each edge computing device before scheduling), each MEC is responsible for the data processing of four road test devices.

[0042] Here, a child node can refer to a child node in a system, such as a child node in a computing power scheduling system. The computing power scheduling system can include a server and multiple child nodes. The child nodes can be configured at road intersections to monitor the traffic status of the road intersections. The server can communicate with the edge computing devices in the child nodes through the switches in each child node to obtain the traffic status of the road intersections monitored by the child nodes.

[0043] In this embodiment of the application, the computing power of each edge computing device can support the operation of more than 8 road test devices. Therefore, the redundant computing power of each edge computing device can be obtained based on the number of road test devices it is responsible for and the total computing power of each edge computing device, so as to schedule computing power based on the redundant computing power.

[0044] In this system, the edge computing devices in each sub-node are connected to the drive testing devices via a switch. The edge computing devices receive drive testing data generated by the drive testing devices through the switch, process the drive testing data, and then send the processed data to the electronic devices in the computing power scheduling system through the switch for data processing and analysis. The electronic devices can refer to servers. The servers process and analyze the data, including scheduling computing power for each edge computing device based on the received data, the functional status of each edge computing device, and the redundant computing power.

[0045] In the embodiments of this application, each edge computing device typically has a fixed computing task. Under normal circumstances, the computing power among edge computing devices is balanced. However, if an edge computing device detects that the traffic status of the target intersection is abnormal and needs to confirm the abnormal status, in order to avoid consuming too much computing power of the edge computing devices, it is necessary to schedule the computing power of multiple edge computing devices at the target intersection. For example, the road test perception data of multiple road test devices that can detect the abnormal scene at the target intersection can be transmitted to the same edge computing device to facilitate a more detailed confirmation of the abnormal scene, while also avoiding consuming the computing power of other edge computing devices.

[0046] Reference Figure 2 This diagram illustrates a flowchart of a computing power scheduling method provided in Embodiment 2 of this application. The computing power scheduling method proposed in this embodiment can be applied to applications involving... Figure 1 In electronic devices connected by neutron nodes, such as servers, Figure 2 As shown, the computing power scheduling method may include the following steps:

[0047] Step 201: Obtain the functional status and redundant computing power of each edge computing device in each sub-node.

[0048] Each sub-node can refer to a sub-node corresponding to each road intersection. A sub-node includes multiple edge computing devices and multiple road testing devices surrounding the road intersection. The sub-nodes transmit data to the server through switches connecting the edge computing devices and the road testing devices. The sub-nodes are also connected to each other through switches. It should be noted that when monitoring the traffic status of road intersections, the devices included in the sub-nodes can be installed at the road intersections. If monitoring the traffic status of other roads, the installation location of the devices can be selected according to the actual application scenario; this application does not impose any restrictions on this.

[0049] Edge computing devices refer to computing devices installed on the roadside to process roadside sensing data generated by roadside equipment to determine the traffic status of corresponding road intersections. It should be noted that edge computing devices have limited computing power. Exceeding the maximum computing power of an edge computing device may result in data not being processed in a timely manner and may cause the device to malfunction due to excessive computational burden.

[0050] The functional status of an edge computing device can refer to the usage status of each function of the edge computing device. Its functional status can include normal status and fault status. Under normal status, it means that the edge computing device can operate normally, and under fault status, it means that the edge computing device cannot operate normally.

[0051] The redundant computing power of an edge computing device can refer to the remaining computing power that the edge computing device can still perform. That is, the difference between the maximum computing power of the edge device and the computing power already used can be called the redundant computing power of the edge computing device. For example, if an edge computing device can process road test perception data transmitted by up to 8 road test devices, and has already received road test perception data from 4 road test devices, then the edge computing device can receive road test perception data from up to 4 road test devices and process the data. The road test perception data from these 4 road test devices is the redundant computing power of the edge computing device.

[0052] In this embodiment, since edge computing devices cannot receive redundant road test sensing data when in a faulty state or with insufficient computing power, the server needs to obtain the functional status and redundant computing power of each edge computing device in each sub-node before scheduling computing power for each edge computing device. Specifically, the functional status and redundant computing power of the edge computing devices can be determined by detecting the computing functions of the edge computing devices and the number of road test devices corresponding to the road test sensing data received by the edge computing devices.

