Networking method and device of collaborative network and storage medium

By pre-generating and updating the collaborative network in the terminal equipment, the problems of low data transmission reliability and large delay caused by connection interruption in the vehicle-road collaborative network are solved, and more efficient data transmission is achieved.

CN118921640APending Publication Date: 2024-11-08TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310505864.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-06
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the vehicle-road collaboration network or drone collaboration network, network connections are interrupted due to the movement of terminal devices, resulting in low data transmission reliability and large delay.

Method used

Before the connection between the terminal device and other network nodes is interrupted, a local collaborative network is formed in advance. By obtaining the characteristic data of the terminal device and network nodes, a second local collaborative network is generated based on the neural network model or graph neural network, and Networking and deletion are carried out at different times of connection.

Benefits of technology

It improves the reliability of data transmission, reduces the delay in data transmission, avoids connection interruption, and realizes a smooth networking method.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118921640A_ABST
    Figure CN118921640A_ABST
Patent Text Reader

Abstract

The invention provides a networking method and device of a collaborative network and a storage medium. The invention relates to a vehicle-road cooperation technology, an unmanned aerial vehicle cooperation technology or an unmanned vehicle cooperation technology and the like in a smart traffic technology. The method comprises the following steps: when a first terminal is in a first local collaborative network and the connection of the first terminal is not interrupted, acquiring first feature data of the first terminal and second feature data sent by at least one network node; obtaining a second local collaborative network based on the first feature data and the second feature data; and when the first local collaborative network is different from the second local collaborative network, deleting the first local collaborative network, and networking the second local collaborative network. On one hand, the reliability of data transmission can be ensured, and on the other hand, the smooth networking mode can reduce the time delay of data transmission.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent transportation technology, and in particular to a collaborative network networking method, device and storage medium. Background Art

[0002] In collaborative networks such as vehicle-road collaborative networks or drone collaborative networks, since vehicles or drones are usually in a mobile state, the local collaborative network where they are located will undergo dynamic changes.

[0003] In the related art, when the connection between a vehicle or a drone and other network nodes is interrupted, the vehicle or drone is triggered to re-network.

[0004] However, the current networking mode is carried out under the condition of disconnection, which, on the one hand, leads to low data transmission reliability, and on the other hand, leads to large data transmission delay. Summary of the invention

[0005] The present application provides a collaborative network networking method, device and storage medium. On the one hand, the reliability of data transmission can be guaranteed, and on the other hand, this smooth networking method can reduce data transmission delay.

[0006] In a first aspect, an embodiment of the present application provides a method for networking a collaborative network, which is applied to a first terminal, and the method includes: when the first terminal is in a first local collaborative network and the connection of the first terminal is not interrupted, obtaining first characteristic data of the first terminal and second characteristic data sent by at least one network node; based on the first characteristic data and the second characteristic data, obtaining a second local collaborative network; when the first local collaborative network is different from the second local collaborative network, deleting the first local collaborative network and networking the second local collaborative network.

[0007] In a second aspect, an embodiment of the present application provides a terminal device, which is a first terminal, and the first terminal includes: a transceiver module and a processing module, wherein the transceiver module is used to obtain first characteristic data of the first terminal and second characteristic data sent by at least one network node when the first terminal is in a first local collaborative network and the connection of the first terminal is not interrupted; the processing module is used to: obtain a second local collaborative network based on the first characteristic data and the second characteristic data; when the first local collaborative network is different from the second local collaborative network, delete the first local collaborative network and network the second local collaborative network.

[0008] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, and executing the method in the first aspect or its various implementations.

[0009] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, wherein the computer program enables a computer to execute the method in the first aspect or its various implementations.

[0010] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising computer program instructions, which enable a computer to execute a method as in the first aspect or its various implementations.

[0011] In a sixth aspect, an embodiment of the present application provides a computer program, which enables a computer to execute the method in the first aspect or its various implementations.

[0012] Through the technical solution provided in the embodiment of the present application, since the first terminal can pre-network the second local collaboration before its connection is interrupted, in other words, this method can avoid connection interruption. On the one hand, the reliability of data transmission can be guaranteed. On the other hand, this smooth networking method can reduce data transmission delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of a vehicle-road cooperative network architecture provided in an embodiment of the present application;

[0014] Figure 2 A flowchart of a collaborative network networking method provided in an embodiment of the present application;

[0015] Figure 3 A schematic diagram of a target attribute graph and a second local collaborative network provided in an embodiment of the present application;

[0016] Figure 4 A schematic diagram of a terminal device 600 provided in an embodiment of the present application;

[0017] Figure 5 It is a schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0019] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0020] Before introducing the technical solution of the present application, the following first describes the relevant knowledge of the embodiments of the present application:

[0021] 1. Smart transportation: Based on intelligent transportation, it integrates high-tech technologies such as the Internet of Things, cloud computing, big data, and mobile Internet, collects traffic information through high-tech technologies, and provides traffic information services under real-time traffic data. It makes extensive use of data processing technologies such as data models and data mining to achieve the systematic, real-time, interactive information exchange, and extensive services of smart transportation.

[0022] 2. Digital Twin: It makes full use of physical models, sensor updates, operation history and other data, integrates multi-disciplinary, multi-physical quantity, multi-scale, and multi-probability simulation processes, completes mapping in virtual space, and thus reflects the entire life cycle of the corresponding physical objects.

[0023] 3. Vehicle-road collaborative network: It adopts advanced wireless communication and new generation Internet technologies to implement all-round dynamic real-time information interaction between vehicles and roads, and carries out vehicle active safety control and road collaborative management based on the collection and integration of dynamic traffic information in all time and space, fully realizing the effective coordination of people, vehicles and roads, ensuring traffic safety, and improving traffic efficiency, thus forming a safe, efficient and environmentally friendly road traffic system.

[0024] At present, the information perception and prediction tasks in traffic scenarios have gradually evolved from traditional single-node data collection and analysis to multi-node interconnected vehicle-road collaborative perception and analysis. Vehicle-road collaboration actually helps single-vehicle intelligent perception to achieve a God's perspective. It realizes digital twins through the full digitization of road infrastructure to make up for the lack of perception caused by adding sensors to the vehicle side alone. Through networking and sharing of information between vehicles and vehicles and vehicles and roads, the vehicle-road collaborative network can realize blind spot supplementation and improve perception accuracy. The importance of vehicle-road collaboration for the high-level autonomous driving has gradually become an industry consensus, and it has become an important part of smart transportation planning. It has huge application value in scenarios such as smart transportation, autonomous driving, and digital twins.

[0025] 4. Point Cloud: Also known as point cloud data, a point cloud is a set of irregularly distributed discrete points in space that express the spatial structure and surface properties of a three-dimensional object or scene. Point cloud refers to the geometry of a large number of three-dimensional points. Each point in the point cloud has at least three-dimensional position information. Depending on the application scenario, it may also have additional attributes such as color, material or other information such as reflectivity. Usually, each point in the point cloud has the same number of additional attributes. For example, the point cloud obtained according to the principle of laser measurement includes three-dimensional coordinates (XYZ) and laser reflection intensity (reflectance); the point cloud obtained according to the principle of photogrammetry includes three-dimensional coordinates (XYZ) and color information, such as red, green and blue (Red Green Blue, RGB); the point cloud obtained by combining laser measurement and photogrammetry principles includes three-dimensional coordinates (XYZ), laser reflection intensity (reflectance) and color information (RGB).

