CN114662999B

CN114662999B - Unmanned aerial vehicle cluster transfer processing method and system and cloud platform

Info

Publication number: CN114662999B
Application number: CN202210566327.XA
Authority: CN
Inventors: 杨翰翔; 杨德润
Original assignee: Shenzhen Lianhe Intelligent Technology Co ltd
Current assignee: Shenzhen Lianhe Intelligent Technology Co ltd
Priority date: 2022-05-24
Filing date: 2022-05-24
Publication date: 2022-08-30
Anticipated expiration: 2042-05-24
Also published as: CN114662999A

Abstract

The application discloses an unmanned aerial vehicle cluster transferring processing method, a system and a cloud platform, and the method comprises the steps of obtaining target unmanned aerial vehicle cluster flight path information; determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selection sub-network and a plurality of task policy response sub-networks, wherein the task policy selection sub-networks are not associated with each other, the task policy selection sub-networks are used for selecting task policies to be used from the plurality of task policy response sub-networks according to the cluster flight path information of the unmanned aerial vehicle, and the task policy response sub-networks are used for determining emergency flight tasks to be responded by the unmanned aerial vehicle to be allocated according to the cluster flight path information of the unmanned aerial vehicle; and moving the target to be allocated to the unmanned aerial vehicle to respond to the target emergency flight task. The method can quickly and accurately select and execute the unmanned aerial vehicle cluster task policy, and realizes accurate and efficient movement of the unmanned aerial vehicle to be allocated.

Description

Unmanned aerial vehicle cluster transfer processing method and system and cloud platform

Technical Field

The application relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle cluster transferring processing method, system and cloud platform.

Background

Under the condition of the continuous extension of unmanned aerial vehicle's application, need carry out the linkage allotment (unmanned aerial vehicle cluster allotment) to a plurality of unmanned aerial vehicle usually, and how to realize that accurate, the high-efficient transfer of unmanned aerial vehicle cluster is a technical problem that needs the improvement at present.

Disclosure of Invention

In order to solve the technical problems in the related art, the application provides an unmanned aerial vehicle cluster transferring processing method, an unmanned aerial vehicle cluster transferring processing system and a cloud platform.

The application provides an unmanned aerial vehicle cluster transferring processing method, which comprises the following steps:

acquiring flight path information of a target unmanned aerial vehicle cluster;

determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selecting sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information;

and moving the target to be allocated, and responding to the target emergency flight mission by the unmanned aerial vehicle.

In an embodiment that can be implemented independently, the determining, by the periodic task scheduling network, a target emergency flight task according to the target drone cluster flight path information includes:

detecting whether a target task policy response sub-network in use exists at present; if yes, responding to a sub-network through the target task policy, and determining the target emergency flight task according to the target unmanned aerial vehicle cluster flight path information; if not, selecting a sub-network through the task policy, and selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the target unmanned aerial vehicle cluster flight path information to serve as a target task policy response sub-network;

responding to a sub-network through the target task policy, and determining the target emergency flight task according to the target unmanned aerial vehicle cluster flight path information;

correspondingly, after the sub-network is responded by the target task policy and the target emergency flight task is determined according to the target unmanned aerial vehicle cluster flight path information, the method further includes:

determining whether to suspend using the target task policy response subnetwork according to the target unmanned aerial vehicle cluster attribute; the target drone cluster attributes include at least one of:

the target emergency flight mission, the flight path of the target unmanned aerial vehicle to be allocated, the flight path of the unmanned aerial vehicle related to the target unmanned aerial vehicle to be allocated, and the accumulated use time of the target mission policy response sub-network;

correspondingly, the task policy response sub-networks comprise an aerial survey task response sub-network and a rescue task response sub-network; the determining whether to suspend responding to subnetworks with the target task policy based on target drone cluster attributes includes:

when the target task policy response subnetwork is the aerial survey task response subnetwork, detecting whether the target emergency flight task corresponds to a termination label of the aerial survey task response subnetwork, and if so, suspending use of the aerial survey task response subnetwork; when the target task policy response sub-network is the rescue task response sub-network, detecting whether the target unmanned aerial vehicle to be deployed is interfered by the emergency flight task of the associated unmanned aerial vehicle, and if so, suspending the use of the rescue task response sub-network; or when the target task policy response sub-network is the rescue task response sub-network, detecting whether the accumulated use time of the rescue task response sub-network exceeds a set use duration value, and if so, suspending the use of the rescue task response sub-network.

In a separately implementable embodiment, the task guideline response sub-network is trained by:

acquiring flight path information of a training cluster in the flight training process of the unmanned aerial vehicle;

acquiring a task policy response sub-network to be trained to train an emergency flight task determined by the training cluster flight path information;

acquiring an unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process;

generating first training sample information based on the training cluster flight path information, the training emergency flight task and the unmanned aerial vehicle task allocation training result;

and using the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the task policy to respond to the sub-network to be mobilized, as a training requirement condition for successful task allocation, and training the task policy responding sub-network by using the first training sample information.

In a separately implementable embodiment, the task guideline response sub-network is generated by:

and generating a task node network as the task policy response sub-network based on the corresponding relation between the emergency flight task response condition under the unmanned aerial vehicle cluster task policy corresponding to the task policy response sub-network and the response emergency flight task.