[0053] In one possible implementation, the functional status and redundant computing power of each edge computing device in each child node are obtained, including:

[0054] Based on the heartbeat connection between each edge computing device and the server, the functional status of each edge computing device in each sub-node is obtained;

[0055] Based on the number of road test devices corresponding to the road test perception data received by each edge computing device and the maximum computing power of each edge computing device, the redundant computing power of each edge computing device is obtained.

[0056] In this embodiment of the application, the status of the heartbeat connection can monitor whether the link between the two devices that are making the heartbeat connection is connected. If the connection status is abnormal, the edge computing device and the server will no longer be able to communicate. Therefore, based on the heartbeat connection between each edge computing device and the server, the functional status of each edge computing device in each sub-node can be obtained. It can be understood that if the edge computing device cannot send data to the server, it is determined that the edge computing device has failed.

[0057] In this embodiment, the computing power of the edge computing device is usually determined based on the number of road test devices it is responsible for. The responsible road test devices are those responsible for processing the road test perception data of the road test devices. Since the maximum computing power of the edge computing device is limited, the redundant computing power of the edge computing device can be determined by the maximum computing power of the edge computing device and the number of road test devices corresponding to the road test perception data received by the edge computing device.

[0058] For example, suppose an edge computing device can process road test sensing data transmitted from up to eight road test devices, such as Figure 1 Of the eight road test devices, this road test device has already received road test perception data from four of the road test devices, i.e. Figure 1 If we consider road test devices 1, 2, 3, and 4, then the edge computing device can receive road test perception data from a maximum of 4 road test devices and process the data. Therefore, the redundant computing power of the edge computing device is the ability to process the road test perception data of 4 road test devices.

[0059] In one possible implementation, the functional status of each edge computing device in each child node is obtained based on the heartbeat connection between each edge computing device and the server, including:

[0060] If the heartbeat connection status between the edge computing device and the server is successful, then the functional status of the edge computing device is determined to be normal.

[0061] If the heartbeat connection status between the edge computing device and the server is "connection failed", then the functional status of the edge computing device is determined to be "faulty".

[0062] In this embodiment, the server can send a command to the edge computing device to establish a heartbeat connection. If the edge computing device responds to the command and sends feedback to the server, it indicates that the heartbeat connection between the edge computing device and the server is successful, and the edge computing device can work normally. If the edge computing device fails to respond to the command, that is, the server does not receive feedback from the edge computing device within a specified time period, it indicates that the heartbeat connection between the edge computing device and the server is failed, and the current functional state of the edge computing device is a fault state. In this case, to ensure that the server can obtain comprehensive information about the intersection, the drive test equipment handled by the faulty edge computing device should be distributed to other edge computing devices in normal state in the sub-node.

[0063] To ensure the server can accurately obtain comprehensive information about the intersection even if any edge computing device fails, one possible implementation of the computing power scheduling method includes:

[0064] For any child node, if the functional status of an edge computing device is in a fault state, the original road test device corresponding to that edge computing device will be assigned to other edge computing devices in the child node so that the other edge computing devices can process the road test perception data sent by the original road test device.

[0065] In this embodiment of the application, since the edge computing device in a faulty state can no longer report the traffic status of the road intersection to the server, the traffic status of the road intersection obtained by the server is missing. This may result in a situation where an accident occurs at the road intersection, but the server does not obtain the accident information. Therefore, once the edge computing device fails, the original road test device corresponding to the edge computing device should be immediately assigned to other edge computing devices in the child node.

[0066] For example, such as Figure 3The diagram shows the computing power scheduling of the fault edge computing devices. Drive test devices 1, 2, 3, and 4 are all managed by MEC1, while drive test devices 5, 6, 7, and 8 are managed by MEC2. When MEC1 fails, the server sends a command to the switch to disconnect the communication links between drive test devices 1, 2, 3, and 4 and MEC1. Then, it establishes communication links between drive test devices 1, 2, 3, and 4 and MEC2, and assigns drive test devices 1, 2, 3, and 4 to MEC2, meaning that MEC2 takes over the computation. It should be noted that before allocating to MEC2, the redundant computing power of MEC2 needs to be determined. If the redundant computing power of MEC2 can support the road test perception data of 4 road test devices, it can be directly allocated to MEC2. If the redundant computing power of MEC2 cannot support the road test perception data of 4 road test devices, after the redundant computing power of MEC2 is exhausted, the unallocated road test devices will be allocated to edge computing devices in other child nodes that are in normal functional status. The allocation will be carried out according to this rule until the allocation is completed.