[0026] Among them, point clouds can be divided into two categories according to their uses: machine-perceived point clouds, such as those used in autonomous navigation systems, real-time inspection systems, geographic information systems, visual sorting robots, emergency rescue robots and other scenarios; human-eye-perceived point clouds, such as those used in digital cultural heritage, free viewpoint broadcasting, three-dimensional immersive communication, three-dimensional immersive interaction and other point cloud application scenarios.

[0027] Among them, point clouds can be classified according to the way they are acquired: the first type is static point cloud: the object is stationary and the device for acquiring the point cloud is also stationary; the second type is dynamic point cloud: the object is moving, but the device for acquiring the point cloud is stationary; the third type is dynamically acquired point cloud: the device for acquiring the point cloud is moving.

[0028] 5. Point Cloud Compression (PCC): Point cloud is a collection of massive points. Storing these point cloud data will not only consume a lot of memory, but also be inconvenient for transmission. The relevant technology does not have such a large bandwidth to support direct transmission of point clouds at the network layer without compression. Therefore, it is necessary to compress the point cloud.

[0029] 7. Geometry-based Point Cloud Compression (G-PCC), which compresses the first type of static point cloud and the third type of dynamically acquired point cloud. The corresponding point cloud media can be called point cloud media compressed based on geometry features, abbreviated as G-PCC point cloud media.

[0030] 8. Video-based Point Cloud Compression (V-PCC) based on traditional video coding. V-PCC compresses the second type of dynamic point cloud, and the corresponding point cloud media obtained can be called point cloud media compressed based on traditional video coding, referred to as V-PCC point cloud media.

[0031] 9. Heterogeneous graph: It means that there is more than one type of node or relationship in the graph.

[0032] 10. Attribute graph: Compared with heterogeneous graph, attribute graph adds additional attribute information to graph data, which can be nodes or edges in the graph.

[0033] 11. Graph Neural Network (GNN) is an optimizable transformation of all attributes on a graph, which can preserve the symmetric information of the graph, namely permutation invariances. GNN is a "graph-in, graph-out" structure that transforms the attributes and / or edges of the input graph, but does not change its connectivity.

[0034] 12. Convex optimization: Under the optimization requirements of minimization or maximization, the objective function is a convex function and the feasible domain set formed by the constraints is an optimization method that is a convex set. Therefore, there are two judgment conditions for convex optimization: the function domain is a convex set and the objective function is a convex function.

[0035] The technical problems to be solved by the technical solution of this application and the inventive concept are described below:

[0036] As mentioned above, the current networking mode is carried out under the condition of disconnection, which, on the one hand, leads to low data transmission reliability, and on the other hand, leads to large data transmission delay.

[0037] In order to solve this technical problem, an embodiment of the present application proposes that before the connection between the terminal device and other network nodes is interrupted, the terminal device can pre-establish a local collaborative network.

[0038] In some possible implementations, the cooperative network provided in the embodiments of the present application may be a vehicle-road cooperative network, a multi-drone cooperative network, a multi-unmanned vehicle cooperative network, etc., but not limited thereto. Based on this, the terminal device in the cooperative network may be a vehicle-mounted terminal, an airborne terminal, etc., but not limited thereto.

[0039] In some implementations, the local collaborative network provided in the embodiments of the present application may include: at least one vehicle, drone or unmanned vehicle, a hub device and an access network device, but is not limited thereto. In other words, the local collaborative network may include: at least one terminal device, a hub device and an access network device, but is not limited thereto.

[0040] Among them, any terminal device in the local collaborative network can be connected to at least one other terminal device, and at least one terminal device in the local collaborative network can be connected to a central device, and the central device is also connected to an access network device in the local collaborative network.

[0041] It should be understood that different local cooperative networks may use the same access network device, or may use the same access network device.

[0042] In some implementations, each terminal device can communicate with sensors on the vehicle or drone it is in to obtain data collected by the sensors. For example, these sensors may include: laser radar, surround view camera, etc. Through these sensor devices, the terminal device can understand the position, direction, travel distance, speed and acceleration of the vehicle or drone in real time. These sensors may also include: various environmental sensors, through which the terminal device can perceive information about the external environment, including temperature, humidity, light, distance, etc.

[0043] In some possible implementations, the central device can extract various elements of traffic status, such as kinematic information of traffic participants, information for determining traffic event triggers, supporting information for calculating traffic flow related indicators, etc. At the same time, it can share data information or command information with surrounding terminal devices or user devices through wireless networks.

[0044] In some implementations, the user device may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a smart sensor, etc., but is not limited thereto.

[0045] In some possible implementations, if the local collaborative network is a local vehicle-road collaborative network or a local unmanned vehicle collaborative network, then the central device included in the local collaborative network may be a road-end device, and the road-end device may be an intelligent street lamp, traffic light, road sign, etc.

[0046] In some possible implementations, the access network device can be an access point (AP) in a wireless local area network (WLAN), a base station (Base Transceiver Station, BTS) in a global system of mobile communication (GSM) or a code division multiple access (CDMA) system, a base station (NodeB, NB) in a wideband code division multiple access (WCDMA) system, an evolved base station (Evolutional Node B, eNB or eNodeB) in a long term evolution (LTE) system, and a network device (gNB) in a new radio (NR) network, etc.

[0047] In some implementations, the access network device can access the Internet, thereby being able to access the server.

[0048] In some implementations, the server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. For example, a server in a public cloud may be a cloud server. A network device in a private cloud branch may also be a cloud server.

[0049] For example, Figure 1 A schematic diagram of a vehicle-road cooperative network architecture provided in an embodiment of the present application is shown in FIG. Figure 1 As shown in FIG. 1 , the network architecture includes: three local vehicle-road cooperative networks, among which, Figure 1 As shown, the area enclosed by each ellipse and the access network device connected to the area constitute a local vehicle-road cooperative network. In other words, each local vehicle-road cooperative network includes multiple vehicles, a road-end device and an access network device.

[0050] Among them, for any of the three local vehicle-road cooperative networks, any vehicle in the local vehicle-road cooperative network can have a vehicle-to-vehicle (V2V) connection with at least one vehicle, and these vehicles can all be connected to the road-side equipment, and the road-side equipment can also be connected to the access network equipment.

[0051] It should be noted that Figure 1 The system to which the present application is applicable is merely illustrated in the form of an example. Of course, the method shown in the embodiment of the present application can also be applied to other systems, and the embodiment of the present application does not make any specific limitation on this.

[0052] The technical solutions of the embodiments of the present application are described in detail below through some embodiments. Among them, the solutions provided by the embodiments of the present application may involve vehicle-road collaboration, drone collaboration or unmanned vehicle collaboration technology in intelligent transportation technology, etc. The following embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0053] Figure 2 A flowchart of a collaborative network networking method provided in an embodiment of the present application, the method can be executed by a first terminal, the first terminal can be a vehicle-mounted terminal or an airborne terminal, but is not limited thereto, such as Figure 2 As shown, the method may include:

[0054] S210: When the first terminal is in a first local cooperative network and the connection of the first terminal is not interrupted, obtain first characteristic data of the first terminal and second characteristic data sent by at least one network node;

[0055] In some implementations, the first terminal is any terminal in the first local cooperative network, but is not limited thereto.

[0056] In some possible implementations, the first local cooperative network is a current local cooperative network in which the first terminal is located.