In an independently implementable embodiment, the plurality of task guideline response sub-networks comprise an aerial survey task response sub-network and a rescue task response sub-network; training the aerial survey task response subnetwork and the rescue task response subnetwork by:

in the flight training process of the unmanned aerial vehicle, performing emergency task interaction processing by using the aerial survey task response sub-network and the rescue task response sub-network to be trained;

acquiring a training aerial survey emergency flight task determined by the aerial survey task response sub-network according to the training cluster flight path information;

acquiring a training rescue emergency flight task determined by the rescue task response sub-network according to the flight path information of the training cluster;

generating second training sample information based on the training cluster flight path information, the training aerial survey emergency flight task and the unmanned aerial vehicle task allocation training result;

generating third training sample information based on the training cluster flight path information, the training rescue emergency flight task and the unmanned aerial vehicle task allocation training result;

for the aerial survey task response sub-network, successfully distributing the unmanned aerial vehicle task distribution training result of the unmanned aerial vehicle to be deployed, which is mobilized by the aerial survey task response sub-network, as a training requirement condition, and training the aerial survey task response sub-network by using the second training sample information;

for the rescue task response sub-network, successfully distributing the unmanned aerial vehicle task distribution training result of the unmanned aerial vehicle to be deployed, which is mobilized by the rescue task response sub-network, as a training requirement condition, and training the rescue task response sub-network by using the third training sample information;

correspondingly, in the flight training process of the unmanned aerial vehicle, the emergency task interaction processing is performed by using the aerial survey task response sub-network and the rescue task response sub-network to be trained, and the emergency task interaction processing comprises the following steps:

in the flight training process of the unmanned aerial vehicle, performing emergency task interaction processing by using the aerial survey task response sub-network in the J-th round and the rescue task response sub-network in the K-th round; j and K are integers greater than or equal to 0; the J is equal to the K, or the difference between the J and the K is 1;

when J is smaller than or equal to K, responding to the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the aerial survey task response subnetwork to be mobilized, as a training requirement condition for successful task allocation, and training the aerial survey task response subnetwork by using the second training sample information;

when J is larger than K, responding to the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the rescue task response sub-network to be mobilized, as a training requirement condition, and training the rescue task response sub-network by using the third training sample information;

correspondingly, the task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a rescue task response sub-network; different aerial survey task response sub-networks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; the rescue task response sub-network is used for determining an emergency flight starting task according to the unmanned aerial vehicle cluster flight path information under different aerial survey standards; or the plurality of task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a plurality of rescue task response sub-networks; different aerial survey task response subnetworks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; different rescue task response sub-networks correspond to different rescue standards, and the different rescue standards are used for starting measurement strategies or measurement strategy combinations under different aerial survey standards.

In an independently implementable embodiment, after training the plurality of task guideline response sub-networks, training the task guideline selector sub-network by:

acquiring a training task policy response sub-network selected by the task policy selection sub-network to be trained from the plurality of task policy response sub-networks according to the flight path information of the training cluster;

generating fourth training sample information based on the training cluster flight path information, the training task policy response sub-network and the unmanned aerial vehicle task allocation training result;

and training the task policy selection sub-network by utilizing the fourth training sample information, wherein the result of the task allocation training of the unmanned aerial vehicle to be allocated, which enables the task policy selection sub-network to be mobilized, is the successful task allocation as a training requirement condition.

in the flight training process of the unmanned aerial vehicle, selecting a sub-network by using the task policy to be trained to perform emergency task interaction processing with the sub-network;

taking one of the task policy selection sub-networks as a target task policy selection sub-network, and acquiring a training task policy response sub-network selected by the target task policy selection sub-network in the plurality of task policy response sub-networks according to the training cluster flight path information;

generating fifth training sample information based on the training cluster flight path information, the training task policy response sub-network and the unmanned aerial vehicle task allocation training result;

and training the target task policy selection sub-network by utilizing the fifth training sample information, wherein the result of the task allocation training of the unmanned aerial vehicle to be allocated, which enables the target task policy selection sub-network to be mobilized, is the successful task allocation as a training requirement condition.

In an independently implementable embodiment, maneuvering a target drone to be commissioned in response to the target emergency flight mission comprises:

acquiring a visual record of local flight data and a visual record of global flight data in the target unmanned aerial vehicle to be allocated; matching the local flight data visual record and the global flight data visual record in the target unmanned aerial vehicle to be allocated based on the visual record correlation between the local flight data visual record and the global flight data visual record in the target unmanned aerial vehicle to be allocated to obtain a visual record matching result;

determining the abnormal global flight data visual record as a to-be-matched global flight data visual record, and determining the unmanned aerial vehicle flight attitude characteristic matched with the to-be-matched global flight data visual record according to the visual record quantitative commonality value between the global flight data visual record in the visual record matching result and the to-be-matched global flight data visual record;

matching the flight attitude characteristics of the unmanned aerial vehicle matched with the to-be-matched global flight data visual record to obtain an attitude matching result; determining a clustered unmanned aerial vehicle in the target unmanned aerial vehicle to be allocated and unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle according to the attitude pairing result and the visual record pairing result;

and determining a target task path of the target unmanned aerial vehicle to be allocated according to a clustered unmanned aerial vehicle in the target unmanned aerial vehicle to be allocated and the unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle.

The application also provides an unmanned aerial vehicle cluster mobilizing processing system, the system includes intercommunication's unmanned aerial vehicle cluster mobilizing cloud platform and a plurality of unmanned aerial vehicle:

the unmanned aerial vehicle cluster transferring cloud platform is used for acquiring flight path information of a target unmanned aerial vehicle cluster; determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selection sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information; and moving the target to be allocated to the unmanned aerial vehicle to respond to the target emergency flight task.

