Data processing method, device, equipment, system and storage medium

Abstract

The invention provides a data processing method, device, equipment and system and a storage medium, and relates to the technical field of unmanned aerial vehicles. The method comprises the following steps: a network element of a core network acquires an inference model sent by a model enabling server, and then divides the inference model into a plurality of sub-models; sending the corresponding sub-model to each unmanned aerial vehicle in the unmanned aerial vehicle group; and the unmanned aerial vehicle performs sensing data reasoning by using the received sub-models to obtain reasoning results of the sub-models. In the scheme, the network element of the core network divides the inference model into a plurality of sub-models and sends the sub-models to the corresponding unmanned aerial vehicles, and the unmanned aerial vehicles perform inference of the sensing data by using the sub-models, so that the situation that the unmanned aerial vehicles send a large amount of sensing data to the ground user side is avoided, the time delay of data inference is reduced, and the time delay of the sensing service of the unmanned aerial vehicles is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of wireless communication technologies, and in particular to a data processing method, apparatus, device, system, and storage medium. Background Art

[0002] The Unmanned Aerial Vehicle (UAV) business is a comprehensive set of technical applications and services that use drone platforms to perform specific tasks through remote control or autonomous control.

[0003] In related technologies, a ground user terminal is typically responsible for controlling and managing a swarm of drones. The swarm transmits a large amount of sensory data to the ground user terminal, which then performs inference on the data.

[0004] Ground user terminals can only rely on local computing resources to process large amounts of perception data. Due to the simple equipment of ground user terminals, they cannot provide powerful computing resources to integrate and analyze and reason about large amounts of perception data, resulting in long reasoning delays. Summary of the Invention

[0005] The present disclosure provides a data processing method, apparatus, device, system and storage medium, which at least to some extent overcome the problem of prolonged inference time in related technologies.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a data processing method is provided, which is applied to a core network network element, including: obtaining an inference model sent by a model enabling server; dividing the inference model into multiple sub-models; and sending the corresponding sub-model to each drone in a drone group, so that the drone uses the received sub-model to perform data processing.

[0008] In an exemplary embodiment of the present disclosure, dividing the reasoning model into multiple sub-models includes: dividing the reasoning model into multiple sub-models according to at least one of the configuration information of the drone group, the configuration information of the reasoning model, and the task requirement information.

[0009] In an exemplary embodiment of the present disclosure, dividing the inference model into a plurality of sub-models includes: dividing the inference model into the plurality of sub-models, and determining a correspondence between a sub-model identifier and a drone identifier; broadcasting the correspondence to a drone group so that the drone determines its corresponding sub-model identifier; and sending the corresponding sub-model to each drone in the drone group includes: receiving, for each drone in the drone group, a sub-model request message sent by the drone, wherein the sub-model request message includes the sub-model identifier; Send the sub-model corresponding to the sub-model identifier to the drone.

[0010] In an exemplary embodiment of the present disclosure, the configuration information of the drone group includes the priority of each drone in the drone group; determining the correspondence between the sub-model identifier and the drone identifier includes: determining the correspondence between the sub-model identifier and the drone identifier based on the granularity of the sub-model and the priority of the drone.

[0011] In an exemplary embodiment of the present disclosure, the priority of the drone is determined by the remaining energy of the drone and / or the distance between the drone and the core network element.

[0012] In an exemplary embodiment of the present disclosure, the granularity of the sub-model is inversely proportional to the priority of the drone.

[0013] In an exemplary embodiment of the present disclosure, obtaining the inference model transmitted by the model enabling server includes: determining the model identifier of the inference model that matches the drone group; sending a model selection signaling to the model enabling server, wherein the model selection signaling includes the model identifier of the inference model; and obtaining the inference model corresponding to the model identifier sent by the model enabling server.

[0014] In an exemplary embodiment of the present disclosure, obtaining the inference model sent by the model enabling server includes: obtaining an updated inference model sent by the model enabling server, wherein the updated inference model is determined by updated configuration information of the drone group received by the model enabling server.

[0015] In an exemplary embodiment of the present disclosure, determining the model identification of the inference model that matches the drone group includes: determining the model identification of the inference model that matches the drone group based on at least one of task requirement information, configuration information of the drone group, and configuration information of the model enabling server.

[0016] In an exemplary embodiment of the present disclosure, a model identifier of an inference model that matches a drone group is determined based on at least one of task requirement information, configuration information of the drone group, and configuration information of a model enabling server, including: receiving a drone group registration request message sent by a drone, wherein the drone group registration request message includes: task requirement information and configuration information of the drone group; sending a drone group registration request message to the model enabling server; after receiving a drone group registration reply message sent by the model enabling server, determining the model identifier of an inference model that matches the drone group based on at least one of task requirement information, configuration information of the drone group, and configuration information of the model enabling server; and sending a drone group registration reply message to the drone group.

[0017] In an exemplary embodiment of the present disclosure, the configuration information of the drone group includes the user information to which the drone group belongs; sending a drone group registration request message to the model enabling server includes: determining a model enabling server that meets the requirements based on the user information to which the drone group belongs; and sending a drone group registration request message to the model enabling server that meets the requirements.

[0018] In an exemplary embodiment of the present disclosure, the method further includes: sending the inference model to the ground user terminal corresponding to the drone group, so that the ground user terminal receives and aggregates the inference results sent by each drone.

[0019] According to one aspect of the present disclosure, a data processing method is provided, which is applied to a drone, including: receiving a sub-model sent by a core network network element, where the sub-model is obtained by the core network network element by splitting an inference model, and the inference model is a model matching a drone group obtained by the core network network element from a model enabling server; obtaining perception data; performing inference based on the sub-model and the perception data to obtain an inference result of the sub-model; and sending the inference result of the sub-model to a ground user terminal, so that the ground user terminal can aggregate the inference results of each sub-model.

[0020] In some exemplary embodiments of the present disclosure, receiving a sub-model sent by a core network network element includes: receiving the correspondence between the sub-model identifier broadcast by the core network network element and the drone identifier; determining the sub-model identifier corresponding to the drone based on the correspondence; sending a sub-model request message to the core network network element, wherein the sub-model request message includes the sub-model identifier; and receiving the sub-model corresponding to the sub-model identifier sent by the core network network element.

[0021] In some exemplary embodiments of the present disclosure, it also includes: obtaining shared information of each drone in the drone group; determining the configuration information of the drone group based on the shared information of each drone; sending a drone group registration request message to the core network network element, wherein the drone group registration request message is used to instruct the core network network element to forward the drone group registration request message to the model enabling server, so that the core network network element obtains the model identifier of the inference model matching the drone group according to at least one of the task requirement information, the configuration information of the drone group and the configuration information of the model enabling server after receiving the drone group registration reply message sent by the model enabling server; and receiving the drone group registration reply message sent by the core network network element.

[0022] In some exemplary embodiments of the present disclosure, it also includes: receiving updated configuration information of the drone group; sending the updated configuration information to the model enabling server, so that the model enabling server determines the updated reasoning model according to the updated configuration information of the drone group, and sends the updated reasoning model to the core network.

[0023] In some exemplary embodiments of the present disclosure, it also includes: periodically sending the configuration information of the drone group to the model enabling server, so that the model enabling server determines whether the configuration information of the drone group is updated; when the configuration information of the drone group is updated, determining an updated inference model according to the updated configuration information of the drone group, and sending the updated inference model to the core network.

[0024] According to another aspect of the present disclosure, a data processing method is provided, which is applied to a model enabling server, including: determining an inference model that matches a drone group; transmitting the inference model to a core network network element so that the core network network element divides the inference model into multiple sub-models, and sending the corresponding sub-model to each drone in the drone group so that the drone uses the received sub-model to perform data processing.

[0025] In some exemplary embodiments of the present disclosure, determining the inference model that matches the drone group includes: receiving model selection signaling sent by a core network element, wherein the model selection signaling includes a model identifier of the inference model that matches the drone group; and determining the inference model corresponding to the model identifier.

[0026] In some exemplary embodiments of the present disclosure, determining a reasoning model that matches a drone group includes: receiving updated configuration information of the drone group; and determining an updated reasoning model based on the updated configuration information of the drone group.

[0027] In some exemplary embodiments of the present disclosure, determining an inference model that matches a drone group includes: periodically receiving configuration information of the drone group sent by drones; and when the configuration information of the drone group is updated, determining an updated inference model based on the updated configuration information of the drone group.

[0028] In some exemplary embodiments of the present disclosure, it also includes: receiving a drone group registration request message sent by a core network network element; sending a drone group registration reply message to the core network network element, so that after the core network network element receives the drone group registration reply message sent by the model enabling server, it obtains the model identifier of the inference model matching the drone group based on the task requirement information, the configuration information of the drone group and at least one of the configuration information of the model enabling server, and forwards the drone group registration reply message to the drone group.

[0029] According to another aspect of the present disclosure, a data processing system is provided, which includes a core network network element, a model enabling server, a drone and a ground user terminal; the model enabling server is used to determine an inference model that matches a drone group; the core network network element is used to obtain the inference model sent by the model enabling server; divide the inference model into multiple sub-models; send the corresponding sub-model to each drone in the drone group; the drone is used to use the received sub-model to perform data processing to obtain the inference results of each sub-model; the ground user terminal is used to receive and aggregate the inference results of each sub-model.

[0030] According to another aspect of the present disclosure, a data processing device is provided, which is configured in a core network network element and includes: a model acquisition module for acquiring an inference model sent by a model enabling server; a model segmentation module for dividing the inference model into multiple sub-models; and a sub-model sending module for sending a corresponding sub-model to each drone in a drone group, so that the drone uses the received sub-model to perform data processing.

[0031] According to another aspect of the present disclosure, a data processing device is provided, which is configured on a drone and includes: a sub-model receiving module, used to receive a sub-model sent by a core network network element, where the sub-model is obtained by the core network network element by splitting an inference model, and the inference model is a model that matches the drone group and is obtained by the core network network element from a model enabling server; a perception data acquisition module, used to obtain perception data; a data inference module, used to perform inference based on the sub-model and the perception data to obtain an inference result of the sub-model; and an inference result sending module, used to send the inference result of the sub-model to a ground user terminal, so that the ground user terminal can aggregate the inference results of each sub-model.

[0032] According to another aspect of the present disclosure, a data processing device is provided, which is configured on a model enabling server and includes: an inference model determination module, used to determine an inference model that matches a drone group; an inference model sending module, used to transmit the inference model to a core network network element, so that the core network network element divides the inference model into multiple sub-models, and sends the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to perform data processing.