[0067] Step 202: Receive the traffic status of the corresponding road intersection monitored by each edge computing device in each sub-node.

[0068] In this embodiment, each edge computing device in each sub-node detects the traffic status of the corresponding road intersection and sends it to the server. The traffic status of the road intersection is determined by the edge computing device at the intersection after calculating all received roadside sensing data based on its own computing power. This roadside sensing data is generated by the roadside sensing devices at the road intersection.

[0069] In one possible implementation, the road testing equipment includes at least one of a camera, millimeter-wave radar, and lidar; the road testing perception data includes traffic participant status data at road intersections and image data of road intersections; the traffic participant status data includes at least one of real-time vehicle driving status data, real-time pedestrian driving status data, and environmental data of road intersections; the real-time vehicle driving status data includes at least one of real-time speed, real-time position, and real-time heading angle.

[0070] In this embodiment of the application, a road intersection can be equipped with various types of road testing devices, including cameras that can capture images of the road intersection, millimeter-wave radar and lidar that can acquire the real-time speed, real-time position and real-time heading angle of vehicles, and edge computing devices that, after acquiring the above-mentioned road testing perception data, can determine the traffic status of the road intersection based on the multi-directional data.

[0071] For example, based on the vehicle's real-time driving status data, it can be determined whether an abnormality has occurred at the road intersection. If the real-time speed of all vehicles is within the preset range and their real-time positions are constantly changing, the traffic status of the road intersection can be determined to be normal. However, if the real-time speed of the vehicles is no longer within the preset range, such as if the real-time speed has not changed for a long time, then there may be congestion or traffic accidents at the road intersection. Congestion or traffic accidents are both determined to be abnormal traffic status at the road intersection.

[0072] Step 203: If the traffic status of the target intersection is abnormal, then the computing power of multiple edge computing devices at the target intersection is scheduled according to the functional status and redundant computing power of all edge computing devices.

[0073] In this embodiment, when the traffic status of the target intersection is abnormal, it is necessary to confirm the abnormality of the target intersection. If multiple road test devices that can detect the accident transmit data to different edge computing devices, there is a problem of low accuracy in the abnormality confirmation. Therefore, in this case, computing power scheduling among edge computing devices can be performed based on the functional status and redundant computing power of all edge computing devices. That is, when the server receives a message that the traffic status of the target intersection is abnormal, it will send a computing power scheduling instruction to the switch.

[0074] In one possible implementation, when the traffic flow at the target intersection is abnormal, the scheduling of computing power under abnormal conditions may include, for example: Figure 4 The steps shown are as follows:

[0075] Step 401: If the traffic status of the target intersection is abnormal, obtain the location coordinates of the location where the abnormal event occurred at the target intersection.

[0076] In this embodiment, since the road testing device can be a camera, the location coordinates of the abnormal event at the target intersection can be obtained from images of the target intersection captured by at least one camera. These location coordinates are in the world coordinate system.

[0077] In one possible implementation, if the traffic status of the target intersection is abnormal, the location coordinates of the location where the abnormal event occurred at the target intersection are obtained, including:

[0078] If the traffic status of the target intersection is abnormal, a positioning command is sent to the first edge computing device monitoring the target intersection.

[0079] Control at least two cameras located in different positions to transmit the perceived target intersection image data to a first edge computing device;

[0080] The system receives target intersection image data generated by at least two cameras located in different directions from the first edge computing device, and obtains the location coordinates of the abnormal event at the target intersection based on the target intersection image data from different directions.

[0081] The first edge computing device is any one of the multiple edge computing devices at the target intersection.

[0082] In this embodiment of the application, the server can send a positioning command to the first edge computing device at the target intersection through a switch. The roadside sensing data received by the first edge computing device is roadside sensing data located in different directions. Therefore, the positioning command sent by the server can enable the switch to control at least two cameras located in different directions to transmit the target intersection image data generated by sensing to the first edge computing device, so as to obtain the location coordinates of the abnormal event at the target intersection more accurately through the target intersection image data from different directions.