The application also provides an unmanned aerial vehicle cluster mobilizing cloud platform which comprises a processor and a memory; the processor is connected with the memory in communication, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method.

The technical scheme provided by the embodiment of the application can have the following beneficial effects.

In the method, target unmanned aerial vehicle cluster flight path information is obtained firstly; secondly, determining a target emergency flight task according to the cluster flight path information of the target unmanned aerial vehicle through a staged task scheduling network, wherein the staged task scheduling network comprises a task policy selection sub-network and a plurality of task policy response sub-networks which are not related to each other, the task policy selection sub-network is used for selecting a task policy response sub-network which needs to be used from the plurality of task policy response sub-networks according to the cluster flight path information of the unmanned aerial vehicle, and the task policy response sub-network is used for determining the emergency flight task which needs to be responded by the unmanned aerial vehicle according to the cluster flight path information of the unmanned aerial vehicle; and finally, moving the target to be allocated with the unmanned aerial vehicle to respond to the target emergency flight task.

The method simplifies the problem of relatively complex emergency flight task scheduling in the unmanned aerial vehicle cluster to a certain extent, simulates the emergency flight task scheduling process of a real unmanned aerial vehicle in the flight process of the unmanned aerial vehicle cluster, performs independent processing on task policy selection and task policy response to a certain extent, and respectively realizes task policy selection and task policy response through a task policy selection sub-network and a task policy response sub-network which are not associated with each other.

On the one hand, the sub-network is selected through the task policy to select the task policy required to be used from the plurality of task policy response sub-networks to respond to the sub-networks, the emergency flight task to be allocated for the emergency flight task of the unmanned aerial vehicle is determined through the selected task policy response sub-network, the stage scheduling of the emergency flight task of the unmanned aerial vehicle is realized, the emergency flight task scheduling mode can effectively reduce the scheduling duration of the emergency flight task, the scheduling difficulty of the emergency flight task is reduced, and the method is helpful for rapidly coping with the flight conditions of the complex and changeable unmanned aerial vehicles in the real-time unmanned aerial vehicle cluster.

On the other hand, the task policy selection and the task policy response are adaptively disassembled, and a task policy selection sub-network and a task policy response sub-network which are not associated with each other are established, so that the task policy selection sub-network and the task policy response sub-network are respectively trained by adopting proper training requirement conditions and network training strategies, and thus, the training complexity and the resource overhead of the periodic task scheduling network can be reduced, the periodic task scheduling network obtained by training can be guaranteed to have network performance with stronger flight scene adaptability of the unmanned aerial vehicle as much as possible, and accurate and efficient movement of the unmanned aerial vehicle to be deployed can be realized.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.

Fig. 1 is a schematic diagram of a hardware structure of a cloud platform mobilized by a cluster of unmanned aerial vehicles according to an embodiment of the present application.

Fig. 2 is a schematic flowchart of a method for handling cluster mobilization of unmanned aerial vehicles according to an embodiment of the present application.

Fig. 3 is a schematic view of a communication architecture of a drone cluster maneuver processing system according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

The method provided by the embodiment of the application can be executed in an unmanned aerial vehicle cluster-mobilized cloud platform, computer equipment or similar operation devices. Taking an example of an unmanned aerial vehicle cluster maneuvering cloud platform operating on the unmanned aerial vehicle cluster maneuvering cloud platform, fig. 1 is a hardware structure block diagram of the unmanned aerial vehicle cluster maneuvering cloud platform implementing the unmanned aerial vehicle cluster maneuvering processing method in the embodiment of the application. As shown in fig. 1, the drone cluster mobilization cloud platform 10 may include one or more (only one shown in fig. 1) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data, and optionally, may further include a transmission device 106 for communication functions. Those skilled in the art will understand that the structure shown in fig. 1 is only an illustration, and does not limit the structure of the unmanned aerial vehicle cluster mobilizing cloud platform. For example, the drone cluster mobilizing cloud platform 10 may also include more or fewer components than shown in fig. 1, or have a different configuration than shown in fig. 1.

The memory 104 may be configured to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the method for drone cluster commissioning processing in this embodiment of the present application, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located from the processor 102, which may be connected to the drone cluster mobilizer cloud platform 10 through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the drone cluster deployment cloud platform 10. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.

Based on this, please refer to fig. 2, fig. 2 is a schematic flow chart of a method for handling cluster mobilization of unmanned aerial vehicles according to an embodiment of the present invention, and the method may specifically include the technical solutions described in steps 210 to 230.

And step 210, acquiring the flight path information of the target unmanned aerial vehicle cluster.

In particular, the target drone cluster flight path information may include a single flight path and a coordinated flight path of different drones in the drone cluster, in other words, the target drone cluster flight path information may include multiple flight paths, and the flight paths may correspond to different flight tasks, such as an aerial survey task, a rescue task, a delivery task, and the like.

For example, in an actual application process, the unmanned aerial vehicle cluster mobilizing cloud platform may communicate with flight control of the unmanned aerial vehicle, so as to obtain a flight path of the unmanned aerial vehicle based on the flight control, so as to obtain the above-mentioned target unmanned aerial vehicle cluster flight path information.

And step 220, determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network.

In specific implementation, the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selecting sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; and the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information.

For example, the periodic task scheduling network may be an artificial intelligence-based Neural network model, such as a Convolutional Neural network model (CNN), a feed-forward Neural network model (FNN), a Deep Learning Neural network model (DLNN), a Long-Short Term Memory Neural network model (LSTM), and the like.