[0057] It should be understood that the connection of the first terminal refers to all connections involved in the first local collaborative network. Assuming that terminal 1 is connected to terminal 2, terminal 3 and road-end equipment 4 in the current local collaborative network, then the connection of terminal 1 includes: the connection between terminal 1 and terminal 2, the connection between terminal 1 and terminal 3, and the connection between terminal 1 and road-end equipment 4.

[0058] In the embodiment of the present application, the first terminal may obtain the first characteristic data of the first terminal and the second characteristic data sent by at least one network node in any of the following implementable ways, but is not limited thereto:

[0059] In a first implementation mode, the first terminal obtains, in real time, first characteristic data of the first terminal and second characteristic data sent by at least one network node.

[0060] In a second implementation mode, the first terminal periodically obtains the first characteristic data of the first terminal and the second characteristic data sent by at least one network node. For example, the first terminal obtains the first characteristic data of the first terminal and the second characteristic data sent by at least one network node every 2 seconds (Sesond, S).

[0061] In some possible implementations, the first feature data includes at least one of the following, but is not limited to: the driving speed, driving direction, driving plan, and other connectable terminals in the surrounding area of ​​the vehicle or drone where the first terminal is located.

[0062] In some possible implementations, the first terminal may obtain the first characteristic data based on a navigation system, but is not limited thereto.

[0063] In some possible implementations, the driving speed, driving direction, and driving plan of the vehicle or drone where the first terminal is located can all come from a navigation system. The driving plan can also be called a navigation plan, for example, the navigation plan of the first terminal from starting point A to destination B.

[0064] In some possible implementations, the first terminal can obtain the positions of surrounding terminals from a navigation system, determine the distance between the first terminal and the surrounding terminals based on the first terminal's own position and the positions of the surrounding terminals, and determine the surrounding terminals whose distance is less than a preset distance as other connectable terminals around.

[0065] For example, terminal 1 can obtain the positions of its surrounding terminals from the navigation system, including the positions of terminal 2 and terminal 3. Terminal 1 calculates the distance between terminal 1 and terminal 2 based on its own position and the position of terminal 2. Assuming the distance is 10 meters, terminal 1 calculates the distance between terminal 1 and terminal 3 based on its own position and the position of terminal 3. Assuming the distance is 100 meters, if the preset distance is 30 meters, then for the terminal, terminal 2 is the other terminal that can be connected around it.

[0066] In some other possible implementations, the first terminal may obtain the distance between the surrounding terminals and the first terminal from the navigation system, and determine the surrounding terminals whose distance is less than a preset distance as other surrounding connectable terminals.

[0067] For example, terminal 1 can obtain the distances between its surrounding terminals and terminal 1 from the navigation system, including: the distances between terminal 2 and terminal 3 and terminal 1 respectively. Assume that the distance between terminal 1 and terminal 2 is 10 meters, and the distance between terminal 1 and terminal 3 is 100 meters. If the preset distance is 30 meters, then for the terminal, terminal 2 is the only other terminal that can be connected to it.

[0068] In some possible implementations, the at least one network node includes, but is not limited to: a hub device in the first local collaborative network, and an access network device to which the hub device is connected (ie, an access network device in the first local collaborative network).

[0069] In some implementations, the second characteristic data acquired by the first terminal from the hub device includes, but is not limited to: traffic flow information collected by the hub device. For example, in a vehicle-road cooperative network, the traffic flow information may be current vehicle flow.

[0070] In some implementations, the second characteristic data obtained by the first terminal from the access network device includes at least one of the following, but is not limited to: information about a cellular cell where the access network device is located, and information about neighboring cellular cells of the cellular cell.

[0071] In some implementations, the information of the cellular cell where the access network device is located includes at least one of the following, but is not limited to: the location of the cellular cell, the distance between the cellular cell and the first terminal, and the coverage of the cellular cell. The distance between the cellular cell and the first terminal may be the distance between the center of the cellular cell and the first terminal, but is not limited to this.

[0072] In some implementations, the information of the neighbor cell includes at least one of the following, but is not limited thereto: the location of the neighbor cell, the distance between the neighbor cell and the first terminal, and the coverage of the neighbor cell. The distance between the neighbor cell and the first terminal may be the distance between the center of the neighbor cell and the first terminal, but is not limited thereto.

[0073] S220: Obtaining a second local collaborative network based on the first feature data and the second feature data;

[0074] In a first implementation mode, S220 may include:

[0075] S220-1A: Input the first feature data and the second feature data into the first neural network model to obtain a second local collaborative network.

[0076] In other words, the input of the first neural network model is the first feature data and the second feature data, and the output is the second local collaborative network.

[0077] For example, the vehicle terminal 1 obtains its first characteristic data, including: the current speed of the vehicle it is in is 50 kilometers per hour (km / h), the driving direction is from east to west, the driving plan is from the current position to destination A, and the surrounding connectable terminals are vehicle terminal 2, vehicle terminal 3 and vehicle terminal 4. And the second characteristic data obtained by the vehicle terminal 1 from the road-side device it is connected to includes: the current traffic volume is 20 vehicles passing through per minute, and the second characteristic data obtained by the vehicle terminal 1 from the access network device includes: the position of the cellular cell 01 where the vehicle terminal 1 is located is (x0, y0), the distance between the cellular cell 01 and the vehicle terminal 1 is 1km, and the coverage range of the cellular cell 01 is centered on the access network device with a radius of 10km. The position of the neighboring cellular cell 02 is (x1, y1), the distance between the vehicle terminal 1 and the neighboring cellular cell 02 is 5km, and the neighboring cellular cell is centered on the access network device corresponding to the neighboring cellular cell with a radius of 10km. The vehicle terminal 1 can input the first feature data and the second feature data into the first neural network model, and the obtained second local collaborative network is as follows: the vehicle terminal 1, the vehicle terminal 2, the vehicle terminal 3 and the vehicle terminal 4 are interconnected with each other, and they are connected to the access network device corresponding to the neighboring cellular cell 01 through a road end device, and the road end device is connected to the access network device.

[0078] In some possible implementations, the training method for the first neural network model may be a supervised training method or an unsupervised training method.

[0079] In the training phase of the first neural network model, the training device can calculate the training loss. Further, the training device can minimize the loss to adjust the parameters of the first neural network model until the loss converges or the number of training times reaches a preset number.

[0080] In some implementations, the loss may be a cross entropy loss, a one-norm (L1) loss, a mean square error loss (MSE), etc., but is not limited thereto.

[0081] In a second implementation mode, S220 may include:

[0082] S220-1B: Obtaining a second local collaborative network based on the nodes within the preset range and the connection relationship between the nodes, the first feature data, and the second feature data;

[0083] The preset range covers the first local collaborative network and the second local collaborative network. In other words, when predicting the second local collaborative network, the first terminal may first determine an approximate range where the second local collaborative network may be located, i.e., the preset range. For example, the preset range may be a circular range with the first terminal as the center and 100 meters as the preset radius.

[0084] In some implementations, each node within the preset range may include: the first terminal, at least one other terminal, at least one road end device, and at least one access network device.