[0033] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any one of the above-mentioned data processing methods by executing the executable instructions.

[0034] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above-mentioned data processing methods is implemented.

[0035] According to another aspect of the present disclosure, a computer program product is provided, including: a computer program or an instruction, and any one of the above data processing methods when the computer program or the instruction is executed by a processor.

[0036] In the data processing method provided in the embodiments of the present disclosure, a core network element obtains an inference model sent by a model enabling server, then divides the inference model into multiple sub-models. The corresponding sub-models are then sent to each drone in the drone group. The drones then use the received sub-models to perform data processing and obtain inference results for each sub-model. In this solution, the core network element divides the inference model into multiple sub-models and sends them to the corresponding drones. The drones then use the sub-models to infer perception data, avoiding the need for drones to send large amounts of perception data to ground users, reducing data inference latency, and lowering the latency of drone perception services.

[0037] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0039] Figure 1 A flow chart of a data processing method applied in a core network element according to an embodiment of the present disclosure is shown; Figure 2 A flow chart showing another data processing method applied in a core network element according to an embodiment of the present disclosure is shown; Figure 3 A flow chart of a data processing method applied in a core network element according to another embodiment of the present disclosure is shown; Figure 4 A flow chart showing another data processing method applied in a core network element according to an embodiment of the present disclosure is shown; Figure 5 A flowchart of a drone group registration method according to an embodiment of the present disclosure is shown; Figure 6 A flow chart of a data processing method applied in a drone according to an embodiment of the present disclosure is shown; Figure 7 A flow chart showing another data processing method applied in a drone according to an embodiment of the present disclosure is shown; Figure 8A flow chart of a data processing method applied in a drone according to another embodiment of the present disclosure is shown; Figure 9 A flow chart of another data processing method applied in a drone according to an embodiment of the present disclosure is shown; Figure 10 A flow chart of a data processing method applied in a model enabling server according to an embodiment of the present disclosure is shown; Figure 11 A flow chart showing another data processing method applied in a model enabling server according to an embodiment of the present disclosure is shown; Figure 12 A flow chart of a data processing method applied in a model enabling server according to another embodiment of the present disclosure is shown; Figure 13 A flow chart of another data processing method applied in a model enabling server according to an embodiment of the present disclosure is shown; Figure 14 A data interaction flow chart applied in a data processing system according to an embodiment of the present disclosure is shown; Figure 15 A block diagram showing an application of the present disclosure in a data processing system is shown; Figure 16 A schematic diagram showing a drone group registration process in an embodiment of the present disclosure; Figure 17 A schematic diagram showing a drone group management process in an embodiment of the present disclosure; Figure 18 A schematic diagram of a data processing device according to an embodiment of the present disclosure is shown; Figure 19 A schematic diagram of another data processing device according to an embodiment of the present disclosure is shown; Figure 20 A schematic diagram of another data processing device according to an embodiment of the present disclosure is shown; Figure 21 A structural block diagram of a computer device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0040] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0041] In addition, the accompanying drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0042] For ease of understanding, before introducing the embodiments of the present disclosure, the terms involved in the embodiments of the present disclosure are first explained as follows: Data Sensing: Detecting and sensing the surrounding environment through user devices and / or base stations, then storing the data locally or uploading it to the network for processing to obtain deeper insights. Data sensing includes, but is not limited to, target location, speed, and trajectory tracking. It is widely used in in-vehicle internet, drone control, smart homes, health monitoring, environmental monitoring, and large-scale Internet of Things.

[0043] Figure 1 A data processing method flow chart of an embodiment of the present disclosure is shown. An embodiment of the present disclosure provides a data processing method that can be executed by any core network element with computing and processing capabilities. Figure 1 As shown, the data processing method provided in the embodiment of the present disclosure includes the following steps.

[0044] S102: Obtain the inference model sent by the model enabling server.

[0045] Core network elements can be understood as the basic units or components that make up a telecommunications network, including but not limited to base stations and switches. Each element has specific functions, such as signal processing and user data management. Optionally, this embodiment uses the Network Data Analytics Function (NWDAF) as an example. The NWDAF is a standardized functional entity in the core network that collects and analyzes network data to provide intelligent decision support.

[0046] In the fields of artificial intelligence and machine learning, a model can be understood as a computing system generated by training data, capable of making predictions or decisions based on sensory data. Inference refers to the process of using a trained model to predict or classify new sensory data. An inference model can be understood as a trained machine learning or deep learning model used to perform inference tasks. Inference models include, but are not limited to, image recognition models, classification models, and language models. Optionally, the inference model described above is an AI (artificial intelligence) model.

[0047] A model-enabling server can be understood as a server or service framework that provides specific support and services to facilitate model deployment, operation, and management in the field of artificial intelligence and machine learning. A model-enabling server provides the necessary environment and support for model usage, such as distributing inference models, optimizing inference model parameters, and monitoring status. Exemplarily, a model-enabling server may include an AI-enabling server.

[0048] In some exemplary embodiments of the present disclosure, a model enabling server is used to store and manage inference models. When a core network element needs to use an inference model, the core network element sends a model selection instruction to the model enabling server, and the model enabling server sends the inference model corresponding to the model selection instruction to the core network element. The core network element's use of the inference model can be understood as the core network element forwarding the inference model to the drone group.

[0049] In some exemplary embodiments of the present disclosure, when the model enabling server determines that the configuration information of the drone group is updated, the model parameters of the inference model are updated according to the updated configuration information to obtain an updated inference model, and the updated inference model is sent to the core network element.

[0050] S104: Divide the inference model into multiple sub-models.

[0051] Submodels can be thought of as dividing a larger inference model into several smaller parts or modules. Each submodel processes a specific task or portion of data. In the context of a drone swarm, different drones receive different submodels to perform specific data processing tasks.

[0052] In this embodiment, each sub-model corresponds to a drone, that is, each sub-model will be sent to a drone and deployed in the drone, so that the drone can perform a specific reasoning task.

[0053] In an exemplary embodiment of the present disclosure, the inference model is divided into multiple sub-models according to the number of drones in the drone group, wherein the number of sub-models is the same as the number of drones in the drone group.

[0054] In an exemplary embodiment of the present disclosure, an inference model is divided into multiple sub-models according to the model's hierarchical structure. For example, a CNN (Convolutional Neural Network) model is divided into three sub-models according to the model's hierarchical structure: Sub-model 1 is used for feature extraction in the first few layers; Sub-model 2 is used for feature combination in the middle layers; and Sub-model 3 is used for classification or regression tasks in the last few layers.

[0055] In an exemplary embodiment of the present disclosure, a complete inference model is split based on the user's task requirements in actual applications to generate multiple sub-models. For example, the user's requirements are: rapid detection of abnormal behavior; accurate analysis of the specific type of abnormal behavior. The inference model is divided into sub-model 1 and sub-model 2. Sub-model 1 is a lightweight model used to quickly detect whether a moving object is abnormal and is a low-precision, low-latency model; sub-model 2 is a heavyweight model used to analyze the specific type of abnormal behavior in detail and is a high-precision, high-latency model.

[0056] The inference model is divided into multiple sub-models, each sub-model corresponds to a drone in the drone group.

[0057] S106: Send the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to perform data processing.

[0058] A drone swarm can be understood as a group of drone systems connected via a network and working together. Each drone in the swarm can communicate with each other and work together to complete specific tasks based on pre-set algorithms or instructions.

[0059] The drone uses the received sub-model to perform data processing, which can be understood as the process of analyzing, classifying, and identifying the collected data using the received sub-model. This includes but is not limited to image recognition, environmental perception, path planning, and other tasks.

[0060] The core network elements transmit the multiple sub-models to the corresponding drones via communication links. After receiving the corresponding sub-model, each drone loads it into its local computing environment and uses it to infer the perception data to obtain the inference results.

[0061] The communication technologies used between core network elements and drones include but are not limited to: cellular networks, satellite communications, Wi-Fi, point-to-point communications, ad hoc networks, etc.

[0062] In this embodiment, the core network element obtains the inference model sent by the model enablement server, then divides the inference model into multiple sub-models. The corresponding sub-model is then sent to each drone in the drone group. The drone then processes the received sub-model to obtain the inference results of each sub-model. Because each drone can perform specific inference tasks using its own deployed sub-model, there is no need to send perception data to the ground user end for inference. This avoids the need for drones to send large amounts of perception data to the ground user end, reduces data inference latency, and lowers the latency of drone perception services.

[0063] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 2 As shown, the optimized data processing method includes the following steps.

[0064] S202: Obtain the inference model sent by the model enabling server.

[0065] S204: Divide the inference model into multiple sub-models according to at least one of the configuration information of the drone group, the configuration information of the inference model, and the task requirement information.

[0066] The configuration information of a drone group can be understood as information that describes the overall and individual capabilities, status, and related parameters of the drone group, including but not limited to: the location information of the drone group, the number of drones in the group, the type of drone, the identification of the drone, the hardware resources of the drone, the endurance information of the drone, etc.

[0067] The configuration information of the inference model can be understood as information that describes the structure, function, performance, and resource requirements of the inference model itself, including but not limited to: the model architecture, input and output parameters, computing resource requirements, communication requirements, etc. of the inference model.

[0068] Mission requirements refer to the goals, constraints, and evaluation criteria set to complete a specific mission, including but not limited to mission objectives, geographical scope, time constraints, accuracy requirements, resource constraints, environmental conditions, etc.

[0069] In an exemplary embodiment of the present disclosure, the segmentation method based on the drone group configuration information includes but is not limited to: segmentation based on the hardware resources of each drone and segmentation based on the sensor type.

[0070] Based on the hardware resource allocation of each drone, if there are differences in hardware resources within the drone group, the model can be divided into lightweight sub-models and complex sub-models. Lightweight sub-models are suitable for running on drones with limited computing resources, while complex sub-models are suitable for running on drones with strong computing power.

[0071] Segmentation by sensor type: a sub-model corresponding to each sensor type. Drones are equipped with different types of sensors, such as cameras, radars, and lidars. Models can be segmented based on the characteristics of the sensor input data. The segmentation results include: an image processing sub-model for processing camera data, a signal processing sub-model for processing radar or lidar data, and so on.

[0072] In an exemplary embodiment of the present disclosure, the segmentation method based on the inference model configuration information includes but is not limited to: segmentation by functional modules and segmentation by real-time requirements.