[0083] For example, such as Figure 5 The diagram shown illustrates the location coordinates of the abnormal event. A binocular positioning algorithm can be used. The MEC1 acquires image data captured by road test device 1 and road test device 2, and sends the image data captured by road test device 1 and road test device 2 to the server. The server locates the abnormal event and determines its location coordinates.

[0084] For example, such as Figure 6 Another schematic diagram showing the location coordinates of the abnormal event can be obtained from multiple directions. For example, MEC1 can acquire image data captured by road test device 1, road test device 2 and road test device 3, and send the image data to the server. The server can locate the location of the abnormal event based on the image data from different directions and determine the location coordinates of the abnormal event.

[0085] Step 402: Compare the location coordinates of the abnormal event at the target intersection with the preset detection range of each road testing device at the target intersection, and determine the target road testing device at the target intersection based on the comparison results.

[0086] In this embodiment of the application, since not all road test devices at the target intersection can detect the abnormal event when it occurs, in order for the server to obtain data on the abnormal event more centrally, the target road test devices at the target intersection that can detect the abnormal event can be identified.

[0087] In one possible implementation, the location coordinates of the abnormal event at the target intersection are compared with the preset detection ranges of each road testing device at the target intersection, and based on the comparison results, the target road testing devices at the target intersection are determined, including:

[0088] Compare the location coordinates of the abnormal event at the target intersection with the preset detection range of each road testing device at the target intersection;

[0089] If the location coordinates of an abnormal event at the target intersection are within the preset detection range of any road testing device at the target intersection, then that road testing device is identified as the target road testing device.

[0090] In this embodiment, the predicted detection range of each road testing device at the target intersection is pre-stored in the server. Therefore, it can be directly invoked when determining the target road testing device. Based on the location coordinates of the abnormal event determined above and the preset detection range of each road testing device at the target intersection, the road testing device whose preset detection range includes the location coordinates can be determined as the target road testing device. For example, according to Figure 7 The schematic diagram of the target road test equipment shown in the figure can be used to identify road test equipment 1 and road test equipment 4 as the target road test equipment.

[0091] Step 403: Based on the functional status and redundant computing power of all edge computing devices, transmit the target road test perception data generated by the target road test device at the target intersection to the target edge computing device.

[0092] The target edge computing device can refer to an edge computing device in a normal functional state. It should be noted that this target edge computing device can be an edge computing device in a normal functional state within a sub-node corresponding to the target intersection, or it can be an edge computing device in a normal functional state within a sub-node corresponding to other intersections. The specific determination can be based on the functional state and redundant computing power of the edge computing device.

[0093] There are several ways to identify target edge computing devices. As one possible approach, identifying target edge computing devices can include:

[0094] If there is an edge computing device in the target sub-node corresponding to the target intersection whose functional status is normal, then select any edge computing device in the target sub-node whose functional status is normal as the target edge computing device.

[0095] If all edge computing devices in the target sub-nodes corresponding to the target intersection are in a fault state, then select any edge computing device in a normal state from the sub-nodes corresponding to other intersections as the target edge computing device.

[0096] For example, in practical applications, the maximum computing power of the edge computing device at each road intersection is usually set to be able to process the road-tested perception data of all road-testing devices at the intersection. This is because when determining the target edge computing device, the principle of proximity is typically followed. If the traffic status of the target intersection is abnormal, any edge computing device with a normal functional state is selected from the target sub-nodes corresponding to the target intersection as the target edge computing device. However, if all edge computing devices in the target sub-node malfunction, such as... Figure 8 As shown, at this point, any edge computing device whose functional status is normal should be selected from the child nodes corresponding to other intersections as the target edge computing device, for example... Figure 8 The MEC3 was identified as the target edge computing device, and the road test perception data of road test devices 1 to 8 were all transmitted to the target edge computing device (it is known that road test devices 1 to 8 are all target road test devices).

[0097] In one possible implementation, the target road perception data generated by the target road perception device at the target intersection can be transmitted to the target edge computing device, and the road perception data generated by other road perception devices at the target intersection can be transmitted to other edge computing devices at the target intersection besides the target edge computing device.