[0085] In some implementations, S220-1B may include:

[0086] S220-1B-1a: Generate a target attribute graph based on the nodes within a preset range, the connection relationship between the nodes, the first feature data, and the second feature data;

[0087] In some possible implementations, the first terminal may use each node within a preset range as a node of a target attribute graph, use the connection relationship between each node as an edge of the target attribute graph, use the first feature data as the attribute corresponding to the node corresponding to the first terminal, and use the second feature data as the attribute corresponding to the corresponding node. For example, assuming that a certain second feature data comes from the road-end device 001, the first terminal may use the second feature data as the attribute of the node corresponding to the road-end device 001 in the target attribute graph. For another example, assuming that a certain second feature data comes from the access network device 0001, the first terminal may use the second feature data as the attribute of the node corresponding to the access network device 0001 in the target attribute graph.

[0088] S220-1B-2a: Based on the target attribute graph, obtain a second local collaborative network.

[0089] In some implementations, the first terminal may input the target attribute graph into the graph neural network to obtain a second local collaborative network.

[0090] As mentioned above, a graph neural network is an optimizable transformation of all attributes on a graph, which can maintain the symmetric information of the graph, namely, permutation invariances. A graph neural network is a "graph-in, graph-out" structure, which transforms the nodes and edges of the input graph, but does not change its connectivity. It can be seen that the nodes involved in the second local collaborative network are exactly the same as those in the target attribute graph, that is, the connectivity is the same. The difference is that the second local collaborative network may have changed attributes and / or edges relative to the target attribute graph.

[0091] For example, Figure 3 A schematic diagram of a target attribute graph and a second local collaborative network provided in an embodiment of the present application, such as Figure 3 As shown, the target attribute graph and the nodes included in the second local collaborative network are exactly the same. The difference is that in the target attribute graph, terminal 2 is connected to terminal 1, terminal 3 and road end device 01, and in the second local collaborative network, terminal 2 is connected to terminal 5, terminal 6 and road end device 03.

[0092] In some other possible implementations, the first terminal may construct a convex optimization problem based on the target attribute graph, wherein the objective function of the convex optimization problem may be to construct a second local collaborative network with the least number of edges based on the target attribute graph, and the second local collaborative network needs to satisfy that the sum of the signal transmission delays in the entire network is less than a preset threshold. The first terminal may use a heuristic algorithm to calculate the convex optimization problem, and finally obtain the second local collaborative network.

[0093] In some other possible implementations, S220-1B may include:

[0094] S220-1B-1b: Input each node within a preset range, the connection relationship between each node, the first feature data and the second feature data into a second neural network model to obtain a second local collaborative network.

[0095] In other words, the input of the second neural network model is the nodes within a preset range and the connection relationship between the nodes, the first feature data and the second feature data, and the output is the second local collaborative network.

[0096] In some implementations, the training method for the second neural network model may be a supervised training method or an unsupervised training method.

[0097] In the training phase of the second neural network model, the training device can calculate the training loss. Further, the training device can minimize the loss to adjust the parameters of the second neural network model until the loss converges or the number of training times reaches a preset number.

[0098] In some implementations, the loss may be a cross entropy loss, a one-norm (L1) loss, a MSE, etc., but is not limited thereto.

[0099] In a third implementation mode, S220 may include:

[0100] S220-1C: Process the second characteristic data from the central device to obtain third characteristic data;

[0101] In some implementations, the second characteristic data from the hub device may include: traffic flow information collected by the hub device. Based on this, the first terminal may determine the current road condition information based on the traffic flow information.

[0102] In some possible implementations, when the traffic flow information is greater than a preset flow, the first terminal determines that the current road condition information is a congested state; when the traffic flow information is less than or equal to the preset flow, the first terminal determines that the current road condition information is a smooth state.

[0103] For example, assuming that the traffic flow information obtained by the vehicle terminal 1 from the road-side device 01 is 20 vehicles passing through per minute, and the above-mentioned preset flow rate is 15, based on this, it can be determined that the current road condition information is that the current road condition is in a congested state.

[0104] S220-2C: Input the first feature data, the third feature data and the second feature data from the access network device into a third neural network model to obtain a second local collaborative network.

[0105] In other words, the input of the first neural network model is the first feature data, the third feature data and the second feature data from the access network device, and the output is the second local collaborative network.

[0106] For example, the vehicle terminal 1 obtains its first characteristic data, including: the current speed of the vehicle it is in is 50 kilometers per hour (km / h), the driving direction is from east to west, the driving plan is from the current position to destination A, and the surrounding connectable terminals are vehicle terminal 2, vehicle terminal 3 and vehicle terminal 4. And the second characteristic data obtained by the vehicle terminal 1 from the road-side equipment it is connected to includes: the current traffic volume is 20 vehicles passing through per minute, and the vehicle terminal obtains the third characteristic data based on the second characteristic data, that is, the current road condition information is that the current road condition is congested. The vehicle terminal 1 also obtains the second characteristic data from the access network equipment, including: the position of the cellular cell 01 where the vehicle terminal 1 is located is (x0, y0), the distance between the cellular cell 01 and the vehicle terminal 1 is 1km, and the coverage range of the cellular cell 01 is centered on the access network equipment and has a radius of 10km. The location of the neighboring cell 02 is (x1, y1), the distance between the neighboring cell 02 and the vehicle terminal 1 is 5km, and the neighboring cell is centered on the access network device corresponding to the neighboring cell with a radius of 10km. The vehicle terminal 1 can input the first feature data, the third feature data and the second feature data from the access network device into the third neural network model, and the obtained second local collaborative network is as follows: the vehicle terminal 1, the vehicle terminal 2, the vehicle terminal 3 and the vehicle terminal 4 are interconnected with each other, and they are connected to the access network device corresponding to the neighboring cell 01 through a road end device, and the road end device is connected to the access network device.

[0107] In some possible implementations, the training method for the third neural network model may be a supervised training method or an unsupervised training method.

[0108] In the training phase of the third neural network model, the training device can calculate the training loss. Further, the training device can minimize the loss to adjust the parameters of the third neural network model until the loss converges or the number of training times reaches a preset number.

[0109] In some implementations, the loss may be a cross entropy loss, a one-norm (L1) loss, a MSE, etc., but is not limited thereto.

[0110] In a fourth implementation mode, S220 may include:

[0111] S220-1D: Process the second characteristic data from the central device to obtain third characteristic data;

[0112] S220-2D: Obtain a second local collaborative network based on the nodes within the preset range and the connection relationship between the nodes, the first feature data, the third feature data, and the second feature data from the access network device;

[0113] It should be understood that the explanation of the method for obtaining the third characteristic data and the preset range can be referred to above, and the embodiments of the present application will not be repeated here.

[0114] In some implementations, S220-2D may include:

[0115] S220-2D-1a: Generate a target attribute graph based on the nodes within a preset range and the connection relationship between the nodes, the first feature data, the third feature data, and the second feature data from the access network device;

[0116] In some possible implementations, the first terminal can use each node within a preset range as a node of a target attribute graph, and use the connection relationship between each node as an edge of the target attribute graph, use the first characteristic data as the attribute corresponding to the node corresponding to the first terminal, use the second characteristic data from the access network device as the attribute corresponding to the access network device, and use the third characteristic data as the attribute corresponding to the node corresponding to the central device.

[0117] S220-2D-2a: Based on the target attribute graph, obtain a second local collaborative network.

[0118] In some implementations, the first terminal may input the target attribute graph into the graph neural network to obtain a second local collaborative network.

[0119] In some other possible implementations, the first terminal may construct a convex optimization problem based on the target attribute graph, wherein the objective function of the convex optimization problem may be to construct a second local collaborative network with the least number of edges based on the target attribute graph, and the second local collaborative network needs to satisfy that the sum of the signal transmission delays in the entire network is less than a preset threshold. The first terminal may use a heuristic algorithm to calculate the convex optimization problem, and finally obtain the second local collaborative network.