[0073] If an inference model contains multiple independent functional modules, such as feature extraction, classification, and regression, it can be segmented according to the functional boundaries of the modules. The segmentation results include: a feature extraction sub-model for extracting high-level features from the raw data, and a classification / regression sub-model for performing classification or regression predictions.

[0074] If some modules in the inference model have high real-time requirements, they can be treated as a separate sub-model and ensure that this part can run under the conditions that meet the real-time requirements.

[0075] In an exemplary embodiment of the present disclosure, the segmentation method based on task requirement information includes but is not limited to: segmentation according to task priority and segmentation according to task complexity.

[0076] If a task contains multiple subtasks with different priorities, the model can be split into high-priority and low-priority submodels. The high-priority submodel is used to perform urgent tasks, such as rapid target recognition. The low-priority submodel is used to perform non-urgent tasks, such as background analysis.

[0077] For complex and simple tasks, sub-models of different complexity can be divided. Simple task sub-models are used to handle basic tasks such as obstacle detection. Complex task sub-models are used to handle difficult tasks such as multi-target tracking.

[0078] Segmenting the inference model based on the configuration information of the drone group, the configuration information of the inference model, and the mission requirement information can not only improve the overall performance of the drone group, but also enhance its adaptability and flexibility to better meet diverse mission requirements.

[0079] S206: Determine the correspondence between the sub-model identifier and the drone identifier.

[0080] A submodel ID can be understood as a unique identifier for each submodel separated from the inference model. This ID is used to distinguish different submodels and manage and schedule them. A drone ID can be understood as a unique identifier for each drone in a drone group. Each drone may have different hardware configurations, location information, or mission status. The drone ID is used to identify and locate individual drones within the group.

[0081] The correspondence can be understood as associating each sub-model ID with the ID of the drone it is suitable for operation. This includes but is not limited to: matching the computational requirements of the sub-model with the computational capabilities of the drone; adapting the data processing type of the sub-model to the type of sensor carried by the drone; and assigning different sub-models to drones covering different areas.

[0082] Analyze the resource requirements of each sub-model and the available resources of each drone to ensure that the sub-model can run efficiently on the corresponding drone. Prioritize the sub-models on which drones to deploy based on the specific mission requirements. For example, for sub-models that require fast response times, select drones with strong computing power and close to the mission area.

[0083] Create a mapping table or use a data structure to record the correspondence between each sub-model ID and the drone ID.

[0084] S208: Broadcast the corresponding relationship to the drone group so that the drones can determine their corresponding sub-model identifiers.

[0085] Broadcasting is a one-to-many communication mode that can be used to efficiently transmit information to all drone members in a drone group. Broadcasting methods include but are not limited to: radio signal broadcasting, dedicated protocol broadcasting, multi-hop network broadcasting, and Internet-based broadcasting.

[0086] In an exemplary embodiment of the present disclosure, if a group of drones is relatively concentrated and operating within a short distance, radio signals can be broadcast directly from the core network element to all drones. For a more widely distributed group of drones, multi-hop network technology can be used, which uses drones to forward information to each other to expand signal coverage.

[0087] In an exemplary embodiment of the present disclosure, a message containing the correspondence between sub-model identifiers and drone identifiers is transmitted via direct radio broadcast or multi-hop network propagation. To ensure that each drone successfully receives the correspondence, a confirmation mechanism can be incorporated into the communication protocol. For example, after receiving the correspondence, the drone transmits a confirmation signal back to the core network element to indicate receipt of the correspondence.

[0088] After receiving the correspondence, each drone parses the message content and extracts the sub-model identifier corresponding to its own drone identifier.

[0089] S210: Receive a sub-model request message from each drone in the drone group, wherein the sub-model request message includes a sub-model identifier.

[0090] After each drone receives the broadcasted correspondence, it generates a sub-model request message. The sub-model request message includes but is not limited to: sub-model identifier and drone identifier. The sub-model identifier is used to indicate the sub-model that the drone needs to load, and the drone identifier is used to identify the drone that sends the request. The UAV directly sends a sub-model request message to the core network element, and the core network element receives the sub-model request message sent by each UAV.

[0091] S212: Send the sub-model corresponding to the sub-model identifier to the UAV.

[0092] The core network element searches for the corresponding sub-model data in local storage based on the sub-model identifier in the sub-model request message. The sub-model data includes but is not limited to: model structure, weight parameters, configuration files and other information, so that the drone can correctly load and run the sub-model.

[0093] After the core network element locates the corresponding sub-model data, it sends it to the drone. Upon receiving the sub-model data from the core network element, the drone performs an integrity check. If the data passes the integrity check, the drone saves the sub-model data to local storage and prepares to load the sub-model to perform inference tasks using the sub-model. If the integrity check fails, the drone notifies the core network element and requests that the problematic data be resent.

[0094] In this embodiment, by broadcasting the corresponding relationship, the drone requests the sub-model it needs, avoiding indiscriminate acquisition of data of the entire inference model, reducing the amount of data transmitted over the network, and reducing the risk of network congestion.

[0095] Based on the above embodiment, this embodiment further optimizes the above step S206, such as Figure 3 As shown, the method for determining the optimized corresponding relationship includes the following steps.

[0096] S302: Obtain the priority of each drone in the drone group from the configuration information of the drone group. The priority of the drone is determined by the remaining energy of the drone and / or the distance between the drone and the core network element.

[0097] The priority of a drone can be understood as the degree to which each drone is sorted or classified when performing a task in a drone group, which determines whether the drone should be given priority to perform certain tasks, obtain resources or receive instructions.

[0098] A drone's remaining energy can be understood as the amount of energy remaining in its battery or energy storage system. When allocating resources in the sub-model, drones with high remaining energy levels are given higher priority because they can perform longer missions without returning to recharge or replace batteries, allowing them to take on more critical or time-consuming tasks. Conversely, drones with low remaining energy levels are assigned a lower priority to ensure their limited energy resources are used only for essential operations, avoiding mission failure or drone loss due to energy depletion.

[0099] The distance between a drone and core network elements refers to the distance from the drone to the core network elements responsible for command, control, and data transmission. Drones with closer proximity are given higher priority, enabling more efficient communication, reducing signal latency, and consuming less energy for data transmission. Drones farther from core network elements are given lower priority due to higher communication costs, unless a specific mission requires a drone farther from the core network.

[0100] In an exemplary embodiment of the present disclosure, the drones in a drone group are networked and information is shared between the drones. One drone in the drone group receives the shared information from other drones, sorts the drones from large to small according to their remaining energy, obtains the priority of each drone, and sends the priority of each drone to the core network element.

[0101] In an exemplary embodiment of the present disclosure, the drones in a drone group are networked and information is shared between the drones. One drone in the drone group receives the shared information from other drones, and sorts the drones from small to large according to the distance between the drones and the core network element to obtain the priority of each drone, and sends the priority of each drone to the core network element.

[0102] In this embodiment, the remaining energy of drones is considered to determine their priority. This allows drones with sufficient power to download high-granularity sub-models first, maximizing the utilization of each drone's energy resources and preventing mission interruptions or failures due to insufficient power. The distance between drones and core network elements is also considered a factor in the priority process, allowing closer drones to be assigned tasks requiring frequent communication. This reduces communication latency and minimizes the energy consumption associated with long-distance transmission.

[0103] S304: Determine the correspondence between the sub-model identifier and the drone identifier based on the granularity of the sub-model and the priority of the drone. The granularity of the sub-model is inversely proportional to the priority of the drone.

[0104] The granularity of a submodel can be understood as the functionality, data volume, or computational complexity of each submodel when a large inference model is split into multiple submodels. Granularity can be either coarse or fine. Coarse granularity means each submodel contains more functionality or data, while fine granularity means each submodel contains less functionality or data.

[0105] When determining the correspondence between sub-model identifiers and drones, the granularity of each sub-model is first determined, and then matched based on the priority of each drone. Sub-models with larger granularity and higher complexity are assigned to drones with more remaining energy and located closer to the core network to ensure efficient task completion. Meanwhile, sub-models with smaller granularity and relatively simpler features can be flexibly assigned to drones with slightly lower priorities based on actual conditions, effectively utilizing resources while ensuring that all tasks are properly handled.

[0106] In this embodiment, task offloading of different granularities is achieved in the drone group, which effectively reduces the latency of drone communication, perception and reasoning integration, optimizes user experience, and reduces the energy consumption of drone services to a certain extent.

[0107] In this embodiment, the optimal allocation of resources is achieved according to the granularity of the sub-model and the priority of the drones, ensuring the effective execution of the tasks while improving the overall operational efficiency and flexibility.

[0108] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 4 As shown, the optimized data processing method includes the following steps.

[0109] S402: Determine the model identifier of the inference model that matches the drone group.

[0110] A model identifier is a unique identifier or name for an inference model, used to distinguish different models. Each model identifier corresponds to a specific model structure, parameter configuration, and other characteristics, including but not limited to: model name, ID number, version number, or other identifiers.

[0111] The inference model that matches the drone group can be understood as a model that needs to be deployed in the drone group and performs data inference based on the perception data collected by the drone group.

[0112] In some exemplary embodiments of the present disclosure, an inference model corresponding to a drone group is determined based on the mission requirements of the drone group. For example, if a drone group needs to perform large-scale farmland monitoring, an inference model that processes large amounts of image data and identifies crop health status is selected as the inference model that matches the drone group.

[0113] In some exemplary embodiments of the present disclosure, the inference model that matches the drone group is determined based on the drones' sensor types, computing capabilities, battery life, flight speed, communication capabilities, etc. For example, if the drone group has strong local computing capabilities, a more complex inference model is selected as the inference model that matches the drone group; if computing resources are limited, a more efficient lightweight model is selected as the inference model that matches the drone group.

[0114] In some exemplary embodiments of the present disclosure, the computation of the inference model may rely on a model-enabled server. Therefore, the inference model that matches the drone group can be determined based on the model server's processor performance, memory size, storage capacity, network bandwidth, and other factors. For example, if the model-enabled server is very powerful, it can support complex and computationally intensive models; if server resources are limited, a simplified model more suitable for remote execution may be selected.

[0115] S404: Send a model selection signaling to the model enabling server, where the model selection signaling includes a model identifier of the inference model.

[0116] Model selection signaling is a communication instruction used to notify the model enabling server to select and issue an inference model.

