A distributed collaborative task processing method, device, equipment and storage medium

Abstract

The embodiments of the present application provide a distributed collaborative task processing method, device, equipment and storage medium, which belongs to the field of computer. The method includes: when obtaining the target task to be executed by the drone cluster, judging whether the computing resources required for the target task exceed the computing threshold; if not, assigning the target task to the first node for processing; if exceeded, performing distributed task processing; based on the processing result of the first node or the distributed task processing result, planning the flight path of the drone cluster in the target task. The present application aims to improve the accuracy and real-time performance of the drone cluster in processing tasks.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and more specifically, to a distributed collaborative task processing method, apparatus, device, and storage medium. Background Art

[0002] Drone swarms have the advantages of low cost, large number, flexibility and convenience. When natural disasters such as earthquakes, tsunamis or forest fires occur, the communication situation in the disaster area is generally complicated, and it is not easy for rescue personnel to enter the disaster area. At this time, through the advantages of drone swarms, it is possible to carry out communication networking, conduct preliminary exploration of the situation in the disaster area or conduct preliminary fire extinguishing during emergency rescue. However, when performing these tasks, the route planning and decision-making processing methods of drone swarms need to be more intelligent. If they only rely on manual operation, the relatively poor communication environment will cause the drone swarm to be unable to complete the above-mentioned emergency rescue and forest fire extinguishing tasks.

[0003] When a drone cluster needs to perform route planning or online decision-making, target recognition is required to guide the drone cluster to make more intelligent decisions. However, due to the computing resource limitations of a single drone, it is difficult to guarantee the accuracy and real-time performance of a single drone in a drone cluster when performing online recognition of large-pixel, high-resolution image data. Summary of the invention

[0004] The embodiments of the present application provide a distributed collaborative task processing method, apparatus, device and storage medium, which aim to improve the accuracy and real-time performance of drone clusters when processing tasks.

[0005] In a first aspect, an embodiment of the present application provides a distributed collaborative task processing method, which is applied to a drone cluster consisting of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, the method is applied to the master control node, and the method includes:

[0006] When the target task to be performed by the drone cluster is obtained, it is determined whether the computing resources required for the target task exceed the computing threshold; wherein the target task is a task of performing path planning for the fire-fighting task of the multiple drones in the forest fire-fighting / or emergency rescue task;

[0007] If not, the target task is assigned to the first node for processing;

[0008] If it exceeds, distributed task processing is performed, and the distributed task processing includes: based on the real-time occupancy rate of computing resources of each node in the drone cluster, determining whether each node in the drone cluster meets the execution of multiple target subtasks; the target subtask is a task obtained by executing distributed task processing on the target task; if so, sending the multiple target subtasks to the corresponding second nodes respectively; if not, repeating the above distributed task processing steps until each node in the drone cluster meets the execution of the multiple target subtasks obtained by re-dividing;

[0009] Based on the processing result of the first node or the distributed task processing result, plan the flight path of the drone cluster in the target task.

[0010] Optionally, the determining whether the computing resources required for the target task exceed a computing threshold includes:

[0011] Obtaining pixel number information of image data in the target task;

[0012] Based on the pixel number information of the image data and a preset pixel number threshold, it is determined whether the computing resources required for the target task exceed the computing threshold.

[0013] Optionally, determining whether the computing resources required for the target task exceeds the computing threshold based on the pixel number information of the image data and a preset pixel number threshold includes:

[0014] If the number of pixels of the target task is greater than or equal to the preset pixel number threshold, the computing resources required for the target task exceed the computing threshold;

[0015] If the number of pixels of the target task is less than the preset pixel number threshold, the computing resources required for the target task do not exceed the computing threshold.

[0016] Optionally, determining whether each node in the drone cluster satisfies the execution of the multiple target subtasks based on the real-time occupancy rate of computing resources of each node in the drone cluster includes:

[0017] Constructing and training a computing resource occupancy rate database, wherein the computing resource occupancy rate database is used to characterize the computing resource occupancy of each node in the drone cluster when calculating various forest fire fighting tasks;

[0018] Based on the computing resource occupancy rate database, calculating the estimated computing resource occupancy rate of the target task;

[0019] Determine whether there are multiple nodes in the drone cluster whose real-time computing resource occupancy rate exceeds the estimated computing resource occupancy rate;

[0020] If so, each node in the drone cluster satisfies the execution of the multiple target subtasks;

[0021] If not, each node in the drone cluster does not meet the execution of the multiple target subtasks.