[0120] It should be understood that the explanation of S220-2D-2a can refer to the explanation of S220-1B-2a, and the embodiments of the present application will not go into details.

[0121] In some other possible implementations, S220-2D may include:

[0122] S220-2D-1b: Input each node within a preset range, the connection relationship between each node, the first feature data, the third feature data and the second feature data from the access network device into the fourth neural network model to obtain a second local collaborative network.

[0123] In other words, the input of the fourth neural network model is the nodes within a preset range and the connection relationship between the nodes, the first feature data, the third feature data and the second feature data from the access network device, and the output is a second local collaborative network.

[0124] In some implementations, the training method for the fourth neural network model may be a supervised training method or an unsupervised training method.

[0125] In the training phase of the fourth neural network model, the training device can calculate the training loss. Further, the training device can minimize the loss to adjust the parameters of the fourth neural network model until the loss converges or the number of training times reaches a preset number.

[0126] In some implementations, the loss may be a cross entropy loss, a one-norm (L1) loss, a MSE, etc., but is not limited thereto.

[0127] S230: When the first local collaborative network is different from the second local collaborative network, the first local collaborative network is deleted, and the second local collaborative network is networked.

[0128] In some implementations, if the first local collaborative network and the second local collaborative network do not have the same connection, the first terminal can completely delete the first local collaborative network and form a new second local collaborative network. If the first local collaborative network and the second local collaborative network have some of the same connections, the first terminal can delete the different parts of the first local collaborative network from the second local collaborative network, and form the second local collaborative network based on the undeleted parts, which can improve the networking efficiency.

[0129] The first terminal may delete all connections of the first local cooperative network in the following manner, but is not limited thereto:

[0130] In some possible implementations, the first terminal deletes the connection between the first terminal and other terminals in the first local collaborative network, and deletes the connection between the first terminal and the hub device in the first local collaborative network. When the connection between the hub device and the first terminal is deleted, the hub device can delete the connection between it and the access network device. At the same time, the hub device can delete the connection between it and other terminals in the first local collaborative network. When the connection between these other terminals and the road-end devices is deleted, these other terminals will continue to delete the connection between themselves and other terminals until all connections in the first local collaborative network are deleted.

[0131] In other possible implementations, the first terminal deletes the connection between the first terminal and a hub device in the first local collaborative network. When the connection between the hub device and the first terminal is deleted, the hub device can delete the connection between it and the access network device. At the same time, the hub device can delete the connection between it and other terminals in the first local collaborative network. When the connection between these other terminals and the road-end devices is deleted, these other terminals will continue to delete the connection between themselves and other terminals until all connections in the first local collaborative network are deleted.

[0132] The first terminal may delete some connections of the first local cooperative network in the following manner, but is not limited thereto:

[0133] In some possible implementations, the first terminal can determine the connection to be deleted and send an indication message to a central device in the first local collaborative network. The central device can send the indication message to all terminals associated with the connection to be deleted. The central device can delete the connection to be deleted associated with the central device based on the indication message, and all terminals that have received the indication message can also delete their corresponding connections to be deleted based on the indication message.

[0134] In some implementations, the indication message is used to indicate the connection to be deleted determined by the first terminal.

[0135] In some possible implementations, the first terminal may use the following method to completely network the second local cooperative network, but is not limited thereto:

[0136] In some possible implementations, the first terminal sends network information of the second local collaborative network to a central device in the second local collaborative network. The central device establishes a connection between the central device and the access network device based on the network information, and establishes a connection with each terminal. At the same time, the central device sends the network information to the associated terminals so that these terminals can establish connections with each other.

[0137] In some possible implementations, the first terminal may partially network the second local cooperative network in the following manner, but is not limited thereto:

[0138] In some possible implementations, the first terminal sends a connection establishment request to other terminals in the second local collaborative network that have not yet been connected, so as to establish a connection between the first terminal and other terminals in the second local collaborative network that have not yet been connected. At the same time, the first terminal can also send a connection establishment request to a road-end device in the second local collaborative network, so as to establish a connection between the first terminal and the road-end device. The first terminal can also send network information of the second local system network to the road-end device, so that the road-end device sends a connection request to other corresponding terminals based on the network information, so as to establish a connection between other terminals and the road-end device. At the same time, the road-end device can send the network information to other terminals, so as to establish a connection between other terminals. And the road-end device can also establish a connection with an access network device in the second local collaborative network based on the network information.

[0139] In other possible implementations, the first terminal can determine the connection to be added and send an indication message to a central device in the second local collaborative network. The central device can send the indication message to all terminals associated with the connection to be added. The central device can add the connection to be added associated with the central device based on the indication message, and all terminals that have received the indication message can also add their corresponding connections to be added based on the indication message.

[0140] In some implementations, the indication message is used to indicate the connection to be added determined by the first terminal.

[0141] In some implementations, the network information of the second local system network includes: nodes included in the second local system network and connection relationships between the nodes, but is not limited thereto.

[0142] In some implementations, when the first local cooperative network is the same as the second local cooperative network, the first terminal may not perform any operation.

[0143] The embodiment of the present application provides a method for networking a collaborative network, including: when a first terminal is in a first local collaborative network and the connection of the first terminal is not interrupted, obtaining the first characteristic data of the first terminal and the second characteristic data sent by at least one network node; based on the first characteristic data and the second characteristic data, obtaining the second local collaborative network; when the first local collaborative network is different from the second local collaborative network, deleting the first local collaborative network and networking the second local collaborative network. Since the first terminal can pre-network the second local collaborative network before its connection is interrupted, in other words, this method can avoid the situation of connection interruption. On the one hand, the reliability of data transmission can be guaranteed, and on the other hand, this smooth networking method can reduce data transmission delay.

[0144] In some implementations, after the above S230, the networking method may further include:

[0145] S410: compressing original point cloud data collected by the first terminal to obtain first compressed point cloud data;

[0146] It should be understood that each point of the original point cloud data has at least three-dimensional position information, and may also have additional attributes such as color, material or other information such as reflectivity, depending on the application scenario. Usually, each point in the point cloud has the same number of additional attributes. For example, the point cloud obtained according to the principle of laser measurement includes three-dimensional coordinates (XYZ) and laser reflection intensity (reflectance); the point cloud obtained according to the principle of photogrammetry includes three-dimensional coordinates (XYZ) and color information, such as red, green and blue (Red Green Blue, RGB); the point cloud obtained by combining laser measurement and photogrammetry principles includes three-dimensional coordinates (XYZ), laser reflection intensity (reflectance) and color information (RGB).

[0147] It should be understood that in the embodiments of the present application, the point cloud data may be referred to as a machine-perceived point cloud, which may be used in an autonomous navigation system, but is not limited thereto.

[0148] In some possible implementations, the original point cloud data may be a second type of dynamic point cloud, or a third type of dynamic point cloud. Based on this, when the first terminal compresses the original point cloud data, it may use the G-PCC compression technology or the V-PCC compression technology, which is not limited in the embodiments of the present application.

[0149] It should be understood that the compression process of the original point cloud data by the first terminal is also called the feature extraction process for the original point cloud data, which can be understood as feature extraction of the original point cloud data through a feature extraction model, and the embodiment of the present application does not limit the feature extraction model.