[0117] The core network element embeds the model identifier into the model selection signaling, and sends the model selection signaling to the model enabling server through the network or other communication technologies. After receiving the model selection signaling, the model enabling server parses the model selection signaling to obtain the model identifier, and sends the inference model corresponding to the model identifier to the core network element.

[0118] S406: Obtain the inference model corresponding to the model identifier transmitted by the model enabling server.

[0119] The core network element sends a model selection signaling to the model enabling server. The model selection signaling includes a model identifier of the inference model. The model selection signaling is used to request to obtain the inference model corresponding to the model identifier.

[0120] After the model enabling server receives the model selection signaling, it finds the corresponding inference model according to the model identifier, and sends the found inference model to the core network element through the network or other communication methods.

[0121] The model enabling server sends the inference model including: sending the model architecture, model parameters, model identifier, relevant information of the drone group, and relevant information of the ground user terminal corresponding to the drone group.

[0122] In this embodiment, by sending model selection signaling to the model enabling server, all inference models are uniformly managed and distributed by the server, reducing the need for local storage and maintenance of models by drones and reducing the complexity of the system.

[0123] S408: Divide the inference model into multiple sub-models.

[0124] S410: Send the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to perform perception data inference.

[0125] S412: Send the inference model to the ground user terminal corresponding to the drone group, so that the ground user terminal receives and aggregates the inference results sent by each drone.

[0126] A ground user terminal (UTE) refers to a user device or software terminal located on the ground. It can communicate with core network elements or drones. Examples include, but are not limited to, smartphones, tablets, laptops, desktops, dedicated remote controls, vehicle-mounted terminals, wearable devices, and industrial control systems.

[0127] The inference result is the output generated by the drone based on its sensor data and sub-models. For example, the result of object detection might be “object A was found”, or the result of path planning might be “optimal flight path recommended”.

[0128] Aggregation refers to the integration, analysis, and processing of inference results from multiple drones to generate a global inference result. Aggregation methods include, but are not limited to, the fusion of inference results, the fusion of inference parameters, and the fusion of multi-dimensional inference parameters and inference results.

[0129] In a potential application scenario, during a search and rescue mission, each drone may only cover a small area. By aggregating the search results of all drones, the ground user can generate a complete search map and determine the location of the target.

[0130] After perception, inference and identification, each drone in the drone group sends the inference results of each sub-model to the ground user, which aggregates them to obtain detailed and accurate inference results. Since only the inference results are sent to the ground user, the transmission of large amounts of perception data is avoided, communication traffic is reduced, and communication latency is shortened.

[0131] S414: Obtain the updated reasoning model transmitted by the model enabling server, and return to execute S408.

[0132] The updated inference model is determined by the updated configuration information of the drone group received by the model enabling server.

[0133] While a drone group is executing a mission, the drones report their configuration information to the model server. If the configuration is updated, the model enablement server adjusts the parameters of the inference model based on the updated configuration information, generating an updated inference model. The model enablement server then sends the updated inference model to the core network element. The core network element receives the updated inference model from the model enablement server, divides it into multiple sub-models, and sends them to the corresponding drones. Each drone then uses the received sub-model to execute its own inference task.

[0134] In this embodiment, while a drone group is executing a mission, the inference model can be updated based on the group's configuration information, thereby updating the sub-models of each drone. When a drone leaves the group or a new drone joins, the updated inference model can quickly adjust the strategy, ensuring stable operation of the entire drone group and improving inference efficiency.

[0135] Based on the above embodiment, this embodiment provides a drone group registration method, such as Figure 5 As shown, the drone group registration method provided in this embodiment includes the following steps.

[0136] S502: Receive a drone group registration request message sent by a drone, where the drone group registration request message includes: mission requirement information and drone group configuration information.

[0137] A registration request message can be understood as a message sent by a drone group to the model enablement server to register or apply for participation in a specific mission. Mission requirement information can be understood as the specific information of the task to be completed, including but not limited to: mission objectives, expected results, execution timeframe, etc.

[0138] In one possible implementation, the drone group is initialized after startup, and each drone shares information such as location, hardware resources, and remaining energy. The drone with the highest remaining energy communicates with the core network element and registers on the model enablement server platform.

[0139] In one possible implementation, the drone sends a drone group registration request message to the NWDAF in the core network through the base station. The drone group registration request message includes the location of the drone group, the hardware resources of each drone, the remaining energy of each drone, the user information of the drone group, and the mission requirement information.

[0140] S504: Send a drone group registration request message to the model enabling server.

[0141] In some exemplary embodiments of the present disclosure, sending a drone group registration request message to a model enabling server includes the following steps.

[0142] S5042. Determine a model enabling server that meets the requirements based on the user information of the drone group.

[0143] The model-enabled server that meets the requirements can be understood as the model-enabled server closest to the location of the drone group and the model-enabled server closest to the ground user terminal.

[0144] In one possible implementation, NWDAF searches for a model-enabled server based on the various information included in the drone group registration request message, and searches for a deployed model-enabled server that is closest to the ground user terminal or the drone group.

[0145] S5044. Send a drone group registration request message to the model enabling server that meets the requirements.

[0146] Based on the search results in step S5022, a drone group registration request message is sent to the model-enabled server closest to the ground user terminal or the drone group. The drone group registration request message includes the location of the drone group, the hardware resources of each drone, the remaining energy of each drone, information about the users to whom the drone group belongs, and mission requirements.

[0147] In this embodiment, by selecting a model-enabled server that is closest to a ground user terminal or a drone group, data transmission delay can be reduced and the response speed of the drone group can be improved.

[0148] S506. After receiving the drone group registration reply message sent by the model enabling server, determine the model identifier of the inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server.

[0149] After receiving the drone group registration request message, the model enabling server stores the configuration information and user information of the drone group locally, generates the corresponding drone group number information, and then sends a drone group registration reply message to the core network element. The drone group registration reply message contains the drone group number information and registration success information.

[0150] After receiving the drone group registration reply message, the core network element matches the model according to the configuration information of the drone group, the configuration information of the model enabling server, and the mission requirement information, and determines the inference model that matches the drone group.

[0151] Send model selection signaling to the model enabling server to inform the model enabling server of the model identifier of the selected reasoning model, so that the model enabling service can send the reasoning model corresponding to the model identifier to the core network network element.

[0152] Determining the model identifier of an inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server includes: determining a first inference model set corresponding to the drone group based on the task requirement information to be performed by the drone group. Based on the task requirement, a model such as a linear regression model, a convolutional model, or a large language model is selected as the inference model that matches the drone group. For example, if the drone group needs to perform large-scale farmland monitoring, an inference model that processes large amounts of image data and identifies crop health status is selected as the inference model that matches the drone group. Then, based on the configuration information of the model enabling server, a second inference model set is selected from the first inference model set. For example, the second inference model set includes configuration information of the model enabling server that meets the hardware requirements of each inference model in the second inference model set, or the location of the model enabling server is within the service range of a core network element. Then, based on the configuration information of the drone group, an inference model that matches the drone group is selected from the second inference model set. For example, an inference model that matches the hardware resources of the drones is selected from the second inference model set as the inference model that matches the drone group.

[0153] In this embodiment, the optimal inference model is selected by comprehensively considering the task requirement information, model enabling server configuration information and drone group configuration information, thereby improving the task execution efficiency and system adaptability and enhancing the user experience.

[0154] S508: Send a drone group registration reply message to the drone group.

[0155] The core network element broadcasts the drone group registration reply message to the drones in the drone group through the base station, indicating that the drone registration is successful, and informs the registered model enabling server and the selected model number.

[0156] In this embodiment, a process is provided for a drone group to register with a model enabling server through a core network, thereby realizing centralized management of the drone group, effectively allocating resources and scheduling tasks, and ensuring smoother coordination and cooperation between different tasks.

[0157] Figure 6 A data processing method flow chart of an embodiment of the present disclosure is shown. An embodiment of the present disclosure provides a data processing method that can be executed by any drone with computing and processing capabilities. Figure 6 As shown, the data processing method provided in the embodiment of the present disclosure includes the following steps.

[0158] S602. Receive a sub-model sent by a core network element. The sub-model is obtained by the core network element by dividing the inference model. The inference model is a model that matches the drone group and is obtained by the core network element from the model enabling server.

[0159] In this embodiment, the core network element divides the inference model into multiple sub-models, each sub-model corresponds to a drone, the core network element sends the corresponding sub-model to each drone, and the drone receives the sub-model sent by the core network element.

[0160] The process of the core network element dividing the inference model into multiple sub-models can refer to the description in the above embodiment.

[0161] S604: Acquire perception data.

[0162] Perception data can be understood as information about the environment, objects, or events collected by sensors or other devices deployed on drones. This includes, but is not limited to, images, videos, sounds, temperature, humidity, air pressure, and location information. These sensors include, but are not limited to, cameras, infrared sensors, lidar, ultrasonic sensors, inertial measurement units, and barometers.

[0163] S606: Perform inference based on the sub-model and the perception data to obtain an inference result of the sub-model.

[0164] The sensory data collected in step S604 is input into the sub-model, which processes and analyzes the sensory data based on its internal algorithm or learned knowledge, and finally generates an inference result for the sensory data. The inference result may include: classification label, prediction value, anomaly detection mark, etc.

[0165] S608: Send the inference results of the sub-models to the ground user terminal, so that the ground user terminal aggregates the inference results of the sub-models.

[0166] Aggregation refers to the integration, analysis, and processing of inference results from multiple drones to generate a global inference result. Aggregation methods include, but are not limited to, the fusion of inference results, the fusion of inference parameters, and the fusion of multi-dimensional inference parameters and inference results.

[0167] In a potential application scenario, during a search and rescue mission, each drone may only cover a small area. By aggregating the search results of all drones, the ground user can generate a complete search map and determine the location of the target.

[0168] After performing perception, inference, and recognition, each drone in the group sends its inference results to the ground user, who aggregates them to produce detailed and accurate inference results. Sending only the inference results to the ground user avoids the transmission of large amounts of perception data, reduces communication traffic, and lowers communication latency.

[0169] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 7 As shown, the optimized data processing method includes the following steps.

[0170] S702: Acquire shared information of each drone in the drone group.

[0171] Shared information includes but is not limited to location data, drone identification, remaining energy, etc.

[0172] In an exemplary embodiment of the present disclosure, the drones in a drone group are networked and share information with each other. A drone in the drone group receives the shared information from other drones, sorts the drones from highest to lowest according to their remaining energy, obtains the priority of each drone, and transmits the priority of each drone to a core network element. Exemplarily, the priority of each drone is transmitted to the core network element using the drone with the highest priority.