Further, a task policy picking sub-network and a plurality of task policy responding sub-networks that are not associated with each other may be understood as a task policy picking sub-network and a plurality of task policy responding sub-networks that are independent of each other.

In the embodiments of the present application, there are many types of emergency flight missions, such as rescue type, aerial survey type, logistics distribution type, and other types, and the embodiments of the present application are not limited thereto.

Furthermore, the phased network described above may be understood as a hierarchical network.

In a possible implementation manner, the step 220 of determining the target emergency flight mission according to the target drone cluster flight path information through the periodic mission scheduling network may specifically include the contents described in the steps 2201 and 2202.

Step 2201, detecting whether a target task policy response sub-network in use exists at present; if yes, responding to a sub-network through the target task policy, and determining the target emergency flight task according to the target unmanned aerial vehicle cluster flight path information; and if not, selecting a sub-network through the task policy, and selecting a task policy response sub-network which needs to be used from the plurality of task policy response sub-networks according to the target unmanned aerial vehicle cluster flight path information to serve as a target task policy response sub-network.

Step 2202, responding to a sub-network through the target task policy, and determining the target emergency flight task according to the target unmanned aerial vehicle cluster flight path information.

Executing the content described in the step 2201 and the step 2202, selecting a sub-network by using the task policy, selecting the unmanned aerial vehicle cluster task policy applicable to the current unmanned aerial vehicle cluster flight task according to the acquired target unmanned aerial vehicle cluster flight path information, understandably selecting a target task policy response sub-network from each task policy response sub-network, further responding to the sub-network by using the target task policy, and determining a target emergency flight task which the target to be allocated unmanned aerial vehicle should execute under the unmanned aerial vehicle cluster task policy corresponding to the target task policy response sub-network according to the target unmanned aerial vehicle cluster flight path information, thereby ensuring the accuracy of the determined target emergency flight task and the authenticity of the emergency degree of the flight task.

Further, after determining the target emergency flight mission according to the target drone cluster flight path information through the target mission policy response subnetwork as described in step 2202, the method further includes the content described in step 310.

Step 310, determining whether to suspend responding to the sub-network with the target task policy according to the target unmanned aerial vehicle cluster attribute. In this embodiment of the present application, the target drone cluster attribute may include at least one of: the target emergency flight mission, the flight path of the target unmanned aerial vehicle to be allocated, the flight path of the unmanned aerial vehicle related to the target unmanned aerial vehicle to be allocated, and the accumulated use time (accumulated running time length) of the target mission policy response sub-network.

In one possible implementation, the plurality of task guideline response sub-networks include an aerial survey task response sub-network and a rescue task response sub-network. Based on this, the determining whether to suspend using the target task policy response subnetwork according to the target drone cluster attribute described in step 310 may specifically include the following: when the target task policy response subnetwork is the aerial survey task response subnetwork, detecting whether the target emergency flight task corresponds to a termination label of the aerial survey task response subnetwork, and if so, suspending use of the aerial survey task response subnetwork; when the target task policy response sub-network is the rescue task response sub-network, detecting whether the target unmanned aerial vehicle to be deployed is interfered by the emergency flight task of the associated unmanned aerial vehicle, and if so, suspending the use of the rescue task response sub-network; or when the target task policy response sub-network is the rescue task response sub-network, detecting whether the accumulated use time of the rescue task response sub-network exceeds a set use duration value, and if so, suspending use of the rescue task response sub-network.

In this way, the target task policy response sub-networks are respectively detected under different conditions, different judgments are made according to different detected results, and the use of the rescue task response sub-networks is suspended until the accumulated use time of the rescue task response sub-networks exceeds the set use duration value, so that different judgments can be made according to the information of different branches, the accuracy of the judgment result can be ensured, and the condition that the final network use judgment result is inaccurate due to the error of the information of a certain branch can be avoided.

Step 230, the target to be allocated unmanned aerial vehicle is mobilized to respond to the target emergency flight mission.

In a specific implementation, the target drone to be deployed may be understood as a drone ready to perform a target emergency flight task (in this embodiment, the target drone to be deployed may be multiple drones). Further, the target emergency flight mission may be medical rescue, emergency rescue, disaster relief, and the like. For example, the number of the unmanned aerial vehicles to be allocated by the targets can be multiple, and the unmanned aerial vehicles to be allocated by the multiple targets can efficiently and quickly complete corresponding target emergency flight tasks through cooperative flight and clustered flight.

In one possible implementation manner, the embodiment of the present application provides three exemplary implementation manners of training the task policy response sub-network in the periodic task scheduling network, and a specific training process is as follows.

The first training method comprises the steps of obtaining flight path information of a training cluster in the flight training process of an unmanned aerial vehicle; acquiring a task policy response sub-network to be trained to train an emergency flight task determined by the training cluster flight path information; acquiring an unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process; generating first training sample information based on the training cluster flight path information, the training emergency flight task and the unmanned aerial vehicle task allocation training result; and using the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the task policy to respond to the sub-network to be mobilized, as a training requirement condition for successful task allocation, and training the task policy responding sub-network by using the first training sample information.