[0150] S420: Acquire a correspondence relationship between at least one sub-area in the blind area of ​​the first terminal and other terminals;

[0151] In some implementations, the blind area of ​​the first terminal may be an area within a preset area, excluding the visible area of ​​the first terminal. Alternatively, the blind area of ​​the first terminal may be an area within a preset area, excluding the best visible area of ​​the first terminal.

[0152] Based on this, in the embodiment of the present application, the first terminal may determine the blind area of ​​the first terminal in any of the following achievable ways, but is not limited thereto:

[0153] In a first implementation mode, the first terminal may determine an area other than the visible area in the preset area as a blind area of ​​the first terminal.

[0154] In some implementations, the preset area may be a circular area with the first terminal as the center and a preset radius as the radius. For example, the value of the preset radius may be 3 meters, etc., but is not limited thereto.

[0155] For example, assume that the preset area is a circular area with a preset radius of 3 meters centered on the first terminal, and the visible area of ​​the first terminal includes: the area observable in front of the driving direction and the area observable through the rearview mirror. In addition, the remaining area is the blind spot of the first terminal.

[0156] A second implementation method is that the first terminal can determine the blind area of ​​the first terminal based on the first compressed point cloud data.

[0157] The first terminal may generate a data distribution diagram of the first compressed point cloud data; and delete an area corresponding to a peak of the data distribution diagram in a preset area to obtain a blind area of ​​the first terminal.

[0158] It should be understood that each point in the first compressed point cloud data has a position coordinate. Based on this, the data distribution diagram of the first compressed point cloud data refers to the position distribution of the first compressed point cloud data. Its horizontal coordinate can represent different areas, and the vertical coordinate can represent the number of points in the area. For example, in area 1, the number of points it corresponds to is 100, and in area 2, the number of points it corresponds to is 1000. Based on this, the first terminal can find the area corresponding to the peak of the data distribution diagram, and this area is the best visible area of ​​the first terminal. Furthermore, the first terminal can delete the best visible area from the preset area to obtain the blind area of ​​the first terminal.

[0159] For example, assuming that the preset area is a circular area with a preset radius of 3 meters centered on the first terminal, and the best viewing area of ​​the first terminal is based on the vehicle's driving direction, from 60 degrees to the left to 60 degrees to the right, the remaining area in the preset area except the best viewing area is the blind spot of the first terminal.

[0160] In the embodiment of the present application, the first terminal may obtain the correspondence between at least one sub-area in the blind area of ​​the first terminal and other terminals in any of the following achievable ways, but is not limited thereto:

[0161] In a first implementation mode, the first terminal locally obtains a correspondence between at least one sub-area in the blind area of ​​the first terminal and other terminals.

[0162] In some implementations, the first terminal may establish a correspondence between at least one sub-area within the blind spot of the first terminal and other terminals, and obtain the correspondence locally.

[0163] A second implementation method is that the first terminal obtains the correspondence between at least one sub-area in the blind area of ​​the first terminal and other terminals from a central device in the second local cooperative network.

[0164] In some implementations, a hub device in the second local cooperative network may establish a correspondence between at least one sub-area within the blind spot of the first terminal and other terminals, and the first terminal may obtain the correspondence from the hub device.

[0165] S430: Sending a first request message to other terminals corresponding to at least one sub-area based on the corresponding relationship;

[0166] The first request message is used to request to obtain the second compressed point cloud data of other terminals.

[0167] It should be understood that, for other terminals, they also compress the original point cloud data collected by themselves in a similar manner as the first terminal to form their own second compressed point cloud data.

[0168] For example, assuming that the blind area of ​​terminal 1 includes: sub-area A and sub-area B, where sub-area A corresponds to terminal 2 and sub-area B corresponds to terminal 3, based on this, terminal 1 can send a first request message to terminal 2 corresponding to sub-area A, and the first request message is used to request to obtain the second compressed point cloud data of terminal 2. Terminal 1 can also send a first request message to terminal 3 corresponding to sub-area B, and the first request message is used to request to obtain the second compressed point cloud data of terminal 3.

[0169] S440: Receive second compressed point cloud data respectively sent by other terminals corresponding to at least one sub-area;

[0170] S450: Fusing the first compressed point cloud data and the second compressed point cloud data.

[0171] In some possible implementations, the first terminal may combine the first compressed point cloud data and the second compressed point cloud data to achieve fusion of the first compressed point cloud data and the second compressed point cloud data.

[0172] For example, the first compressed point cloud data is the point cloud data corresponding to the area formed from 60 degrees on the left to 60 degrees on the right based on the driving direction of the vehicle where terminal 1 is located, while the point cloud data from terminal 2 is the point cloud data corresponding to the area formed from 60 degrees on the right to 120 degrees based on the driving direction of the vehicle where terminal 1 is located. Based on this, the fused point cloud data is the point cloud data corresponding to the area formed from 60 degrees on the left to 120 degrees on the right.

[0173] In some possible implementations, the first terminal may also receive a second request message sent by the second terminal; wherein the second request message is used to request acquisition of the first compressed point cloud data; and in response to the second request message, the first compressed point cloud data is sent to the second terminal.

[0174] In an embodiment of the present application, the first terminal can compress its original point cloud data to obtain first compressed point cloud data, and can specifically request second compressed point cloud data from other terminals corresponding to each sub-area in the blind spot of the first terminal, and fuse the first compressed point cloud data and the second compressed point cloud data. Since the fused data has a larger corresponding range and higher precision than the first compressed point cloud data, and the fused data is used for the perception of the first terminal, the first terminal can obtain a perception result with a larger range and higher precision to achieve efficient collaborative perception. In addition, compared with the method of directly transmitting original point cloud data between terminal devices, in an embodiment of the present application, what is transmitted between terminals is compressed point cloud data, and the compressed point cloud data is transmitted in a targeted manner, thereby reducing the load of the local collaborative network.

[0175] In some implementations, after the above S230, the networking method may further include:

[0176] S510: Acquire network characteristics between the first terminal and the third terminal;

[0177] The third terminal is any terminal connected to the first terminal in the second local cooperative network.

[0178] In some implementations, the network characteristic includes at least one of the following, but is not limited thereto: throughput of the first terminal, throughput of the third terminal, available bandwidth between the first terminal and the third terminal, and signal strength.

[0179] It should be understood that throughput refers to the maximum rate that a device can accept without frame loss.

[0180] In some possible implementations, null packets may be transmitted between the first terminal and the third terminal; and network characteristics may be obtained based on the transmission of the null packets.

[0181] S520: predicting a transmission delay between the first terminal and the third terminal based on network characteristics to obtain a predicted transmission delay;

[0182] In some implementations, the first terminal may input network characteristics into a machine learning model to obtain a predicted transmission delay.

[0183] In some implementations, the training method for the machine learning model can be a supervised training method or an unsupervised training method.

[0184] In the training phase of the machine learning model, the training device can calculate the training loss. Further, the training device can minimize the loss to adjust the parameters of the machine learning model until the loss converges or the number of training times reaches a preset number.

[0185] In some implementations, the loss may be a cross entropy loss, a one-norm (L1) loss, a MSE, etc., but is not limited thereto.

[0186] In other possible implementations, the first terminal may establish a correspondence between network characteristics and transmission delays based on acquiring historical network characteristics and corresponding transmission delays, and further, may predict transmission delays based on the acquired network characteristics and the correspondence.