[0173] It is achieved through wireless communication technologies such as Wi-Fi, radio waves or other dedicated protocols to obtain shared information of each drone in the drone group.

[0174] S704: Determine the configuration information of the drone group based on the shared information of each drone.

[0175] The shared information of each drone is integrated to obtain the configuration information of the drone group. The relevant information of the users belonging to the drone group and the mission requirement information can also be obtained. The shared information of each drone, the relevant information of the users belonging to the drone group, and the mission requirement information are integrated to obtain the configuration information of the drone group.

[0176] S706. Send a drone group registration request message to the core network network element, wherein the drone group registration request message is used to instruct the core network network element to forward the drone group registration request message to the model enabling server, so that after the core network network element receives the drone group registration reply message sent by the model enabling server, it obtains the model identifier of the inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group and the configuration information of the model enabling server.

[0177] A registration request message can be understood as a message sent by a drone group to the model enablement server to register or apply for participation in a specific mission. Mission requirement information can be understood as the specific information of the task to be completed, including but not limited to: mission objectives, expected results, execution timeframe, etc.

[0178] In one possible implementation, the drone group is initialized after startup, and each drone shares information such as location, hardware resources, and remaining energy. The drone with the highest remaining energy communicates with the core network element and registers on the model enablement server platform.

[0179] In one possible implementation, the drone sends a drone group registration request message to the NWDAF in the core network through the base station. The drone group registration request message includes the location of the drone group, the hardware resources of each drone, the remaining energy of each drone, the user information of the drone group, and the mission requirement information.

[0180] In one possible implementation, the NWDAF searches for model-enabled servers based on the information included in the drone group registration request message, finding the deployed model-enabled server closest to the ground user or the drone group. The drone group registration request message is sent to the model-enabled server closest to the ground user or the drone group. The drone group registration request message includes the location of the drone group, the hardware resources of each drone, the remaining energy of each drone, information about the users to whom the drone group belongs, and mission requirements.

[0181] After receiving the drone group registration request message, the model enabling server stores the configuration information and user information of the drone group locally, generates the corresponding drone group number information, and then sends a drone group registration reply message to the core network element. The drone group registration reply message contains the drone group number information and registration success information.

[0182] After receiving the drone group registration reply message, the core network element matches the model based on the drone group configuration information, the model enablement server configuration information, and the mission requirement information to determine the inference model that matches the drone group. It then sends a model selection signaling to the model enablement server, informing the model enablement server of the model identifier of the selected inference model, so that the model enablement service can send the inference model corresponding to the model identifier to the core network element.

[0183] S708. Receive the drone group registration reply message sent by the core network element.

[0184] After receiving the drone group registration reply message, the core network element broadcasts the drone group registration reply message to the drones in the drone group through the base station, indicating that the drone registration is successful, and informs the registered model enabling server and the selected model number.

[0185] In this embodiment, a process is provided for a drone group to register with a model enabling server through a core network, thereby realizing centralized management of the drone group, effectively allocating resources and scheduling tasks, and ensuring smoother coordination and cooperation between different tasks.

[0186] S710. Receive the correspondence between the sub-model identifier and the drone identifier broadcast by the core network element.

[0187] After the Model Enabling Service sends the inference model corresponding to the model identifier to the core network element, the core network element divides the inference model into multiple sub-models and determines the correspondence between the sub-model identifiers and the drone identifiers. This correspondence is broadcast to the drone group, and each drone can receive the correspondence, allowing the drone to determine its corresponding sub-model identifier.

[0188] S712: Determine the sub-model identifier corresponding to the UAV according to the corresponding relationship.

[0189] After receiving the correspondence, each drone parses the message content and extracts the sub-model identifier corresponding to its own drone identifier.

[0190] S714: Send a sub-model request message to the core network element, where the sub-model request message includes a sub-model identifier.

[0191] After each drone receives the broadcasted correspondence, it generates a sub-model request message containing the following information: sub-model ID and drone ID. The sub-model ID is used to indicate the sub-model that the drone needs to load, and the drone ID is used to identify the drone that sends the request.

[0192] The UAV directly sends a sub-model request message to the core network element, and the core network element receives the sub-model request message sent by each UAV.

[0193] S716: Receive the sub-model corresponding to the sub-model identifier sent by the core network element.

[0194] The core network element searches for the corresponding sub-model data in local storage based on the sub-model identifier in the sub-model request message. The sub-model data includes but is not limited to: model structure, weight parameters, configuration files and other necessary information to ensure that the drone can correctly load and run the sub-model.

[0195] After the core network element locates the corresponding sub-model data, it sends it to the drone. Upon receiving the sub-model data from the core network element, the drone performs an integrity check. If the data passes the integrity check, the drone saves the sub-model data to local storage and prepares to load the sub-model to begin the mission. If the integrity check fails, the drone notifies the core network element and requests that the problematic data be resent.

[0196] In this embodiment, by broadcasting the corresponding relationship, the drone requests the sub-model it needs, avoiding indiscriminate acquisition of data of the entire inference model, reducing the amount of data transmitted over the network, and reducing the risk of network congestion.

[0197] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 8 As shown, the optimized data processing method includes the following steps.

[0198] S802: Receive updated configuration information of the drone group.

[0199] During a drone group mission, each drone in the group shares its location and online information. When refreshing via the ground user terminal, drones can discover newly added drones or offline drones in the group. Based on the information about newly added or offline drones, the drone group's updated configuration is retrieved.

[0200] S804. Send the updated configuration information to the model enabling server, so that the model enabling server determines the updated reasoning model according to the updated configuration information of the drone group, and sends the updated reasoning model to the core network.

[0201] The drones report the updated configuration information of the drone group to the model server. The model enabling server adjusts the model parameters of the inference model based on the updated configuration information of the drone group, generating an updated inference model. The model enabling server sends the updated inference model to the core network element. The core network element receives the updated inference model from the model enabling server, divides the updated inference model into multiple sub-models, and sends them to the corresponding drones. Each drone uses the received sub-model to perform its own inference task.

[0202] In this embodiment, the drone reports the configuration information of the drone group, and the model enabling server can quickly adjust the inference model through the updated configuration information, ensuring that the sub-model sent to the drone is always consistent with the current status of the drone group, thereby improving the inference accuracy.

[0203] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 9 As shown, the optimized data processing method includes the following steps.

[0204] S902: Periodically receive configuration information of the drone group.

[0205] During the process of the drone group performing a mission, each drone in the drone group shares location information and online information, and the drone obtains the configuration information of the drone group in real time or periodically.

[0206] S904. Periodically send the configuration information of the drone group to the model enabling server so that the model enabling server can determine whether the configuration information of the drone group is updated. When the configuration information of the drone group is updated, determine the updated inference model based on the updated configuration information of the drone group, and send the updated inference model to the core network element.

[0207] Drones periodically report their group's configuration information to the model server. If the group's configuration is updated, the model enablement server adjusts the parameters of the inference model based on the updated configuration information, generating an updated inference model. The model enablement server then sends the updated inference model to the core network element. The core network element receives the updated inference model from the model enablement server, divides it into multiple sub-models, and sends them to the corresponding drones. Each drone then uses the received sub-model to perform its own inference task.

[0208] In this embodiment, the drone reports the configuration information of the drone group, and the model enables the server to check whether the configuration information of the drone group is updated. If updated, the inference model can be quickly adjusted through the updated configuration information to ensure that the sub-model sent to the drone is always consistent with the current status of the drone group, thereby improving the inference accuracy.

[0209] Figure 10 A data processing method flow chart of an embodiment of the present disclosure is shown. An embodiment of the present disclosure provides a data processing method that can be executed by any model-enabled server with computing and processing capabilities. Figure 10 As shown, the data processing method provided in the embodiment of the present disclosure includes the following steps.

[0210] S1002: Determine an inference model that matches the drone group.

[0211] In an exemplary embodiment of the present disclosure, a core network element embeds a model identifier into model selection signaling and transmits the model selection signaling to a model enablement server via a network or other communication technology. Upon receiving the model selection signaling, the model enablement server parses the signaling to obtain the model identifier and acquires the inference model corresponding to the model identifier.

[0212] In an exemplary embodiment of the present disclosure, the drones report updated configuration information of the drone group to the model server, and the model enabling server adjusts the model parameters of the inference model according to the updated configuration information of the drone group to obtain an updated inference model.

[0213] In an exemplary embodiment of the present disclosure, drones periodically report configuration information of the drone group to the model server. If the configuration information of the drone group is updated, the model enabling server adjusts the model parameters of the inference model according to the updated configuration information of the drone group to obtain an updated inference model.

[0214] S1004. Transmit the inference model to the core network element so that the core network element divides the inference model into multiple sub-models, and sends the corresponding sub-model to each drone in the drone group so that the drone uses the received sub-model to process data.

[0215] The core network element sends a model selection signaling message to the model enabling server. The model selection signaling message includes the model identifier of the inference model. The model selection signaling message is used to request the inference model corresponding to the model identifier. After receiving the model selection signaling message, the model enabling server finds the corresponding inference model based on the model identifier. The model enabling server then sends the inference model to the core network element via the network or other communication methods. The model enabling server transmits the inference model, including the model architecture, model parameters, model identifier, information related to the drone group, and information related to the ground user terminal corresponding to the drone group.

[0216] After receiving the inference model, the core network element divides the inference model into multiple sub-models and sends the sub-models to the corresponding drones. The drones use the sub-models to infer the perception data and obtain the inference results of the sub-models.

[0217] In this solution, the core network element divides the inference model into multiple sub-models and sends them to the corresponding drones. The drones use the sub-models to infer the perception data, avoiding the drones sending large amounts of perception data to the ground user end, reducing the delay in data inference, and lowering the delay of the drone's perception service.

[0218] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 11 As shown, the optimized data processing method includes the following steps.

[0219] S1102. Receive a drone group registration request message sent by a core network element.

[0220] In an exemplary embodiment of the present disclosure, the drones in a drone group are networked and share information with each other. A drone in the group receives the shared information from other drones and ranks them from highest to lowest according to their remaining energy to determine their respective priorities. The shared information from each drone and the relevant information about the users in the drone group are integrated to obtain the configuration information for the drone group. A drone group registration request message is generated based on the configuration information, the relevant information about the users in the drone group, and the mission requirements.