Therefore, through the training mode, the sub-network for responding to the task policy to be trained carries out multi-round training, namely after the model configuration of the sub-network for responding to the task policy is updated once, the sub-network for responding to the task policy after the model configuration is updated is used for scheduling the unmanned aerial vehicle to be deployed to take part in training again, training cluster flight path information in the flight training process of the unmanned aerial vehicle is obtained, a training emergency flight task determined by the sub-network for responding to the task policy according to the training cluster flight path information is obtained through the task policy, the unmanned aerial vehicle task distribution training result in the flight training process of the unmanned aerial vehicle is obtained, and further, based on first training sample information consisting of the training cluster flight path information, the training emergency flight task and the unmanned aerial vehicle task distribution training result, network parameters of the sub-network for responding to the task policy are optimized again. After a plurality of rounds of iterative training are carried out according to the training process, a task policy response sub-network which can be put into practical use can be obtained. And then can ensure the stability that the task policy responds to the subnetwork, can guarantee simultaneously that follow-up transfer target waits to allocate the accuracy of unmanned aerial vehicle response target emergency flight task.

And in the second training method, based on the corresponding relation between the emergency flight task response condition under the unmanned aerial vehicle cluster task policy corresponding to the task policy response sub-network and the response emergency flight task, a task node network (tree structure model) is generated as the task policy response sub-network.

According to the third training method, the task policy response sub-networks comprise an aerial survey task response sub-network and a rescue task response sub-network; the aerial survey task response subnetwork and the rescue task response subnetwork are trained by the following steps.

In the flight training process of the unmanned aerial vehicle, performing emergency task interaction processing by using the aerial survey task response sub-network and the rescue task response sub-network to be trained; acquiring flight path information of a training cluster in the flight training process of the unmanned aerial vehicle; acquiring a training aerial survey emergency flight task determined by the aerial survey task response sub-network according to the training cluster flight path information; acquiring a training and rescue emergency flight task determined by the rescue task response sub-network according to the training cluster flight path information; acquiring an unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process; generating second training sample information based on the training cluster flight path information, the training aerial survey emergency flight task and the unmanned aerial vehicle task allocation training result; generating third training sample information based on the training cluster flight path information, the training rescue emergency flight task and the unmanned aerial vehicle task allocation training result; for the aerial survey task response subnetwork, successfully distributing the task for the unmanned aerial vehicle task distribution training result of the unmanned aerial vehicle to be dispatched, which is mobilized by the aerial survey task response subnetwork, as a training requirement condition, and training the aerial survey task response subnetwork by using the second training sample information; and as for the rescue task response sub-network, successfully distributing the task by using the unmanned aerial vehicle task distribution training result of the unmanned aerial vehicle to be allocated, which is mobilized by the rescue task response sub-network, as a training requirement condition, and training the rescue task response sub-network by using the third training sample information.

In specific implementation, the task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a rescue task response sub-network; different aerial survey task response sub-networks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; the rescue task response sub-network is used for determining an emergency flight starting task according to the unmanned aerial vehicle cluster flight path information under different aerial survey standards; or the plurality of task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a plurality of rescue task response sub-networks; different aerial survey task response sub-networks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; different rescue task response sub-networks correspond to different rescue standards, and the different rescue standards are used for starting measurement strategies or measurement strategy combinations under different aerial survey standards.

In a possible implementation manner, in the flight training process of the unmanned aerial vehicle described above, the navigation task response sub-network and the rescue task response sub-network to be trained are used for performing emergency task interaction processing, and specifically, the emergency task interaction processing may include the contents described in steps B to D.

Step B, in the flight training process of the unmanned aerial vehicle, performing emergency task interaction processing by using the aerial survey task response sub-network in the J-th round and the rescue task response sub-network in the K-th round; j and K are integers greater than or equal to 0; the J is equal to the K, or the difference between the J and the K is 1.

And C, when the J is smaller than or equal to the K, responding to the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the aerial survey task response sub-network to be mobilized, as a training requirement condition for successful task allocation, and training the aerial survey task response sub-network by using the second training sample information.

And D, when J is larger than K, responding to the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the rescue task response sub-network to be mobilized, as a training requirement condition, and training the rescue task response sub-network by utilizing the third training sample information.

In a possible implementation manner, after training of each task policy response sub-network in the staged task scheduling network is completed, the unmanned aerial vehicle cluster mobilizing cloud platform can further train the task policies to select the sub-networks. The embodiment of the application provides two implementation modes of selecting the sub-networks by the task policy in the exemplary staged task scheduling network, and the specific training process is as follows.

In the first implementation mode, flight path information of a training cluster in the flight training process of an unmanned aerial vehicle is acquired; acquiring the task policy selection sub-network to be trained, and selecting a training task policy response sub-network from the plurality of task policy response sub-networks according to the training cluster flight path information; acquiring an unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process; generating fourth training sample information based on the training cluster flight path information, the training task policy response sub-network and the unmanned aerial vehicle task allocation training result; and training the task policy selection sub-network by utilizing the fourth training sample information, wherein the result of the task allocation training of the unmanned aerial vehicle to be allocated, which enables the task policy selection sub-network to be mobilized, is the successful task allocation as a training requirement condition.

In this way, by implementing the above embodiment, the sub-network is selected for the task policy to be trained to perform multiple rounds of iterative training; after the network parameters of the task policy selection sub-network are updated once, the updated network parameters are used for selecting the sub-network to mobilize the unmanned aerial vehicle to be allocated to participate in the unmanned aerial vehicle flight training again, training cluster flight path information in the unmanned aerial vehicle flight training process is obtained, the training task policy selection sub-network selected according to the training cluster flight path information through the task policy selection sub-network responds to the sub-network, the unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process is obtained, and further, the network parameters of the task policy selection sub-network are adjusted again based on fourth training sample information generated by the training cluster flight path information, the training task policy response sub-network and the unmanned aerial vehicle task allocation training result. And (4) after the training is performed for a plurality of rounds according to the steps, obtaining a task policy selection sub-network which can be put into practical use.