[0022] Optionally, the method further comprises:

[0023] Obtaining the real-time occupancy rate and attribute information of computing resources of each node in the drone cluster, wherein the attribute information includes at least: state information, decision target type and location information of each node;

[0024] In response to an online decision task, the online decision task is used to determine a target location of the drone cluster;

[0025] The online decision-making task and the attribute information of each node are sent to a third node with the lowest real-time computing resource occupancy rate, so as to output the target location to be reached by each node in the drone cluster at the next moment through the third node.

[0026] Optionally, each node in the drone cluster is also communicatively connected to a ground auxiliary node, and the method further comprises:

[0027] Acquire the communication status with the ground auxiliary node and the timeliness of the target task;

[0028] If the target task is of the first timeliness and the communication state satisfies the target communication state, sending the target task to the ground auxiliary node for execution;

[0029] Obtain the execution result returned by the ground auxiliary node after executing the target task.

[0030] Optionally, the method further comprises:

[0031] Regularly obtain the communication connection between the nodes in the drone cluster and the master control node;

[0032] If multiple nodes in the drone cluster detect that the communication connection with the master control node is interrupted, one of the nodes in the drone cluster other than the master control node is determined as a new master control node.

[0033] In a second aspect, an embodiment of the present application provides a distributed collaborative task processing device for executing a distributed collaborative task processing method according to any one of the first aspects, which is applied to a drone cluster composed of multiple drones, wherein the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, and the drone cluster includes a master control node. The method is applied to the master control node, and the device includes: a first judgment module, a distributed processing module, and a planning module, wherein:

[0034] A first judgment module is used to judge whether the computing resources required for the target task to be executed by the drone cluster exceed the computing threshold when the target task to be executed by the drone cluster is acquired; wherein the target task is a task of performing path planning for the fire-fighting tasks of the multiple drones in the forest fire fighting; if it does not exceed, the target task is assigned to the first node for processing; if it exceeds, distributed task processing is performed;

[0035] A distributed processing module is used to determine whether each node in the drone cluster meets the execution of the multiple target subtasks based on the real-time occupancy rate of the computing resources of each node in the drone cluster; the target subtask is a task obtained by executing distributed task processing on the target task; if so, the multiple target subtasks are sent to the corresponding second nodes respectively; if not, the above distributed task processing steps are repeated until each node in the drone cluster meets the execution of the multiple target subtasks obtained by re-dividing;

[0036] A planning module is used to plan the flight path of the drone cluster in the target task based on the processing result of the first node or the distributed task processing result.

[0037] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when executed, the processor implements a distributed collaborative task processing method as described in any one of the first aspects.

[0038] In a fourth aspect, an embodiment of the present application provides a storage medium, which stores a computer program that enables a processor to execute a distributed collaborative task processing method as described in any one of the first aspects.

[0039] Beneficial effects: The present application first determines whether the computing resources existing in one of the nodes of the drone cluster can achieve the target task. If the computing resources required for the target task do not exceed the computing threshold, it means that the computing resources of a single first node in the drone cluster can process the target task, and the first node is directly used to process the target task; if the computing resources required for the target task exceed the computing threshold, it means that the computing resources in the first node cannot process the target task, and distributed task processing is performed at this time; when performing distributed task processing, the real-time occupancy rate of the computing resources of each node in the drone cluster is first determined, and the remaining amount of computing resources of each node in the drone cluster can be determined by calculating the real-time occupancy rate of the computing resources; then the target task is divided into multiple target subtasks, and sent to multiple second nodes respectively, and the multiple second nodes process multiple target subtasks respectively. Because the computing resources in multiple second nodes can meet the execution of the target subtasks, the results obtained by multiple second nodes processing the target subtasks are more accurate, and compared with the method of processing by one node in the drone cluster, the processing is faster and more real-time; thereby achieving the effect of improving the accuracy and real-time performance of the drone cluster when processing tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0041] Figure 1 is a flowchart of the steps of a distributed collaborative task processing method proposed in an embodiment of the present application;

[0042] Figure 2 It is a functional module diagram of a distributed collaborative task processing device proposed in another embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] Embodiment 1

[0045] Reference Figure 1, shows a flow chart of steps of a distributed collaborative task processing method in an embodiment of the present invention, the task processing method is applied to a drone cluster consisting of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, and the method is applied to the master control node, such as Figure 1 As shown, the processing method may specifically include the following steps:

[0046] S1, when obtaining the target task to be performed by the drone cluster, determining whether the computing resources required for the target task exceed the computing threshold; wherein the target task is a task of performing path planning for the fire-fighting task of the multiple drones in the forest fire fighting;

[0047] When the drone cluster performs a target task, the master control node first judges the target task to determine whether the computing resources of the target task exceed the computing threshold; the computing threshold is a boundary value set in the master control node, which is the reasonable range of computing resources of a single node in the drone cluster that can handle the target task: when the computing resources required for the target task do not exceed the computing threshold, it means that a single node in the drone cluster can handle the target task; when the computing resources required for the target task exceed the computing threshold, it means that a single node in the drone cluster cannot handle the target task.

[0048] S11, if not exceeded, assigning the target task to the first node for processing;

[0049] As long as a node in the drone cluster has internal computing resources that exceed the computing resources required to perform the target task, the node can be used as the first node.

[0050] S12, if exceeded, distributed task processing is performed;

[0051] The distributed task processing includes:

[0052] S121, based on the real-time occupancy rate of computing resources of each node in the drone cluster, determining whether each node in the drone cluster meets the requirements for executing the multiple target subtasks; the target subtask is a task obtained by executing distributed task processing on the target task;

[0053] The real-time occupancy rate of computing resources is a parameter used to reflect the current remaining computing resources of each node in the drone cluster. By calculating the real-time occupancy rate of computing resources, it is possible to determine whether each node in the drone cluster can execute the target subtask. The target subtask is divided from the target task. Compared with the target task, a single target subtask requires fewer computing resources. By jointly processing the target task through each node in the drone cluster, the real-time performance of the target task processing can be improved.

[0054] For example, in a forest fire fighting mission, the purpose of the target task is to plan a flight path for forest fire fighting, then the target subtask is to plan the flight path for forest fire fighting, for example, based on the image acquired by the drone cluster, identify the fire point in the image; when a single node in the drone cluster is unable to process the acquired image, the image is divided into multiple sub-images according to distributed task processing, and the task of identifying the fire points in multiple sub-images is multiple target subtasks. By processing the multiple target subtasks separately, the fire points in the images corresponding to the multiple target subtasks can also be obtained, and the identification is completed, thereby planning the flight path for the drone cluster.

[0055] S1211, if yes, sending the multiple target subtasks to the corresponding second nodes respectively;

[0056] S1212: If not, repeat the above steps of distributed task processing until each node in the drone cluster satisfies the execution of multiple target subtasks obtained by re-division.

[0057] After all target tasks are divided into multiple target subtasks, it is determined whether each node in the drone cluster can execute the target subtask. If it can, the node used to execute multiple target subtasks is the second node; if it cannot, it means that the level of target task division is not enough, and there are not enough nodes in the drone cluster to process the target subtasks. Then, it jumps back to the division step and re-divides the target task. After the division again, the computing resources required to process a single target subtask are less. Through the process of division and judgment, the target task can be divided into multiple target subtasks and sent to the second node for processing, thereby completing the processing of the target task.

[0058] S2: Based on the processing result of the first node or the distributed task processing result, plan the flight path of the drone cluster in the target task.

[0059] The flight path of the drone cluster can be obtained through the processing results of the first node or the distributed task processing results. The drone cluster can fly to the target location through the flight path to perform fire-fighting tasks in forest fire fighting.

[0060] In this embodiment, it is first determined whether the computing resources existing in one of the nodes in the drone cluster can achieve the target task. If the computing resources required for the target task do not exceed the computing threshold, it means that the computing resources of a single first node in the drone cluster can process the target task, and the first node is directly used to process the target task; if the computing resources required for the target task exceed the computing threshold, it means that the computing resources in the first node cannot process the target task, and distributed task processing is performed at this time; when performing distributed task processing, the real-time occupancy rate of the computing resources of each node in the drone cluster is first determined, and the remaining amount of computing resources of each node in the drone cluster can be determined by calculating the real-time occupancy rate of the resources; then the target task is divided into multiple target subtasks, and sent to multiple second nodes respectively, and the multiple second nodes process the multiple target subtasks respectively. Because the computing resources in the multiple second nodes can meet the execution of the target subtasks, the results obtained by the multiple second nodes processing the target subtasks are more accurate, and compared with the method of using one of the nodes in the cluster for processing, the processing is faster and more real-time; thereby achieving the effect of improving the accuracy and real-time performance of the drone cluster when processing tasks.