[0221] The drones send a drone group registration request message to the NWDAF in the core network via the base station. The drone group registration request message includes the drone group's location, each drone's hardware resources, each drone's remaining energy, information about the users belonging to the drone group, and mission requirements. Based on the various information included in the drone group registration request message, the NWDAF searches for the model-enabling server that is closest to the ground user or the drone group. The drone group registration request message is sent to the model-enabling server closest to the ground user or the drone group. The model-enabling server then receives the drone group registration request message.

[0222] S1104. A drone group registration reply message is sent to the core network network element, so that after receiving the drone group registration reply message sent by the model enabling server, the core network network element obtains the model identifier of the inference model matching the drone group based on the task requirement information, the configuration information of the drone group and at least one of the configuration information of the model enabling server, and forwards the drone group registration reply message to the drone group.

[0223] The model enabling server stores the configuration information and user information of the drone group locally, generates the corresponding drone group number information, and then sends a drone group registration reply message to the core network element. The drone group registration reply message contains the drone group number information and registration success information.

[0224] After receiving the drone group registration reply message, the core network element matches the model according to the configuration information of the drone group, the configuration information of the model enabling server, and the mission requirement information, and determines the inference model that matches the drone group.

[0225] In this embodiment, a process is provided for a drone group to register with a model enabling server through a core network, thereby realizing centralized management of the drone group, effectively allocating resources and scheduling tasks, and ensuring smoother coordination and cooperation between different tasks.

[0226] S1106. Receive model selection signaling sent by a core network element, where the model selection signaling includes a model identifier of an inference model that matches the drone group.

[0227] After receiving the drone group registration reply message, the core network element matches the model based on the drone group configuration information, the model enablement server configuration information, and the mission requirements information, and determines the inference model that matches the drone group. It then sends a model selection signaling to the model enablement server, informing the model enablement server of the model identifier of the selected inference model.

[0228] S1108. Determine the inference model corresponding to the model identifier.

[0229] The model enabling service queries the corresponding inference model according to the model identifier and sends the inference model corresponding to the model identifier to the core network element.

[0230] In this embodiment, by sending model selection signaling to the model enabling server, all inference models are uniformly managed and distributed by the server, reducing the need for local storage and maintenance of models by drones and reducing the complexity of the system.

[0231] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 12 As shown, the optimized data processing method includes the following steps.

[0232] S1202: Receive updated configuration information of the drone group.

[0233] During a drone group mission, each drone in the group shares its location and online information. When refreshing via the ground user terminal, drones can discover newly added drones or offline drones in the group. Based on the information about newly added or offline drones, the drone group's updated configuration is retrieved.

[0234] S1204: Determine an updated inference model based on the updated configuration information of the drone group.

[0235] The drones report the updated configuration information of the drone group to the model server. The model enabling server adjusts the model parameters of the inference model based on the updated configuration information of the drone group, generating an updated inference model. The model enabling server sends the updated inference model to the core network element. The core network element receives the updated inference model from the model enabling server, divides the updated inference model into multiple sub-models, and sends them to the corresponding drones. Each drone uses the received sub-model to perform its own inference task.

[0236] In this embodiment, the drone reports the configuration information of the drone group, and the model enabling server can quickly adjust the inference model through the updated configuration information, ensuring that the sub-model sent to the drone is always consistent with the current status of the drone group, thereby improving the inference accuracy.

[0237] Based on the above embodiment, this embodiment optimizes the data processing method. Figure 13 As shown, the optimized data processing method includes the following steps.

[0238] S1302: Periodically receive configuration information of a drone group sent by a drone.

[0239] During the execution of a mission by a drone group, each drone in the drone group shares its location and online information. The drones obtain the configuration information of the drone group in real time or periodically, and periodically send the configuration information of the drone group to the model enablement server.

[0240] S1304. When the configuration information of the drone group is updated, determine an updated inference model based on the updated configuration information of the drone group.

[0241] If the configuration information of the drone group is updated, the model enabling server adjusts the model parameters of the inference model based on the updated configuration information of the drone group, generating an updated inference model. The model enabling server sends the updated inference model to the core network element. The core network element receives the updated inference model from the model enabling server and divides it into multiple sub-models. These sub-models are then sent to the corresponding drones, allowing each drone to perform its own inference task using the received sub-model.

[0242] In this embodiment, the drone reports the configuration information of the drone group, and the model enables the server to check whether the configuration information of the drone group is updated. If updated, the inference model can be quickly adjusted through the updated configuration information to ensure that the sub-model sent to the drone is always consistent with the current status of the drone group, thereby improving the inference accuracy.

[0243] Based on the above embodiment, this embodiment provides a data processing system, such as Figure 14 As shown in Figure 1, the data processing system includes: core network elements, model enabling servers, drones and ground user terminals. Figure 14 As shown, the interactive method for data processing includes the following steps.

[0244] S1402: The model enabling server determines an inference model that matches the drone group.

[0245] S1404. The core network element obtains the inference model sent by the model enabling server.

[0246] S1406. The core network element divides the inference model into multiple sub-models.

[0247] S1408. The core network element sends the corresponding sub-model to each drone in the drone group.

[0248] S1410: The drone processes data using the received sub-model to obtain an inference result of the sub-model.

[0249] S1412. The UAV sends the inference result of the sub-model to the ground user terminal.

[0250] S1414. The ground user terminal receives and aggregates the inference results of each sub-model.

[0251] The interactive method of the data processing system provided in the embodiments of the present disclosure includes: the core network network element obtains the inference model sent by the model enabling server, and then divides the inference model into multiple sub-models; and sends the corresponding sub-model to each drone in the drone group; the drone uses the received sub-model to perform data processing to obtain the inference results of each sub-model, and sends them to the ground user end, which receives and aggregates the inference results of each sub-model. In this solution, the core network network element divides the inference model into multiple sub-models and sends them to the corresponding drones. The drones use the sub-models to infer the perception data, avoiding the drones from sending large amounts of perception data to the ground user end, reducing the delay of data inference, and lowering the delay of the drone perception service.

[0252] Based on the above embodiment, this embodiment provides a data processing system, such as Figure 15 As shown, the data processing system includes: UAV 1510, core network 1520, model enablement server 1530, and ground user terminal 1540. UAV 1510 includes: model enablement client 1511 and perception client 1512, core network 1520 includes NWDAF unit 1521 and UPF unit 1522. The inference model is deployed in model enablement server 1530.

[0253] In one possible implementation, the perception client 1512 establishes a connection with the model enabling server 1530 via the UPF unit 1522. The model enabling client 1511 collects the processing requirements of the user perception data and requests the required reasoning model from the model enabling server 1530 via the NWDAF unit 1521.

[0254] Perception client 1512 perceives the surrounding multimodal environmental information, including images, video, and audio, locally on the user device. This information is then transmitted via the core network's UPF unit 1522 to the inference model deployed by the model enablement server 1530. The data perceived by perception client 1512 is compressed and encrypted locally on the device, and the inference model is provided with compressed data and corresponding labels for fine-tuning training. After matching, model enablement server 1530 distributes the corresponding model to perception client 1512 and ground user terminal 1540 via the data channel of UPF unit 1522. Perception client 1512 transmits the inference results to ground user terminal 1540, which aggregates them to produce the required recognition inference results.

[0255] In a possible implementation scenario, a drone group registration process is provided, such as Figure 16 As shown, the registration process of the drone group includes the following steps.

[0256] S1602. The drone sends a drone group registration request message to the NWDAF unit.

[0257] After startup, the drone group is initialized. Each drone shares information such as location, hardware, and battery level. The drone with the highest remaining battery level communicates with the base station and registers with the model enablement server. This drone sends a drone group registration request message to the NWDAF in the core network through the base station. The request includes information such as the drone group's location, hardware, and battery level, as well as information about the users to whom the drone group belongs.

[0258] S1604: NWDAF searches for the nearest model-enabled server based on the user's location.

[0259] The NWDAF searches for a model-enabled server based on the information in step S1602 , and searches for a deployed model-enabled server that is closest to the user.

[0260] S1606. NWDAF sends a drone group registration request message to the model enabling server.

[0261] Based on the search result in step S1604, NWDAF sends a drone group registration request message to the selected model enabling server, which includes the drone group information and user information in step S1602.

[0262] S1608. The model enabling server sends a drone group registration reply message to the NWDAF.

[0263] After receiving the drone group registration request message in step S1606, the model enabling server stores the drone group information and user information locally, generates the corresponding drone group number information, and then sends a drone group registration reply message to NWDAF, which includes the drone group number information and registration success information.

[0264] S1610. NWDAF determines the model identifier of the inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server.

[0265] S1612: NWDAF sends a model selection signaling to the model enabling server.

[0266] S1614. NWDAF sends a drone group registration reply message to the drone.

[0267] NWDAF broadcasts a drone group registration reply message to each drone in the drone group through the base station, indicating that the drone registration is successful, and informs the registered model enabling server and the selected model number.

[0268] In a possible implementation scenario, a drone group management process is provided, such as Figure 17 As shown, the registration process of the drone group includes the following steps.

[0269] S1702. The drone sends the updated configuration information of the drone group to the model enabling server.

[0270] The drone group shares location and online information. When the user terminal refreshes, the first-numbered drone detects the appearance of a new drone or the offline of an existing drone, and then sends information to the model-enabled server platform to report the updated information of the drone group.

[0271] Based on the shared location and online information, drones periodically report the information of the drone group to the model enabling server through the base station.

[0272] S1704. The model enabling server updates the local storage information according to the updated configuration information of the drone group.

[0273] The model enabling server updates the locally stored information based on the information reported by the drones, including the location and online information of the drone group in step S1702.

[0274] S1706: The model enabling server sends a group information update message to the drone.

[0275] The model enabling server sends a group information update message to the drone, indicating that the information stored locally on the model enabling server has been updated.

[0276] S1708. The model enabling server updates and issues the inference model according to the updated configuration information.

[0277] In a possible implementation scenario, users use drones to scan the environment and collect information in the air. The perceived data includes photography, video, infrared cameras, radio signal collection, etc.; users and drone groups register on the model enablement server through the network, and search and match the required inference models according to their own needs. Multiple drones are performing tasks in a cellular area, such as image recognition, and they recognize each other; after identifying the model sent by the cloud, the highest-priority devices in the drones save the model and parameters with the smallest granularity in the original model; other devices save the remaining models and parameters in order of priority; the drones perform perception, recognition and inference, and then send the recognition results to the ground user end, which are aggregated to obtain detailed and accurate results.