In the second implementation mode, in the flight training process of the unmanned aerial vehicle, a sub-network is selected by using the task policy to be trained to perform emergency task interaction processing with the sub-network; acquiring flight path information of a training cluster in the flight training process of the unmanned aerial vehicle; taking one of the task policy selection sub-networks as a target task policy selection sub-network, and acquiring a training task policy response sub-network selected by the target task policy selection sub-network in the plurality of task policy response sub-networks according to the training cluster flight path information; acquiring an unmanned aerial vehicle task allocation training result in the unmanned aerial vehicle flight training process; generating fifth training sample information based on the training cluster flight path information, the training task policy response sub-network and the unmanned aerial vehicle task allocation training result; and training the target task policy selection sub-network by utilizing the fifth training sample information, wherein the result of the task allocation training of the unmanned aerial vehicle to be allocated, which enables the target task policy selection sub-network to be mobilized, is the successful task allocation as a training requirement condition. In this way, the above embodiment enables the task policy picking sub-network to perform resistance round training with itself, thereby continuously improving the network performance of the task policy picking sub-network.

In some possible embodiments, the maneuvering of the target drone to be deployed in step 230 in response to the target emergency flight mission may include customizing a mission path of the target drone to be deployed, based on which step 230 may be implemented by the technical solutions described in steps 231-234 below.

231, acquiring a local flight data visual record and a global flight data visual record of the target unmanned aerial vehicle to be allocated; and matching the local flight data visual record and the global flight data visual record in the unmanned aerial vehicle to be allocated by the target based on the visual record correlation between the local flight data visual record and the global flight data visual record in the unmanned aerial vehicle to be allocated by the target to obtain a visual record matching result.

In a possible embodiment, the acquiring the visual record of the local flight data and the visual record of the global flight data in the drone to be deployed in step 231 may specifically include the following descriptions: acquiring at least two local flight data segments and at least two global flight data segments in the target unmanned aerial vehicle to be allocated; obtaining a local flight data segment quantitative commonality value and a local flight data segment difference between the at least two local flight data segments, and obtaining a global segment quantitative commonality value and a global flight data segment difference between the at least two global flight data segments; combining the at least two local flight data segments according to the quantitative commonality value of the local flight data segments and the difference of the local flight data segments to obtain a visual record of the local flight data in the target unmanned aerial vehicle to be allocated; a local flight data visualization record comprises at least one local flight data segment; combining the at least two global flight data segments according to the global segment quantitative commonality value and the global flight data segment difference to obtain a global flight data visual record in the target unmanned aerial vehicle to be allocated; one global flight data visualization record includes at least one global flight data segment. Therefore, the quantitative commonality value of the local flight data segments and the difference of the local flight data segments are combined for at least two local flight data segments, so that the visual record of the local flight data can be accurately obtained, and then at least two global flight data segments are combined according to the quantitative commonality value of the global segments and the difference of the global flight data segments on the premise of ensuring the accuracy of the visual record of the local flight data, so that the accuracy of the visual record of the global flight data obtained after combination can be ensured.

In a possible embodiment, the pairing of the local flight data visualization record and the global flight data visualization record in the target unmanned aerial vehicle to be deployed based on the visualization record correlation between the local flight data visualization record and the global flight data visualization record in the target unmanned aerial vehicle to be deployed described in step 231 to obtain a visualization record pairing result, which may specifically include the following description contents: determining the global flight data visual record in the target unmanned aerial vehicle to be allocated as a global flight data airspace record, and determining the local flight data visual record in the target unmanned aerial vehicle to be allocated as a local flight data airspace record; the global flight data segment in the global flight data airspace record is obtained from a target monitoring log of the unmanned aerial vehicle to be deployed aiming at the target; acquiring a local flight data segment in the target monitoring log; determining a segment quantitative commonality value between a local flight data segment in the target monitoring log and a local flight data segment in the local flight data airspace record as the visual record correlation between the global flight data airspace record and the local flight data airspace record; and when the correlation of the visual records is greater than or equal to a correlation threshold value, pairing the global flight data airspace records and the local flight data airspace records to obtain a visual record pairing result. Therefore, according to the visual record correlation between the local flight data visual record and the global flight data visual record in the unmanned aerial vehicle to be allocated, the local flight data visual record and the global flight data visual record are paired in real time, and the condition that the obtained visual record pairing result is inaccurate can be avoided.

Step 232, determining the global flight data visual record with abnormal pairing as a global flight data visual record to be matched, and determining the flight attitude characteristic of the unmanned aerial vehicle matched with the global flight data visual record to be matched according to the visual record quantitative commonality value between the global flight data visual record in the visual record pairing result and the global flight data visual record to be matched.