[0187] In some possible implementations, the first terminal may take an average value of transmission delays corresponding to the same historical network characteristics to establish a corresponding relationship between the historical network characteristics and the average value.

[0188] For example, assuming that the historical available bandwidth is 20M, the corresponding transmission delays include: 2ms, 4ms, then the first terminal can calculate the average value of 2ms and 4ms, which is 3ms. Based on this, the first terminal can establish a corresponding relationship between the available bandwidth of 20M and the transmission delay of 3ms.

[0189] S530: predicting at least one of a data transmission granularity and a data transmission rate of the first terminal based on the predicted transmission delay, and obtaining at least one of a predicted data transmission granularity and a predicted data transmission rate;

[0190] In some possible implementations, the first terminal may establish a correspondence between the transmission delay and at least one of the data transmission granularity and the data transmission rate, and obtain at least one of the predicted data transmission granularity and the predicted data transmission rate based on the predicted transmission delay and the correspondence.

[0191] The first terminal may establish a corresponding relationship between the transmission delay and the data transmission rate. For example, the corresponding relationship is shown in Table 1:

[0192] Table 1

[0193] Transmission delay Data transfer rate 2ms 10 packets / second 3ms 8 packets / second 4ms 4 packets / second …… ……

[0194] Assuming that the above predicted transmission delay is 3ms, the corresponding data transmission rate can be 8 data packets / second.

[0195] The first terminal may establish a corresponding relationship between the transmission delay and the data transmission granularity. For example, the corresponding relationship is shown in Table 2:

[0196] Table 2

[0197]

[0198]

[0199] Assuming that the above predicted transmission delay is 3ms, the corresponding data transmission rate can be 8 data packets / time.

[0200] The first terminal may establish a corresponding relationship between the transmission delay and the data transmission rate. For example, the corresponding relationship is shown in Table 3:

[0201] Table 3

[0202] Transmission delay Data transfer rate 2ms 10 packets / second 3ms 8 packets / second 4ms 4 packets / second …… ……

[0203] Assuming that the above predicted transmission delay is 3ms, the corresponding data transmission rate can be 8 data packets / second.

[0204] The first terminal may establish a corresponding relationship between the transmission delay, the data transmission granularity, and the data transmission rate. For example, the corresponding relationship is shown in Table 4:

[0205] Table 4

[0206] Transmission delay Data transmission granularity and data transmission speed 2ms 10 packets / each time, 10 packets / second 3ms 8 packets / each, 8 packets / second 4ms 4 packets / each, 4 packets / second …… ……

[0207] Assuming that the above predicted transmission delay is 3ms, the corresponding data transmission rate can be 8 data packets / time and the data transmission granularity is 8 data packets / second.

[0208] S540: Perform data transmission to the third terminal based on at least one of the predicted data transmission granularity and the predicted data transmission rate.

[0209] In some possible implementations, if the above-mentioned predicted transmission delay is obtained based on a machine learning model, then after S540, the networking method also includes: obtaining the actual transmission delay between the first terminal and the third terminal; generating training samples based on the actual transmission delay and network characteristics; and training the machine learning model based on the training samples.

[0210] In the embodiment of the present application, the first terminal can obtain the network characteristics between the first terminal and the third terminal; predict the transmission delay between the first terminal and the third terminal based on the network characteristics to obtain the predicted transmission delay; predict at least one of the data transmission granularity and the data transmission rate of the first terminal based on the predicted transmission delay to obtain at least one of the predicted data transmission granularity and the predicted data transmission rate; and perform data transmission to the third terminal based on at least one of the predicted data transmission granularity and the predicted data transmission rate. Thus, dynamic adaptation to the network environment can be achieved.

[0211] The preferred embodiments of the present application are described in detail above in conjunction with the accompanying drawings. However, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, the technical solution of the present application can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the present application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present application will not further explain various possible combinations. For another example, the various different embodiments of the present application can also be arbitrarily combined, as long as they do not violate the ideas of the present application, they should also be regarded as the contents disclosed in the present application.

[0212] It should also be understood that in the various method embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean 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 the embodiments of the present application.

[0213] The method provided in the embodiment of the present application is described above, and the device provided in the embodiment of the present application is described below.

[0214] Figure 4A schematic diagram of a terminal device 600 provided for an embodiment of the present application, wherein the terminal device 600 is a first terminal, and the first terminal includes: a transceiver module 610 and a processing module 620, wherein the transceiver module 610 is used to obtain first characteristic data of the first terminal and second characteristic data sent by at least one network node when the first terminal is in a first local collaborative network and the connection of the first terminal is not interrupted; the processing module 620 is used to: obtain a second local collaborative network based on the first characteristic data and the second characteristic data; when the first local collaborative network is different from the second local collaborative network, delete the first local collaborative network and network the second local collaborative network.

[0215] In some possible implementations, the processing module 620 is specifically used to obtain a second local collaborative network based on each node within a preset range and the connection relationship between each node, the first feature data and the second feature data; wherein the preset range covers the first local collaborative network and the second local collaborative network.

[0216] In some possible implementations, the processing module 620 is specifically used to: generate a target attribute graph based on each node within a preset range and the connection relationship between each node, the first feature data and the second feature data; and obtain a second local collaborative network based on the target attribute graph.

[0217] In some implementations, the processing module 620 is specifically used to: input the target attribute graph into the graph neural network to obtain a second local collaborative network.

[0218] In some possible implementations, after the processing module 620 networks the second local collaborative network, the processing module 620 is also used to compress the original point cloud data collected by the first terminal to obtain first compressed point cloud data; the transceiver module 610 is also used to obtain the correspondence between at least one sub-area within the blind spot of the first terminal and other terminals; based on the correspondence, a first request message is sent to other terminals corresponding to at least one sub-area; wherein the first request message is used to request to obtain second compressed point cloud data from other terminals; and receive second compressed point cloud data sent by each of the other terminals corresponding to at least one sub-area; the processing module 620 is also used to fuse the first compressed point cloud data and the second compressed point cloud data.

[0219] In some implementations, the processing module 620 is further used to determine the blind spot of the first terminal based on the first compressed point cloud data before the transceiver module 610 obtains the correspondence between at least one sub-area within the blind spot of the first terminal and other terminals.

[0220] In some possible implementations, the processing module 620 is specifically used to: generate a data distribution graph of the first compressed point cloud data; and delete an area corresponding to a peak of the data distribution graph in a preset area to obtain a blind area of ​​the first terminal.

[0221] In some possible implementations, the transceiver module 610 is also used to: receive a second request message sent by a second terminal; wherein the second request message is used to request to obtain the first compressed point cloud data; and in response to the second request message, send the first compressed point cloud data to the second terminal.

[0222] In some possible implementations, the transceiver module 610 is also used to: obtain network characteristics between the first terminal and the third terminal; wherein the third terminal is any terminal connected to the first terminal in the second local collaborative network; the processing module 620 is also used to: predict the transmission delay between the first terminal and the third terminal based on the network characteristics to obtain a predicted transmission delay; predict at least one of the data transmission granularity and the data transmission rate of the first terminal based on the predicted transmission delay to obtain at least one of the predicted data transmission granularity and the predicted data transmission rate; the transceiver module 610 is also used to: perform data transmission to the third terminal based on at least one of the predicted data transmission granularity and the predicted data transmission rate.

[0223] In some implementations, the processing module 620 is specifically used to: input network characteristics into a machine learning model to obtain a predicted transmission delay.