[0098] To determine the severity of an anomaly in an abnormal event, one possible implementation of the computing power scheduling method includes:

[0099] In response to the user's confirmation of the degree of abnormality of the abnormal state, it receives multi-directional image data sent by the target edge computing device in the target intersection;

[0100] Target recognition and image inpainting are performed on multi-directional image data to determine the degree of abnormality in the traffic status of the target intersection.

[0101] In this embodiment of the application, for cases where the image data captured by a single road test device is relatively blurry, at least two road test devices can be controlled to send the captured image data to the target edge computing device. Then, the target edge device sends the multi-directional image data to the server. In the case of blurry image data, the server can use target recognition algorithms and image restoration algorithms to perform target recognition and image restoration on the multi-directional image data to determine the degree of abnormality of the traffic status of the target intersection.

[0102] In one possible implementation, the server can use an image restoration algorithm to restore the image, and after restoration, use a target recognition algorithm to identify the location of the abnormal event at the target intersection in order to determine the degree of abnormality in the traffic status of the target intersection.

[0103] It should be understood that the target recognition algorithm can be any target recognition algorithm in the prior art, such as R-CNN (Region with CNN Feature), YOLO (You Only Look Once), etc., and the embodiments of this application do not limit it.

[0104] It should also be understood that the image restoration algorithm can be any image restoration algorithm in the prior art, such as denoising, super-resolution reconstruction, inpainting, deblurring, JPEG deblocking, etc., and the embodiments of this application do not limit it.

[0105] To determine the trajectories of traffic participants at the target intersection after an abnormal event occurs, one possible implementation of the computing power scheduling method includes:

[0106] In response to the user's operation to obtain the vehicle trajectory of the target intersection, it receives road test perception data sent by all edge computing devices in the target intersection;

[0107] Based on road test data, determine the trajectory of vehicles passing through the target intersection;

[0108] Based on the vehicle's operating trajectory, predict the vehicle's future operating trajectory.

[0109] For example, a user can manually obtain vehicle trajectories at a target intersection. In response to this trajectory acquisition operation, the server can send a data acquisition command to the switch in the target sub-node corresponding to the target intersection, receive roadside sensing data sent by all edge computing devices at the target intersection, extract vehicle-related data from the roadside sensing data, determine the vehicle's trajectory based on the extracted data, and predict the vehicle's future trajectory based on the determined trajectory. Figure 9 The diagram shown illustrates the vehicle's trajectory. In one possible implementation, the vehicle's trajectory can be displayed using a 3D digital twin platform, allowing users to intuitively view the vehicle's trajectory and future trajectory.

[0110] To distribute computing power evenly across all edge computing devices in child nodes, when adding new edge computing devices, the computing power scheduling method may further include, as a possible implementation:

[0111] If a new edge computing device is detected in a child node, the road test perception data originally processed by the edge computing device in the child node will be allocated to the new edge computing device for processing based on the computing power of all edge computing devices in the child node.

[0112] In this embodiment of the application, the drive test devices that each edge computing device in the child node can be assigned to each other on an average basis.

[0113] It should be understood that each child node includes a switch connected to the server. One end of the switch is connected to the drive test equipment, and the other end is connected to the edge computing device. The switch is used to receive instructions from the server to schedule the computing power of the edge computing device.

[0114] In this embodiment, each road intersection is configured with sub-nodes. The server first obtains the functional status and redundant computing power of each edge computing device in each sub-node to determine the function and computing power of each edge computing device. Secondly, it receives the traffic status of the corresponding road intersection detected by each edge computing device in each sub-node. The traffic status of the road intersection is determined by the edge computing device at the road intersection after calculating all received road-testing perception data based on its own computing power. The road-testing perception data is generated by the road-testing devices at the road intersection. Finally, the obtained and received data are summarized. If the traffic status of the target intersection is abnormal, the computing power of multiple edge computing devices at the target intersection is scheduled according to the functional status and redundant computing power of all edge computing devices. Therefore, this embodiment uses the functional status and redundant computing power of edge computing devices to schedule computing power among them, achieving a balanced allocation of computing power among the edge computing devices in the system.

[0115] See Figure 10 The diagram shows a schematic representation of the computing power scheduling device provided in Embodiment 3 of this application. For ease of explanation, only the parts relevant to the embodiments of this application are shown.