In one possible embodiment, the to-be-matched global flight data visualization record includes a first global flight data segment in the target to-be-deployed drone; the number of the visual record pairing results is at least two; and the global flight data visual records in each visual record pairing result respectively comprise a second global flight data segment in the unmanned aerial vehicle to be allocated by the target. Based on this, the determining, according to the visual record quantitative commonality value between the global flight data visual record in the visual record pairing result and the global flight data visual record to be matched, the flight attitude feature of the unmanned aerial vehicle matched with the global flight data visual record to be matched described in step 232 may specifically include the following description contents: acquiring a first visual record description of the visual record of the global flight data to be matched according to the first global flight data segment; respectively acquiring second visual record descriptions of the visual records of the global flight data in each visual record pairing result according to a second global flight data segment included in each visual record pairing result; acquiring Euclidean distances between the first visual record description and second visual record descriptions corresponding to each visual record pairing result respectively; determining a visual record quantitative commonality value between the visual records of the global flight data in each visual record matching result and the visual records of the global flight data to be matched according to the Euclidean distance to which the visual record matching result belongs; when the number of the target visual record pairing results is larger than a first number threshold and smaller than or equal to a second number threshold, determining the flight attitude characteristics of the unmanned aerial vehicle contained in the local flight data visual records in the target visual record pairing results as the flight attitude characteristics of the unmanned aerial vehicle matched with the global flight data visual records to be matched; the target visual record pairing result refers to a visual record pairing result of which the quantitative commonality value of the visual record is greater than or equal to the threshold value of the quantitative commonality value of the visual record. Therefore, the flight attitude characteristics of the unmanned aerial vehicle matched with the global flight data visual records to be matched are determined in a targeted manner based on the visual record quantitative commonality value between the global flight data visual records in each visual record matching result and the global flight data visual records to be matched.

Step 233, matching the flight attitude characteristics of the unmanned aerial vehicle matched with the to-be-matched global flight data visual record to obtain an attitude matching result; and determining a clustered unmanned aerial vehicle in the unmanned aerial vehicles to be allocated by the target and the flight attitude characteristics of the unmanned aerial vehicle corresponding to the clustered unmanned aerial vehicle according to the attitude pairing result and the visual record pairing result.

And 234, determining a target task path of the target unmanned aerial vehicle to be allocated according to a clustered unmanned aerial vehicle in the target unmanned aerial vehicle to be allocated and the unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle.

Executing the technical scheme described in the steps 231-234, and acquiring a local flight data visual record and a global flight data visual record in the target unmanned aerial vehicle to be deployed; matching the local flight data visual record and the global flight data visual record in the target unmanned aerial vehicle to be allocated based on the visual record correlation between the local flight data visual record and the global flight data visual record in the target unmanned aerial vehicle to be allocated to obtain a visual record matching result; determining the global flight data visual record with abnormal pairing as a global flight data visual record to be matched, and determining the flight attitude characteristics of the unmanned aerial vehicle matched with the global flight data visual record to be matched according to the visual record quantitative commonality value between the global flight data visual record in the visual record pairing result and the global flight data visual record to be matched; matching the flight attitude characteristics of the unmanned aerial vehicle matched with the visual record of the global flight data to be matched to obtain an attitude matching result; according to the attitude pairing result and the visual record pairing result, determining a clustered unmanned aerial vehicle in the unmanned aerial vehicles to be allocated and unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle, and determining a target task path of the unmanned aerial vehicle to be allocated through the clustered unmanned aerial vehicle in the unmanned aerial vehicles to be allocated and the unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle.

Therefore, the local flight data visual record and the global flight data visual record in the unmanned aerial vehicle to be allocated to the target can be obtained through the content, the clustered unmanned aerial vehicle in the unmanned aerial vehicle to be allocated to the target can be obtained through the global flight data visual record, the flight attitude characteristic of the unmanned aerial vehicle in the unmanned aerial vehicle to be allocated to the target can be obtained through the local flight data visual record, the abnormal global flight data visual record can be matched with the corresponding flight attitude characteristic of the unmanned aerial vehicle, the accuracy of the clustered unmanned aerial vehicle and the flight attitude characteristic of the unmanned aerial vehicle to be allocated to the obtained target is improved, and the accuracy of the target task path of the unmanned aerial vehicle to be allocated to the determined target is further ensured.

On the basis, please refer to fig. 3, based on the same inventive concept, the present application further provides an unmanned aerial vehicle cluster mobilization processing system 30, which includes an unmanned aerial vehicle cluster mobilization cloud platform 10 and a plurality of unmanned aerial vehicles 20 that are in communication with each other;

the unmanned aerial vehicle cluster transferring cloud platform is used for acquiring flight path information of a target unmanned aerial vehicle cluster; determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selecting sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information; and moving the target to be allocated to the unmanned aerial vehicle to respond to the target emergency flight task.

Further, a readable storage medium is provided, on which a program is stored, which when executed by a processor implements the method described above.

In conclusion, by implementing the technical scheme, the flight path information of the target unmanned aerial vehicle cluster is obtained firstly; secondly, determining a target emergency flight task according to the information of the cluster flight path of the target unmanned aerial vehicle through a staged task scheduling network, wherein the staged task scheduling network comprises a task policy selection sub-network and a plurality of task policy response sub-networks which are not related to each other, the task policy selection sub-network is used for selecting a task policy response sub-network which is required to be used from the plurality of task policy response sub-networks according to the information of the cluster flight path of the unmanned aerial vehicle, and the task policy response sub-network is used for determining the emergency flight task which is required to be allocated to the unmanned aerial vehicle according to the information of the cluster flight path of the unmanned aerial vehicle; and finally, moving the target to be allocated with the unmanned aerial vehicle to respond to the target emergency flight task.