[0061] Embodiment 2

[0062] Reference Figure 1 As shown, a flowchart of the steps of a distributed collaborative task processing method of an embodiment of the present application is shown, the task processing method is applied to a drone cluster composed of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, and the method is applied to the master control node, such as Figure 1 As shown, the processing method may specifically include the following steps:

[0063] S1. When the target task to be executed by the drone cluster is obtained, determine whether the computing resources required for the target task exceed the computing threshold; wherein the target task is the task of path planning for the fire-fighting task of the multiple drones in the forest firefighting.

[0064] Drone swarms have the advantages of low cost, large number, flexibility and convenience. Drone swarms perform given tasks together through autonomous collaboration. The premise of autonomous collaboration of drone swarms is that the drones in the drone swarm can communicate with each other and share identification information and decision-making information. Drone swarms have natural advantages in communication networking, remote sensing earth observation, and collaborative execution of tasks in emergency rescue. For example, when natural disasters occur in a certain area, such as landslides, earthquakes or tsunamis, natural disasters can easily cause communication problems in the affected area. In this case, by using drone swarms, it is possible to temporarily establish a communication connection with the disaster-stricken area, and use drone swarms to take photos of the disaster-stricken area or deliver daily necessities to people trapped in the disaster-stricken area.

[0065] When a drone cluster performs a mission, identification information and decision information are needed to guide the actions of the drone cluster. The target task provides identification information for the drone cluster, and the decision task provides decision information for the drone cluster. Through identification information and decision information, the flight path and target location of the drone cluster can be planned.

[0066] For example, when a drone cluster is performing a forest fire prevention task, it first obtains image information through the sensors installed on the drone cluster and generates a target task. After the drone cluster processes the target task, it judges the distance and direction to the target location and whether there are obstacles based on the identification information obtained during the processing, plans the flight path of the drone cluster, and guides the drone cluster to fly.

[0067] Wherein, judging whether the computing resources required for the target task exceed the computing threshold includes:

[0068] Obtaining pixel number information of image data in the target task;

[0069] Based on the pixel number information of the image data and a preset pixel number threshold, it is determined whether the computing resources required for the target task exceed the computing threshold.

[0070] Including: if the number of pixels of the target task is greater than or equal to a preset pixel number threshold, the computing resources required for the target task exceed the computing threshold;

[0071] If the number of pixels of the target task is less than the preset pixel number threshold, the computing resources required for the target task do not exceed the computing threshold.

[0072] The preset pixel number threshold is pre-set in the master control node. When setting the preset pixel number threshold, the upper limit of the preset pixel number threshold is the upper limit of the image data that a single node in the drone cluster can process, and the lower limit of the preset pixel number threshold is the value of the quotient of the pixel number information of the target task and the number of nodes in the drone cluster. When pre-setting the preset pixel number threshold, it is set according to the actual situation and the tasks that the drone cluster needs to perform.

[0073] When the drone cluster obtains identification information through the target task, it mainly obtains the identification information by collecting image data and processing the image data; for example, the drone cluster collects image data of a mountain, and after identification processing, the identification information obtained is a mountain. Then the master control node of the drone cluster will control the drone cluster to fly upward or to the sides according to the obtained identification information, bypassing the mountain to avoid the drone cluster from crashing into the mountain. Therefore, when judging whether the computing resources required for the target task exceed the computing threshold, the pixel number information of the image data in the target task is used as the judgment condition.

[0074] S11, if not exceeded, assigning the target task to the first node for processing;

[0075] The first node is used to process the target task, and is itself a node in the drone cluster, which can be one of the drones in the drone cluster; it has the computing resources required to process the target task.

[0076] S12, if exceeded, distributed task processing is performed;

[0077] Distributed task processing divides the target task into several target subtasks, and aggregates the results of the distributed processing of several target subtasks to obtain the result of processing the target task; through distributed task processing, the target task can be processed when a single node in the drone cluster cannot process the target task.