[0116] The computing power scheduling device 10 may specifically include the following modules:

[0117] The acquisition module 1001 is used to acquire the functional status and redundant computing power of each edge computing device in each sub-node;

[0118] The receiving module 1002 is used to receive the traffic status of the corresponding road intersection monitored by each edge computing device in each sub-node. The traffic status of the road intersection is determined by the edge computing device of the road intersection after calculating all the received road test perception data according to its own computing power. The road test perception data is generated by the road test device of the road intersection.

[0119] The scheduling module 1003 is used to schedule the computing power of multiple edge computing devices at the target intersection based on the functional status and redundant computing power of all edge computing devices if the traffic status of the target intersection is abnormal.

[0120] In this embodiment of the application, the acquisition module 1001 may specifically include the following sub-modules:

[0121] The status acquisition submodule is used to acquire the functional status of each edge computing device in each sub-node based on the heartbeat connection between each edge computing device and the server.

[0122] The computing power acquisition submodule is used to acquire the redundant computing power of each edge computing device based on the number of road test devices corresponding to the road test perception data received by each edge computing device and the maximum computing power of each edge computing device.

[0123] In this embodiment of the application, the status acquisition submodule may specifically include the following units:

[0124] The normal state determination unit is used to determine the functional state of the edge computing device as normal if the heartbeat connection status between the edge computing device and the server is successful.

[0125] The fault status determination unit is used to determine the functional status of the edge computing device as faulty if the heartbeat connection status between the edge computing device and the server is a connection failure.

[0126] In this embodiment of the application, the computing power scheduling device may further include the following modules:

[0127] The device allocation module is used to allocate the original road test device corresponding to any edge computing device in any child node to other edge computing devices in the child node if the functional status of the edge computing device is in a fault state, so that the other edge computing devices can process the road test perception data sent by the original road test device.

[0128] In this embodiment of the application, the scheduling module 1003 may specifically include the following sub-modules:

[0129] The coordinate acquisition submodule is used to obtain the location coordinates of the abnormal event at the target intersection if the traffic status of the target intersection is abnormal.

[0130] The target determination submodule is used to compare the location coordinates of the abnormal event at the target intersection with the preset detection range of each road test device at the target intersection, and determine the target road test device at the target intersection based on the comparison results.

[0131] The data transmission submodule is used to transmit the target road test perception data generated by the target road test device at the target intersection to the target edge computing device, which is an edge computing device in a normal functional state, based on the functional status and redundant computing power of all edge computing devices.

[0132] In this embodiment of the application, the coordinate acquisition submodule may specifically include the following units:

[0133] The positioning instruction sending unit is used to send a positioning instruction to the first edge computing device monitoring the target intersection if the traffic status of the target intersection is abnormal. The first edge computing device is any one of the multiple edge computing devices at the target intersection.

[0134] A control unit is used to control at least two cameras located in different positions to transmit the perceived target intersection image data to a first edge computing device;

[0135] The image data receiving unit is used to receive target intersection image data generated by at least two cameras located in different directions sent by the first edge computing device, and to obtain the location coordinates of the location where the abnormal event of the target intersection occurs based on the target intersection image data from different directions.

[0136] In this embodiment of the application, the data transmission submodule may specifically include the following units:

[0137] The second device determination unit is used to select any edge computing device with a normal functional state from the target sub-node if there is an edge computing device with a normal functional state in the target sub-node corresponding to the target intersection as the target edge computing device.

[0138] The third device determination unit is used to select any edge computing device with a normal function from the sub-nodes corresponding to other intersections as the target edge computing device if the functional status of all edge computing devices in the target sub-nodes corresponding to the target intersection is in a fault state.

[0139] The perception data transmission unit is used to transmit the target road perception data generated by the target road test device at the target intersection to the target edge computing device, and to transmit the road perception data generated by other road test devices at the target intersection to other edge computing devices at the target intersection other than the target edge computing device.

[0140] In this embodiment of the application, the computing power scheduling device may further include the following modules:

[0141] The first response module is used to respond to the user's confirmation operation on the degree of abnormality of the abnormal state, and to receive multi-directional image data sent by the target edge computing device in the target intersection. The multi-directional image data is collected by at least two road test devices located in multiple directions and then sent to the target edge computing device.

[0142] The anomaly determination module is used to perform target recognition and image repair on multi-directional image data in order to determine the degree of anomaly in the traffic status of the target intersection.