It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. An unmanned aerial vehicle cluster mobilization processing method is characterized by comprising the following steps:

acquiring flight path information of a target unmanned aerial vehicle cluster;

determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selection sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information;

the unmanned aerial vehicle to be allocated for moving the target responds to the target emergency flight task;

the step of determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network comprises the following steps:

correspondingly, the plurality of task policy response sub-networks comprise an aerial survey task response sub-network and a rescue task response sub-network; the determining whether to suspend responding to the subnetwork with the target task policy according to the target drone cluster attribute includes:

when the target task policy response subnetwork is the aerial survey task response subnetwork, detecting whether the target emergency flight task corresponds to a termination label of the aerial survey task response subnetwork, and if so, suspending use of the aerial survey task response subnetwork; when the target task policy response sub-network is the rescue task response sub-network, detecting whether the target unmanned aerial vehicle to be allocated is interfered by the emergency flight task of the associated unmanned aerial vehicle, if so, suspending the use of the rescue task response sub-network; or when the target task policy response sub-network is the rescue task response sub-network, detecting whether the accumulated use time of the rescue task response sub-network exceeds a set use duration value, and if so, suspending use of the rescue task response sub-network.

2. The method of claim 1, wherein the task guideline response subnetwork is trained by:

3. The method of claim 1, wherein the task guideline response subnetwork is generated by:

4. The method of claim 1, wherein the plurality of task guideline response subnetworks comprise a navigation task response subnetwork and a rescue task response subnetwork; training the aerial survey task response subnetwork and the rescue task response subnetwork by:

when J is larger than K, responding to the unmanned aerial vehicle task allocation training result of the unmanned aerial vehicle to be allocated, which enables the rescue task response sub-network to be mobilized, as a training requirement condition, and training the rescue task response sub-network by utilizing the third training sample information;

correspondingly, the task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a rescue task response sub-network; different aerial survey task response sub-networks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; the rescue task response sub-network is used for determining an emergency flight starting task according to the unmanned aerial vehicle cluster flight path information under different aerial survey standards; or the plurality of task policy response sub-networks comprise a plurality of aerial survey task response sub-networks and a plurality of rescue task response sub-networks; different aerial survey task response sub-networks correspond to different aerial survey standards, and the different aerial survey standards comprise different measurement strategies or measurement strategy combinations; different rescue task response sub-networks correspond to different rescue standards, and the different rescue standards are used for starting measurement strategies or measurement strategy combinations under different aerial survey standards.

5. The method of claim 1, wherein after training the plurality of task guideline response sub-networks, training the task guideline selection sub-network by:

6. The method of claim 1, wherein after training the plurality of task guideline response sub-networks, training the task guideline selection sub-network by:

in the flight training process of the unmanned aerial vehicle, selecting a sub-network by using the task policy to be trained to perform emergency task interaction processing with the unmanned aerial vehicle;

7. The method of claim 1, wherein maneuvering a target drone to be deployed in response to the target emergency flight mission comprises:

determining the abnormal global flight data visual record to be a to-be-matched global flight data visual record, and determining the unmanned aerial vehicle flight attitude characteristic matched with the to-be-matched global flight data visual record according to the visual record quantitative commonality value between the global flight data visual record and the to-be-matched global flight data visual record in the visual record matching result;

matching the flight attitude characteristics of the unmanned aerial vehicle matched with the to-be-matched global flight data visual record to obtain an attitude matching result; determining a clustered unmanned aerial vehicle in the target unmanned aerial vehicle to be allocated and the unmanned aerial vehicle flight attitude characteristics corresponding to the clustered unmanned aerial vehicle according to the attitude pairing result and the visual record pairing result;

8. The utility model provides an unmanned aerial vehicle cluster moves processing system which characterized in that, the system includes unmanned aerial vehicle cluster that intercommunicates moves cloud platform and a plurality of unmanned aerial vehicle:

the unmanned aerial vehicle cluster transferring cloud platform is used for acquiring flight path information of a target unmanned aerial vehicle cluster; determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network; the stage task scheduling network comprises a task policy selecting sub-network and a plurality of task policy response sub-networks which are not related to each other; the task policy selection sub-network is used for selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the unmanned aerial vehicle cluster flight path information; the task policy response sub-network is used for determining an emergency flight task required by the unmanned aerial vehicle to be allocated according to the unmanned aerial vehicle cluster flight path information; the unmanned aerial vehicle to be allocated for moving the target responds to the target emergency flight task;

determining a target emergency flight task according to the target unmanned aerial vehicle cluster flight path information through a periodic task scheduling network, wherein the step of determining the target emergency flight task comprises the following steps:

detecting whether a target task policy response sub-network in use exists currently; if yes, responding to a sub-network through the target task policy, and determining the target emergency flight task according to the target unmanned aerial vehicle cluster flight path information; if not, selecting a sub-network through the task policy, and selecting a task policy response sub-network to be used from the plurality of task policy response sub-networks according to the target unmanned aerial vehicle cluster flight path information to serve as a target task policy response sub-network;

correspondingly, after the sub-network is responded through the target task policy and the target emergency flight task is determined according to the target unmanned aerial vehicle cluster flight path information, the system further includes:

when the target task policy response subnetwork is the aerial survey task response subnetwork, detecting whether the target emergency flight task corresponds to a termination label of the aerial survey task response subnetwork, and if so, suspending use of the aerial survey task response subnetwork; when the target task policy response sub-network is the rescue task response sub-network, detecting whether the target unmanned aerial vehicle to be deployed is interfered by the emergency flight task of the associated unmanned aerial vehicle, and if so, suspending the use of the rescue task response sub-network; or when the target task policy response sub-network is the rescue task response sub-network, detecting whether the accumulated use time of the rescue task response sub-network exceeds a set use duration value, and if so, suspending use of the rescue task response sub-network.

9. An unmanned aerial vehicle cluster mobilizing cloud platform is characterized by comprising a processor and a memory; the processor is in communication with the memory, and the processor is configured to read the computer program from the memory and execute the computer program to implement the method of any one of claims 1 to 7.