[0078] The distributed task processing includes:

[0079] S121, based on the real-time occupancy rate of computing resources of each node in the drone cluster, determining whether each node in the drone cluster meets the requirements for executing the multiple target subtasks; the target subtask is a task obtained by executing distributed task processing on the target task;

[0080] Construct and train a computing resource occupancy rate database, which is used to characterize the computing resource occupancy of each node in the drone cluster when calculating multiple target tasks;

[0081] Based on the computing resource occupancy rate database, calculating the estimated computing resource occupancy rate of the target task;

[0082] Determine whether there are multiple nodes in the drone cluster whose real-time computing resource occupancy rate exceeds the estimated computing resource occupancy rate,

[0083] If so, each node in the drone cluster satisfies the execution of the multiple target subtasks;

[0084] If not, each node in the drone cluster does not meet the execution of the multiple target subtasks.

[0085] The real-time computing resource occupancy rate is the actual feedback of the current computing resource occupancy of each node of the drone, and the estimated resource occupancy rate is the estimated computing resources required to execute the target subtask; if the real-time computing resource occupancy rate of multiple nodes exceeds the estimated computing resource occupancy rate, it means that in the current drone cluster, there are enough computing resources on the nodes to process the target subtask; if the real-time computing resource occupancy rate of multiple nodes does not exceed the estimated computing resource occupancy rate, it means that in the current drone cluster, there are not enough computing resources on the nodes to process the target subtask.

[0086] S1211, if yes, sending the multiple target subtasks to the corresponding second nodes respectively;

[0087] The second node is a node for processing the target subtask. Through the above judgment, the real-time occupancy rate of computing resources on this part of the second nodes is greater than the estimated occupancy rate of computing resources, and the target subtask can be processed.

[0088] S1212: If not, repeat the above steps of distributed task processing until each node in the drone cluster satisfies the execution of multiple target subtasks obtained by re-division.

[0089] If the judgment result is "No", it means that there are not enough nodes in the current drone cluster to meet the requirements of processing the target subtasks. In this case, the target task needs to be divided again, and when dividing, the target task needs to be divided into fewer target subtasks than the first division, until there are enough nodes in the drone cluster to process the target subtasks.

[0090] The method further comprises,

[0091] Obtaining the real-time occupancy rate and attribute information of computing resources of each node in the drone cluster, wherein the attribute information includes at least state information, decision target type and location information of each node;

[0092] In response to an online decision-making task, the online decision-making task and the attribute information of each node are sent to a third node with the lowest real-time computing resource occupancy rate, so as to output the target task to be executed by each node in the drone cluster at the next moment through the third node.

[0093] Each node in the drone cluster is also in communication connection with a ground auxiliary node, and the method further comprises:

[0094] Acquire the communication status with the ground auxiliary node and the timeliness of the target task;

[0095] If the target task is of the first timeliness and the communication state satisfies the target communication state, sending the target task to the ground auxiliary node for execution;

[0096] Obtain the execution result returned by the ground auxiliary node after executing the target task.

[0097] The ground auxiliary node set on the ground is used to assist the UAV cluster to complete the target task. When the UAV cluster is performing the task, it first obtains the communication status between the UAV cluster and the ground auxiliary node and the timeliness of the target task; if the target task is the first timeliness, it means that the target task is not a task that needs to achieve high real-time performance and can have a certain delay; and at the same time, the communication status meets the target communication status, which means that under the current situation, the communication status between the UAV cluster and the ground auxiliary node can meet the target task to be sent to the ground auxiliary node for execution, and after the ground auxiliary node completes the target task, it can return the execution result returned after the execution to the UAV cluster; by setting up the ground auxiliary node to assist the UAV cluster, when the communication status is met, sending the target task to the ground auxiliary node for completion can reduce the computing pressure of the UAV cluster, and allocate limited computing resources to other tasks, so as to further improve the accuracy and real-time performance of the UAV cluster.

[0098] The method further comprises:

[0099] Regularly obtain the communication connection between the nodes in the drone cluster and the master control node;

[0100] If multiple nodes in the drone cluster detect that the communication connection with the master control node is interrupted, one of the nodes in the drone cluster is determined as a new master control node.

[0101] In this embodiment, the master control node is a node used to control the entire drone cluster. The master control node needs to continuously allocate target tasks and decision-making tasks and process them through the computing resources of the entire drone cluster, and plan the flight path and target location of the drone cluster according to the returned processing results. The master control node is irreplaceable in this embodiment; during the working process of the drone cluster, by regularly obtaining the communication connection between the nodes in the drone cluster and the master control node, it can be determined whether the master control node in the current drone cluster is separated from the drone cluster. If the master control node is separated from the drone cluster, it is necessary to determine a new node as the master control node; when determining the new master control node, directly determine the node with the lowest current computing resource occupancy rate as the new master control node. Through the above method of determining a new master control node, if the master control node is separated from the drone cluster for some reason, the drone cluster can also quickly generate a new master control node without confusion.