[0143] In this embodiment of the application, the computing power scheduling device may further include the following modules:

[0144] The second response module is used to respond to the user's operation to obtain the vehicle trajectory of the target intersection and to receive the road test perception data sent by all edge computing devices in the target intersection.

[0145] The first trajectory determination module is used to determine the trajectory of vehicles passing through the target intersection based on road test perception data.

[0146] The second trajectory determination module is used to predict the future trajectory of the vehicle based on its existing trajectory.

[0147] In this embodiment of the application, the computing power scheduling device may further include the following modules:

[0148] The data allocation module is used to allocate the road test perception data originally processed by the edge computing devices in the child node to the newly added edge computing devices for processing, based on the computing power of all edge computing devices in the child node.

[0149] The computing power scheduling device provided in this application embodiment can be applied in the aforementioned computing power scheduling method embodiment. For details, please refer to the description of the above computing power scheduling method embodiment, which will not be repeated here.

[0150] Reference Figure 11 The diagram illustrates the structure of the electronic device provided in Embodiment 4 of this application. Figure 11 As shown, the electronic device 1100 of this embodiment includes: at least one processor 1110 ( Figure 11 (Only one is shown in the diagram) a processor, a memory 1120, and a computer program 1121 stored in the memory 1120 and executable on the at least one processor 1110. When the processor 1110 executes the computer program 1121, it implements the steps in the above-described computing power scheduling method embodiment.

[0151] The electronic device 1100 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. This electronic device may include, but is not limited to, a processor 1110 and a memory 1120. Those skilled in the art will understand that... Figure 11 This is merely an example of electronic device 1100 and does not constitute a limitation on electronic device 1100. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0152] The processor 1110 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), graphics processing units (GPUs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0153] In some embodiments, the memory 1120 may be an internal storage unit of the electronic device 1100, such as a hard disk or memory of the electronic device 1100. In other embodiments, the memory 1120 may be an external storage device of the electronic device 1100, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 1100. Furthermore, the memory 1120 may include both internal and external storage units of the electronic device 1100. The memory 1120 is used to store operating systems, applications, boot loaders, data, and other programs, such as the program code of computer programs. The memory 1120 can also be used to temporarily store data that has been output or will be output.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0155] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0156] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0157] In the embodiments provided in this application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings or direct couplings or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0158] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0159] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0160] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to requirements; for example, the computer-readable medium may not include electrical carrier signals and telecommunication signals.

[0161] The implementation of all or part of the processes in the methods of the above embodiments can also be accomplished by a computer program product. When the computer program product is run on an electronic device, the electronic device can implement the steps in the various method embodiments described above.

[0162] The embodiments described above are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A computing power scheduling method, characterized in that, The computing power scheduling method includes: Obtain the functional status and redundant computing power of each edge computing device in each sub-node; where each sub-node corresponds to a sub-node at each road intersection; The system receives traffic status data from each edge computing device in each sub-node, which is monitored by the corresponding road intersection. The traffic status of the road intersection is determined by the edge computing device at the road intersection after calculating all the received road test data based on its own computing power. The road test data is generated by the road test device at the road intersection. If the traffic status of the target intersection is abnormal, then obtain the location coordinates of the location where the abnormal event occurred at the target intersection; The location coordinates of the abnormal event at the target intersection are compared with the preset detection range of each road testing device at the target intersection, and the target road testing device at the target intersection is determined based on the comparison results. Based on the functional status and redundant computing power of all edge computing devices, the target road test perception data generated by the target road test device at the target intersection is transmitted to the target edge computing device, which is the edge computing device whose functional status is normal.

2. The computing power scheduling method as described in claim 1, characterized in that, The process of obtaining the functional status and redundant computing power of each edge computing device in each sub-node includes: Based on the heartbeat connection between each edge computing device and the server, the functional status of each edge computing device in each sub-node is obtained; Based on the number of road test devices corresponding to the road test perception data received by each edge computing device and the maximum computing power of each edge computing device, the redundant computing power of each edge computing device is obtained.

3. The computing power scheduling method as described in claim 2, characterized in that, The process of obtaining the functional status of each edge computing device in each sub-node based on the heartbeat connection between each edge computing device and the server includes: If the heartbeat connection status between the edge computing device and the server is successful, then the functional status of the edge computing device is determined to be normal. If the heartbeat connection between the edge computing device and the server is in the state of connection failure, then the functional state of the edge computing device is determined to be a fault state.