[0278] In this embodiment, through the aerial distributed reasoning architecture, task offloading of different granularities in the drone group is achieved, which effectively reduces the delay of drone communication, perception and reasoning integration, optimizes user experience, and reduces the energy consumption of drone services to a certain extent.

[0279] It should be noted that the acquisition, storage, use, and processing of data in the technical solution disclosed herein are in compliance with the relevant provisions of national laws and regulations. Various types of data such as personal identity data, operation data, behavioral data, etc. related to individuals, customers, and groups obtained in the embodiments of the present disclosure have been authorized.

[0280] Based on the same inventive concept, the present disclosure also provides a data processing device, as described in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0281] Figure 18 A schematic diagram of a data processing device according to an embodiment of the present disclosure is shown, wherein the data processing device is configured in a core network element. Figure 18 As shown, the device includes: a model acquisition module 1810, a model segmentation module 1820 and a sub-model sending module 1830.

[0282] Among them, the model acquisition module 1810 is used to obtain the inference model sent by the model enabling server; the model segmentation module 1820 is used to divide the inference model into multiple sub-models; and the sub-model sending module 1830 is used to send the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to perform data processing.

[0283] In an exemplary embodiment of the present disclosure, the model segmentation module 1820 is specifically configured to divide the inference model into multiple sub-models according to at least one of the configuration information of the drone group, the configuration information of the inference model, and the task requirement information.

[0284] In an exemplary embodiment of the present disclosure, the model segmentation module 1820 is specifically used to divide the inference model into multiple sub-models and determine the correspondence between the sub-model identifier and the drone identifier; broadcast the correspondence to the drone group so that the drone determines its corresponding sub-model identifier; the sub-model sending module 1830 is specifically used to receive a sub-model request message sent by each drone in the drone group, wherein the sub-model request message includes the sub-model identifier; and send the sub-model corresponding to the sub-model identifier to the drone.

[0285] In an exemplary embodiment of the present disclosure, the configuration information of the drone group includes the priority of each drone in the drone group; the model segmentation module 1820 is specifically used to determine the correspondence between the sub-model identifier and the drone identifier based on the granularity of the sub-model and the priority of the drone.

[0286] In an exemplary embodiment of the present disclosure, the priority of the drone is determined by the remaining energy of the drone and / or the distance between the drone and the core network element.

[0287] In an exemplary embodiment of the present disclosure, the granularity of the sub-model is inversely proportional to the priority of the drone.

[0288] In an exemplary embodiment of the present disclosure, the model acquisition module 1810 is specifically used to determine the model identifier of the inference model that matches the drone group; send a model selection signaling to the model enabling server, wherein the model selection signaling includes the model identifier of the inference model; and obtain the inference model corresponding to the model identifier sent by the model enabling server.

[0289] In an exemplary embodiment of the present disclosure, the model acquisition module 1810 is specifically used to obtain an updated reasoning model sent by the model enabling server, wherein the updated reasoning model is determined by the updated configuration information of the drone group received by the model enabling server.

[0290] In an exemplary embodiment of the present disclosure, the model acquisition module 1810 is specifically used to determine the model identifier of the inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server.

[0291] In an exemplary embodiment of the present disclosure, the model acquisition module 1810 is specifically used to receive a drone group registration request message sent by a drone, wherein the drone group registration request message includes: mission requirement information and configuration information of the drone group; send a drone group registration request message to a model enabling server; after receiving a drone group registration reply message sent by the model enabling server, determine the model identifier of the inference model that matches the drone group based on at least one of the mission requirement information, the configuration information of the drone group and the configuration information of the model enabling server; and send a drone group registration reply message to the drone group.

[0292] In an exemplary embodiment of the present disclosure, the configuration information of the drone group includes the user information to which the drone group belongs; the model acquisition module 1810 is specifically used to determine a model-enabled server that meets the requirements based on the user information to which the drone group belongs; and send a drone group registration request message to the model-enabled server that meets the requirements.

[0293] In an exemplary embodiment of the present disclosure, it also includes: an inference model sending module, which is used to send the inference model to the ground user terminal corresponding to the drone group, so that the ground user terminal receives the inference results sent by each drone and aggregates them.

[0294] Figure 19 A schematic diagram of a data processing device according to an embodiment of the present disclosure is shown. The data processing device is configured in a drone, such as Figure 19 As shown, the device includes: a sub-model receiving module 1910, a perception data acquisition module 1920, a data inference module 1930 and an inference result sending module 1940.

[0295] Among them, the sub-model receiving module 1910 is used to receive the sub-model sent by the core network network element. The sub-model is obtained by the core network network element by splitting the inference model. The inference model is a model that matches the drone group and is obtained by the core network network element from the model enabling server; the perception data acquisition module 1920 is used to obtain perception data; the data inference module 1930 is used to perform inference based on the sub-model and perception data to obtain the inference result of the sub-model; the inference result sending module 1940 is used to send the inference result of the sub-model to the ground user terminal, so that the ground user terminal can aggregate the inference results of each sub-model.

[0296] In some exemplary embodiments of the present disclosure, the sub-model receiving module 1910 is specifically used to receive the correspondence between the sub-model identifier broadcast by the core network network element and the drone identifier; determine the sub-model identifier corresponding to the drone based on the correspondence; send a sub-model request message to the core network network element, wherein the sub-model request message includes the sub-model identifier; and receive the sub-model corresponding to the sub-model identifier sent by the core network network element.

[0297] In some exemplary embodiments of the present disclosure, a registration module is used to obtain shared information of each drone in a drone group; determine the configuration information of the drone group based on the shared information of each drone; send a drone group registration request message to a core network network element, wherein the drone group registration request message is used to instruct the core network network element to forward the drone group registration request message to a model enabling server, so that after the core network network element receives the drone group registration reply message sent by the model enabling server, it obtains the model identifier of the inference model matching the drone group according to at least one of the task requirement information, the configuration information of the drone group and the configuration information of the model enabling server; and receives the drone group registration reply message sent by the core network network element.

[0298] In some exemplary embodiments of the present disclosure, it also includes: a configuration information update module for receiving updated configuration information of the drone group; sending the updated configuration information to the model enabling server, so that the model enabling server determines the updated reasoning model according to the updated configuration information of the drone group, and sends the updated reasoning model to the core network.

[0299] In some exemplary embodiments of the present disclosure, the configuration information update module is also used to periodically send the configuration information of the drone group to the model enabling server, so that the model enabling server can determine whether the configuration information of the drone group is updated. When the configuration information of the drone group is updated, the updated inference model is determined according to the updated configuration information of the drone group, and the updated inference model is sent to the core network.

[0300] Figure 20 A schematic diagram of a data processing device according to an embodiment of the present disclosure is shown, wherein the data processing device is configured in a model enabling server. Figure 20 As shown, the device includes: an inference model determination module 2010 and an inference model sending module 2020.

[0301] Among them, the inference model determination module 2010 is used to determine the inference model that matches the drone group; the inference model sending module 2020 is used to transmit the inference model to the core network network element, so that the core network network element divides the inference model into multiple sub-models, and sends the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to process data.

[0302] In some exemplary embodiments of the present disclosure, the inference model determination module 2010 is specifically used to receive model selection signaling sent by a core network element, wherein the model selection signaling includes a model identifier of an inference model matching the drone group; and determine the inference model corresponding to the model identifier.

[0303] In some exemplary embodiments of the present disclosure, the reasoning model determination module 2010 is specifically configured to receive updated configuration information of the drone group; and determine an updated reasoning model according to the updated configuration information of the drone group.

[0304] In some exemplary embodiments of the present disclosure, the inference model determination module 2010 is specifically used to periodically receive configuration information of a drone group sent by drones; when the configuration information of the drone group is updated, determine an updated inference model based on the updated configuration information of the drone group.

[0305] In some exemplary embodiments of the present disclosure, it also includes: a registration module, which is used to receive a drone group registration request message sent by a core network network element; a drone group registration reply message sent to the core network network element, so that after the core network network element receives the drone group registration reply message sent by the model enabling server, it obtains the model identifier of the inference model matching the drone group based on the task requirement information, the configuration information of the drone group and at least one of the configuration information of the model enabling server, and forwards the drone group registration reply message to the drone group.

[0306] It should be noted that the examples and application scenarios implemented by the modules in the above-mentioned apparatus embodiment are the same as those implemented by the corresponding steps in the method embodiment, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned modules, as part of the apparatus, can be executed in a computer system, such as a set of computer-executable instructions.

[0307] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation that combines hardware and software aspects, which may be collectively referred to herein as a "circuit," "module," or "system."

[0308] According to the same inventive concept, an embodiment of the present disclosure further provides an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-described data processing methods by executing the executable instructions. Since the principles for solving the problem in this electronic device embodiment are similar to those in the above-described method embodiment, the implementation of this electronic device embodiment can refer to the implementation of the above-described method embodiment, and repeated parts will not be repeated here.

[0309] Refer to the following Figure 21 2100 according to this embodiment of the present disclosure will be described. Figure 21 The electronic device 2100 shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present disclosure. The electronic device 2100 may include any one of a core network element, a drone, or a model enabling server.

[0310] like Figure 21 As shown, electronic device 2100 is implemented as a general-purpose computing device. Components of electronic device 2100 may include, but are not limited to, the aforementioned processing unit 2110, the aforementioned storage unit 2120, and a bus 2130 connecting various system components (including the storage unit 2120 and the processing unit 2110).

[0311] The storage unit stores program code, which can be executed by the processing unit 2110, so that the processing unit 2110 performs the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 2110 can perform any of the data processing methods in the above embodiments.

[0312] The storage unit 2120 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 21201 and / or a cache memory unit 21202 , and may further include a read-only memory unit (ROM) 21203 .

[0313] The storage unit 2120 may also include a program / utility 21204 having a set (at least one) of program modules 21205, such program modules 21205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0314] The bus 2130 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0315] The electronic device 2100 can also communicate with one or more external devices 2140 (e.g., a keyboard, pointing device, Bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 2100, and / or any device that enables the electronic device 2100 to communicate with one or more other computing devices (e.g., a router, modem, etc.). This communication can occur via an input / output (I / O) interface 2150. Furthermore, the electronic device 2100 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 2160. As shown, the network adapter 2160 communicates with other modules of the electronic device 2100 via a bus 2130. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 2100, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0316] Through the description of the above embodiments, it will be readily understood by those skilled in the art that the example embodiments described herein can be implemented via software or via a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or mobile hard drive) or on a network and includes several instructions for enabling a computing device (such as a personal computer, server, terminal device, or network device) to execute the methods according to the embodiments of the present disclosure.