[0224] In some possible implementations, the processing module 620 is also used to: obtain the actual transmission delay between the first terminal and the third terminal after the transceiver module 610 performs data transmission to the third terminal based on at least one of the predicted data transmission granularity and the predicted data transmission rate; generate training samples based on the actual transmission delay and network characteristics; and train the machine learning model based on the training samples.

[0225] In some possible implementations, the processing module 620 is specifically used to: transmit empty packets to the third terminal; and obtain network characteristics based on the transmission of the empty packets.

[0226] It should be understood that the device embodiment and the method embodiment may correspond to each other, and similar descriptions may refer to the method embodiment. To avoid repetition, they will not be described here. Specifically, Figure 4 The device terminal device 600 shown can perform Figure 2 The corresponding method embodiments, and the aforementioned and other operations and / or functions of each module in the device terminal device 600 are respectively to achieve Figure 2 For the sake of brevity, the corresponding processes in each method are not repeated here.

[0227] The above describes the device terminal device 600 of the embodiment of the present application from the perspective of the functional module in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, and can also be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or a combination of hardware and software modules in the decoding processor to perform. Optionally, the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory, and completes the steps in the above method embodiment in conjunction with its hardware.

[0228] Figure 5 It is a schematic block diagram of an electronic device provided in an embodiment of the present application.

[0229] like Figure 5 As shown, the electronic device may include:

[0230] The memory 710 and the processor 720, the memory 710 is used to store the computer program and transmit the program code to the processor 720. In other words, the processor 720 can call and run the computer program from the memory 710 to implement the method in the embodiment of the present application.

[0231] For example, the processor 720 may be configured to execute the above method embodiments according to instructions in the computer program.

[0232] In some embodiments of the present application, the processor 720 may include but is not limited to:

[0233] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0234] In some embodiments of the present application, the memory 710 includes but is not limited to:

[0235] Volatile memory and / or non-volatile memory. Among them, the non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0236] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory 710 and executed by the processor 720 to complete the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0237] like Figure 5 As shown, the electronic device may also include:

[0238] The transceiver 730 may be connected to the processor 720 or the memory 710 .

[0239] The processor 720 may control the transceiver 730 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices. The transceiver 730 may include a transmitter and a receiver. The transceiver 730 may further include an antenna, and the number of antennas may be one or more.

[0240] It should be understood that the various components in the electronic device are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0241] The present application also provides a computer storage medium on which a computer program is stored, and when the computer program is executed by a computer, the computer can perform the method of the above method embodiment. In other words, the present application embodiment also provides a computer program product containing instructions, and when the instructions are executed by a computer, the computer can perform the method of the above method embodiment.

[0242] When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integration. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a solid state drive (solid state disk, SSD)), etc.

[0243] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0244] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the module is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0245] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. For example, each functional module in each embodiment of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0246] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A collaborative network construction method, characterized in that: The method is applied to a first terminal, and the method includes: When the first terminal is in a first local cooperative network and the connection of the first terminal is not interrupted, acquiring first characteristic data of the first terminal and second characteristic data sent by at least one network node; Based on the first feature data and the second feature data, obtaining a second local collaborative network; When the first local collaborative network is different from the second local collaborative network, the first local collaborative network is deleted, and the second local collaborative network is networked.

2. The method according to claim 1, characterized in that The obtaining a second local collaborative network based on the first feature data and the second feature data includes: Based on the nodes within the preset range and the connection relationship between the nodes, the first feature data and the second feature data, a second local collaborative network is obtained; The preset range covers the first local collaborative network and the second local collaborative network.

3. The method according to claim 2, characterized in that The obtaining of the second local collaborative network based on the nodes within the preset range and the connection relationship between the nodes, the first feature data and the second feature data includes: Generate a target attribute graph based on the nodes within the preset range and the connection relationship between the nodes, the first feature data and the second feature data; Based on the target attribute graph, the second local collaborative network is obtained.

4. The method according to claim 3, characterized in that The obtaining the second local collaborative network based on the target attribute graph includes: The target attribute graph is input into the graph neural network to obtain the second local collaborative network.

5. The method according to any one of claims 1 to 4, characterized in that: After the second local cooperative network is formed, the method further includes: Compressing the original point cloud data collected by the first terminal to obtain first compressed point cloud data; Acquire a correspondence relationship between at least one sub-area in the blind area of ​​the first terminal and other terminals; Sending a first request message to other terminals corresponding to the at least one sub-area based on the corresponding relationship; wherein the first request message is used to request to obtain second compressed point cloud data of other terminals; Receiving second compressed point cloud data respectively sent by other terminals corresponding to the at least one sub-area; The first compressed point cloud data and the second compressed point cloud data are fused.

6. The method according to claim 5, characterized in that Before acquiring the correspondence relationship between at least one sub-area in the blind area of ​​the first terminal and other terminals, the method further includes: Based on the first compressed point cloud data, a blind spot of the first terminal is determined.

7. The method according to claim 6, characterized in that The determining the blind area of ​​the first terminal based on the first compressed point cloud data includes: generating a data distribution map of the first compressed point cloud data; The area corresponding to the peak of the data distribution diagram in the preset area is deleted to obtain the blind area of ​​the first terminal.

8. The method according to claim 5, characterized in that Also includes: Receiving a second request message sent by a second terminal; wherein the second request message is used to request to obtain the first compressed point cloud data; In response to the second request message, the first compressed point cloud data is sent to the second terminal.

9. The method according to any one of claims 1 to 4, characterized in that: Also includes: Acquire a network characteristic between the first terminal and a third terminal; wherein the third terminal is any terminal in the second local cooperative network connected to the first terminal; Predicting a transmission delay between the first terminal and the third terminal based on the network characteristics to obtain a predicted transmission delay; Predicting at least one of a data transmission granularity and a data transmission rate of the first terminal based on the predicted transmission delay to obtain at least one of a predicted data transmission granularity and a predicted data transmission rate; Based on at least one of the predicted data transmission granularity and the predicted data transmission rate, data transmission is performed to the third terminal.

10. The method according to claim 9, characterized in that The predicting the transmission delay between the first terminal and the third terminal based on the network characteristics to obtain the predicted transmission delay includes: The network characteristics are input into a machine learning model to obtain the predicted transmission delay.

11. The method according to claim 10, characterized in that After performing data transmission to the third terminal based on at least one of the predicted data transmission granularity and the predicted data transmission rate, the method further includes: Acquire an actual transmission delay between the first terminal and the third terminal; Based on the actual transmission delay and the network characteristics, generating a training sample; Based on the training samples, the machine learning model is trained.

12. The method according to claim 9, characterized in that The obtaining of the network characteristic between the first terminal and the third terminal includes: Performing null packet transmission with the third terminal; The network characteristic is obtained based on the transmission status of the empty packet.

13. A terminal device, characterized in that: The terminal device is a first terminal, and the first terminal includes: a transceiver module, configured to obtain, when the first terminal is in a first local cooperative network and the connection of the first terminal is not interrupted, first characteristic data of the first terminal and second characteristic data sent by at least one network node; Processing modules for: Based on the first feature data and the second feature data, obtaining a second local collaborative network; When the first local collaborative network is different from the second local collaborative network, the first local collaborative network is deleted, and the second local collaborative network is networked.

14. An electronic device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, and the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed on a computer, causes the computer to execute the method according to any one of claims 1 to 12.