4. The computing power scheduling method as described in claim 3, characterized in that, The computing power scheduling method further includes: For any of the aforementioned sub-nodes, if the functional state of the edge computing device is in a fault state, the original road test device corresponding to the edge computing device is assigned to other edge computing devices in the sub-node, so that the other edge computing devices can process the road test perception data sent by the original road test device.

5. The computing power scheduling method as described in claim 1, characterized in that, If the traffic status of the target intersection is abnormal, then the location coordinates of the location where the abnormal event occurred at the target intersection are obtained, including: If the traffic status of the target intersection is abnormal, a positioning command is sent to the first edge computing device monitoring the target intersection. The first edge computing device is any one of the multiple edge computing devices at the target intersection. Control at least two cameras located in different positions to transmit the perceived target intersection image data to the first edge computing device; The system receives target intersection image data generated by at least two cameras located in different directions, sent by the first edge computing device, and obtains the location coordinates of the location where the abnormal event at the target intersection occurred based on the target intersection image data from different directions.

6. The computing power scheduling method as described in claim 1, characterized in that, The step of transmitting the target roadside perception data generated by the target roadside device at the target intersection to the target edge computing device based on the functional status and redundant computing power of all edge computing devices includes: If there is an edge computing device in the target sub-node corresponding to the target intersection whose functional state is normal, then any edge computing device in the target sub-node whose functional state is normal is selected as the target edge computing device. If the functional status of all edge computing devices in the target sub-nodes corresponding to the target intersection is faulty, then any edge computing device whose functional status is normal is selected from the sub-nodes corresponding to other intersections as the target edge computing device. The target road perception data generated by the target road perception device at the target intersection is transmitted to the target edge computing device, and the road perception data generated by other road perception devices at the target intersection is transmitted to other edge computing devices at the target intersection other than the target edge computing device.

7. The computing power scheduling method as described in claim 1, characterized in that, The computing power scheduling method further includes: In response to the user's confirmation operation on the degree of abnormality of the abnormal state, the system receives multi-directional image data sent by the target edge computing device in the target intersection. The multi-directional image data is collected by at least two road test devices located in multiple directions and then sent to the target edge computing device. Target recognition and image restoration are performed on the multi-directional image data to determine the degree of abnormality in the traffic status of the target intersection.

8. The computing power scheduling method as described in claim 1, characterized in that, The computing power scheduling method further includes: In response to the user's operation to obtain the vehicle trajectory of the target intersection, the system receives road test perception data sent by all edge computing devices in the target intersection. Based on the road test perception data, the trajectory of the vehicle passing through the target intersection is determined; Based on the vehicle's operating trajectory, predict the vehicle's future operating trajectory.

9. The computing power scheduling method as described in claim 1, characterized in that, The computing power scheduling method further includes: If a new edge computing device is detected in the sub-node, the road test perception data originally processed by the edge computing device in the sub-node will be allocated to the new edge computing device for processing based on the computing power of all edge computing devices in the sub-node.

10. A computing power scheduling device, characterized in that, The computing power scheduling device includes: The acquisition module is used to acquire the functional status and redundant computing power of each edge computing device in each sub-node; where each sub-node corresponds to a sub-node at each road intersection. The receiving module is used to receive the traffic status of the corresponding road intersection monitored by each edge computing device in each sub-node. The traffic status of the road intersection is determined by the edge computing device of the road intersection after calculating all the received road test perception data according to its own computing power. The road test perception data is generated by the road test device of the road intersection. The scheduling module is used to: if the traffic status of the target intersection is abnormal, obtain the location coordinates of the location where the abnormal event occurred at the target intersection; compare the location coordinates of the location where the abnormal event occurred at the target intersection with the preset detection range of each road testing device at the target intersection, and determine the target road testing device at the target intersection based on the comparison result; and transmit the target road testing perception data generated by the target road testing device at the target intersection to the target edge computing device based on the functional status and redundant computing power of all edge computing devices, wherein the target edge computing device is the edge computing device whose functional status is normal.

11. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the method as described in any one of claims 1 to 9.

12. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 9.

Images