[0317] Based on the same inventive concept, embodiments of the present disclosure further provide a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements any of the aforementioned data processing methods. Because the principles underlying the problems solved by this computer-readable storage medium embodiment are similar to those of the aforementioned method embodiment, the implementation of this computer-readable storage medium embodiment can be referenced to the implementation of the aforementioned method embodiment, and any repetitions will not be repeated.

[0318] More specific examples of computer-readable storage media in the present disclosure may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0319] In the present disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0320] Alternatively, the program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0321] In a specific implementation, the program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0322] Based on the same inventive concept, embodiments of the present disclosure further provide a computer program product, including a computer program or instructions, which, when executed by a processor, implements the data processing method of any one of the above-described method embodiments. Because the principles for solving the problems of this computer program product embodiment are similar to those of the above-described method embodiments, the implementation of this computer program product embodiment can refer to the implementation of the above-described method embodiments, and any repetitive details will not be repeated.

[0323] It should be noted that although several modules or units of the device for action execution are mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0324] Furthermore, although the steps of the method of the present disclosure are described in a particular order in the accompanying drawings, this does not require or imply that the steps must be performed in this particular order, or that all steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0325] Through the description of the above embodiments, it will be readily understood by those skilled in the art that the example embodiments described herein can be implemented via software or via a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or mobile hard drive) or on a network and includes several instructions for enabling a computing device (such as a personal computer, server, mobile terminal, or network device) to execute the methods according to the embodiments of the present disclosure.

[0326] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the appended claims.

Claims

1. A data processing method, characterized in that: The method is applied to a core network element, including: Obtain the inference model sent by the model enablement server; dividing the inference model into a plurality of sub-models; and The corresponding sub-model is sent to each drone in the drone group, so that the drone uses the received sub-model to perform perception data reasoning.

2. The data processing method according to claim 1, wherein: The dividing the inference model into a plurality of sub-models comprises: The reasoning model is divided into a plurality of sub-models according to at least one of the configuration information of the drone group, the configuration information of the reasoning model, and the task requirement information.

3. The data processing method according to claim 2, characterized in that: The dividing the inference model into a plurality of sub-models comprises: Dividing the inference model into a plurality of sub-models, and determining a correspondence between sub-model identifiers and drone identifiers; Broadcasting the corresponding relationship to the drone group so that the drone can determine its corresponding sub-model identifier; Sending the corresponding sub-model to each drone in the drone group includes: For each drone in the drone group, receiving a sub-model request message sent by the drone, wherein the sub-model request message includes a sub-model identifier; The sub-model corresponding to the sub-model identifier is sent to the UAV.

4. The data processing method according to claim 3, wherein: The configuration information of the drone group includes the priority of each drone in the drone group; Determining the correspondence between the sub-model identifier and the drone identifier includes: The corresponding relationship between the sub-model identifier and the drone identifier is determined according to the granularity of the sub-model and the priority of the drone.

5. The data processing method according to claim 4, characterized in that: The priority of the drone is determined by the remaining energy of the drone and / or the distance between the drone and the core network element.

6. The data processing method according to claim 4, characterized in that: The granularity of the sub-model is inversely proportional to the priority of the drone.

7. The data processing method according to claim 1, wherein: The acquisition of the inference model sent by the model enabling server includes: Determining a model identifier of an inference model that matches the drone swarm; Sending a model selection signaling to the model enabling server, wherein the model selection signaling includes a model identifier of the inference model; Obtain the inference model corresponding to the model identifier sent by the model enabling server.

8. The data processing method according to claim 1, wherein: The acquisition of the inference model sent by the model enabling server includes: Obtain an updated reasoning model sent by the model enabling server, wherein the updated reasoning model is determined by updated configuration information of the drone group received by the model enabling server.

9. The data processing method according to claim 7, characterized in that: The model identification of the inference model that matches the drone group includes: Determine a model identifier of an inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server.

10. The data processing method according to claim 9, characterized in that: Determining a model identifier of a reasoning model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server includes: Receiving a drone group registration request message sent by the drone, wherein the drone group registration request message includes: the mission requirement information and the configuration information of the drone group; Sending the drone group registration request message to the model enabling server; After receiving the drone group registration reply message sent by the model enabling server, determine the model identifier of the inference model that matches the drone group based on at least one of the task requirement information, the configuration information of the drone group, and the configuration information of the model enabling server; Send the drone group registration reply message to the drone group.

11. The data processing method according to claim 10, characterized in that: The configuration information of the drone group includes user information of the drone group; Sending a drone group registration request message to the model enabling server includes: Determine a model enabling server that meets the requirements based on the user information of the drone group; Send the drone group registration request message to the model enabling server that meets the requirements.

12. The data processing method according to claim 1, wherein: Also includes: The inference model is sent to the ground user terminal corresponding to the drone group, so that the ground user terminal receives and aggregates the inference results sent by each drone.

13. A data processing method, characterized in that: The method is applied to a drone, comprising: Receiving a sub-model sent by a core network element, wherein the sub-model is obtained by the core network element by splitting the inference model, and the inference model is a model matching the drone group obtained by the core network element from the model enabling server; Acquire sensory data; Performing reasoning based on the sub-model and the perception data to obtain an inference result of the sub-model; The inference results of the sub-models are sent to a ground user terminal, so that the ground user terminal aggregates the inference results of the sub-models.

14. The data processing method according to claim 13, wherein: The sub-model received from the core network element includes: Receive the correspondence between the sub-model identifier and the drone identifier broadcast by the core network element; Determine the sub-model identifier corresponding to the drone according to the corresponding relationship; Sending a sub-model request message to the core network element, wherein the sub-model request message includes the sub-model identifier; Receive the sub-model corresponding to the sub-model identifier sent by the core network network element.

15. The data processing method according to claim 13, wherein: Also includes: Obtaining shared information of each drone in the drone group; Determining configuration information of a drone group based on the shared information of each drone; Sending a drone group registration request message to the core network network element, wherein the drone group registration request message is used to instruct the core network network element to forward the drone group registration request message to the model enabling server, so that after receiving the drone group registration reply message sent by the model enabling server, the core network network element obtains an inference model matching the drone group according to at least one of task requirement information, configuration information of the drone group, and configuration information of the model enabling server; Receive the drone group registration reply message sent by the core network element.

16. The data processing method according to claim 13, characterized in that: Also includes: receiving updated configuration information of the drone group; Send the updated configuration information to the model enabling server, so that the model enabling server determines an updated reasoning model according to the updated configuration information of the drone group, and sends the updated reasoning model to the core network network element.

17. The data processing method according to claim 13, characterized in that: Also includes: The configuration information of the drone group is periodically sent to the model enabling server so that the model enabling server determines whether the configuration information of the drone group is updated. When the configuration information of the drone group is updated, the updated inference model is determined according to the updated configuration information of the drone group, and the updated inference model is sent to the core network element.

18. A data processing method, characterized in that: The method is applied to a model enabling server, comprising: Determine the inference model that matches the drone swarm; The inference model is sent to a core network element so that the core network element divides the inference model into multiple sub-models, and the corresponding sub-model is sent to each drone in the drone group so that the drone uses the received sub-model to perform perception data inference.

19. The data processing method according to claim 18, characterized in that: The inference model for determining the matching with the drone group includes: Receiving a model selection signaling sent by the core network element, wherein the model selection signaling includes a model identifier of an inference model that matches the drone group; Determine the reasoning model corresponding to the model identifier.

20. The data processing method according to claim 18, wherein: The inference model for determining the matching with the drone group includes: receiving updated configuration information of the drone group; An updated inference model is determined based on the updated configuration information of the drone group.

21. The data processing method according to claim 18, wherein: The inference model for determining the matching with the drone group includes: Periodically receiving configuration information of the drone group sent by the drone; When the configuration information of the drone group is updated, the updated inference model is determined according to the updated configuration information of the drone group.

22. The data processing method according to claim 19, characterized in that: Also includes: Receiving the drone group registration request message sent by the core network element; The drone group registration reply message is sent to the core network network element, so that after receiving the drone group registration reply message sent by the model enabling server, the core network network element determines the model identifier of the inference model matching the drone group according to the task requirement information, the configuration information of the drone group and at least one of the configuration information of the model enabling server, and forwards the drone group registration reply message to the drone group.

23. A data processing system, characterized in that: The system includes core network elements, a model enabling server, a UAV and a ground user terminal; The model enabling server is configured to determine an inference model that matches the drone group; The core network element is configured to obtain the inference model sent by the model enabling server; divide the inference model into a plurality of sub-models; and send the corresponding sub-model to each drone in the drone group; The drone is used to process data using the received sub-models to obtain inference results of each sub-model; The ground user terminal is used to receive and aggregate the inference results of each sub-model.

24. A data processing device, characterized in that: The device is configured in a core network element and includes: A model acquisition module is used to obtain the inference model sent by the model enabling server; A model segmentation module, configured to divide the inference model into a plurality of sub-models; and The sub-model sending module is used to send the corresponding sub-model to each drone in the drone group, so that the drone uses the received sub-model to perform perception data inference.

25. A data processing device, characterized in that: The device is configured on a drone and includes: A sub-model receiving module is configured to receive a sub-model sent by a core network element, where the sub-model is obtained by the core network element by segmenting the inference model, and the inference model is a model matching the drone group obtained by the core network element from the model enabling server; A perception data acquisition module, used to acquire perception data; A data inference module, configured to perform inference based on the sub-model and the perception data to obtain an inference result of the sub-model; The inference result sending module is used to send the inference results of the sub-models to the ground user terminal, so that the ground user terminal aggregates the inference results of each sub-model.

26. A data processing device, characterized in that: The device is configured on a model enabling server and includes: An inference model determination module, used to determine the inference model that matches the drone group; The inference model sending module is used to transmit the inference model to the core network network element so that the core network network element divides the inference model into multiple sub-models, and sends the corresponding sub-model to each drone in the drone group so that the drone uses the received sub-model to perform perception data inference.

27. An electronic device, characterized in that: include: processor; as well as a memory for storing executable instructions of the processor; The processor is configured to perform the data processing method according to any one of claims 1 to 22 by executing the executable instructions.

28. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the data processing method according to any one of claims 1 to 22 is implemented.

29. A computer program product comprising: A computer program or instruction, characterized in that when the computer program or instruction is executed by a processor, it implements the data processing method described in any one of claims 1 to 22.

Images