[0102] In this embodiment, it is first determined whether the computing resources existing in one of the nodes of the drone cluster can achieve the target task. If the computing resources required for the target task do not exceed the computing threshold, it means that the computing resources of a single first node in the drone cluster can process the target task, and the first node is directly used to process the target task; if the computing resources required for the target task exceed the computing threshold, it means that the computing resources in the first node cannot process the target task, and distributed task processing is performed at this time; when performing distributed task processing, the real-time occupancy rate of the computing resources of each node in the drone cluster is first determined, and the remaining amount of computing resources of each node in the drone cluster can be determined by calculating the real-time occupancy rate of the resources; then the target task is divided into multiple target subtasks, and sent to multiple second nodes respectively, and the multiple second nodes process the multiple target subtasks respectively, because the computing resources in the multiple second nodes can meet the execution of the target subtasks, so the results obtained by the multiple second nodes processing the target subtasks are more accurate, and compared with the method of using one of the nodes in the cluster for processing, the processing is faster and more real-time; thereby achieving the effect of improving the accuracy and real-time performance of the drone cluster when processing tasks.

[0103] Embodiment 3

[0104] Based on the same inventive concept, Figure 2 A schematic diagram of a distributed collaborative task processing device is shown, referring to Figure 2 As shown, it is applied to a drone cluster composed of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, the method is applied to the master control node, and the device includes: a first judgment module, a distributed processing module, a planning module, a ground auxiliary module and a master control node determination module, wherein:

[0105] A first judgment module is used to judge whether the computing resources required for the target task to be executed by the drone cluster exceed the computing threshold when the target task to be executed by the drone cluster is acquired; wherein the target task is a task of performing path planning for the fire-fighting tasks of the multiple drones in the forest fire fighting; if it does not exceed, the target task is assigned to the first node for processing; if it exceeds, distributed task processing is performed;

[0106] A distributed processing module is used to determine whether each node in the drone cluster meets the execution of the multiple target subtasks based on the real-time occupancy rate of the computing resources of each node in the drone cluster; the target subtask is a task obtained by executing the distributed task processing of the target task; if so, the multiple target subtasks are sent to the corresponding second nodes respectively; if not, the above distributed task processing steps are repeated until each node in the drone cluster meets the execution of the multiple target subtasks obtained by re-dividing; a planning module

[0107] A planning module is used to plan the flight path of the drone cluster in the target task based on the processing result of the first node or the distributed task processing result.

[0108] Ground auxiliary module, each node in the UAV cluster is also connected to the ground auxiliary node in communication, and the method also includes: obtaining the communication status between the ground auxiliary node and the timeliness of the target task; if the target task is the first timeliness and the communication status meets the target communication status, the target task is sent to the ground auxiliary node for execution; and obtaining the execution result returned by the ground auxiliary node after executing the target task.

[0109] A master control node determination module is configured to periodically obtain the communication connection between the nodes in the drone cluster and the master control node; if multiple nodes in the drone cluster detect that the communication connection with the master control node is interrupted, one of the nodes in the drone cluster is determined as a new master control node;

[0110] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0111] Embodiment 4

[0112] Based on the same inventive concept, embodiment 4 of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes, it implements a distributed collaborative task processing method as described in any one of embodiments 1 and 2.

[0113] Embodiment 5

[0114] Based on the same inventive concept, embodiment 5 of the present application provides a computer-readable storage medium, which stores a computer program that enables a processor to execute a distributed collaborative task processing method as described in any one of embodiments 1 to 2.

[0115] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0116] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0117] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0118] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0120] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0121] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0122] Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A distributed collaborative task processing method, characterized in that: The method is applied to a drone cluster composed of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, and the method is applied to the master control node. The method includes: When the target task to be performed by the drone cluster is obtained, it is determined whether the computing resources required for the target task exceed the computing threshold; wherein the target task is a task of performing path planning for the fire fighting task of the multiple drones in the forest fire fighting and / or emergency rescue task; If not, the target task is assigned to the first node for processing; If it exceeds, distributed task processing is performed, and the distributed task processing includes: based on the real-time occupancy rate of computing resources of each node in the drone cluster, determining whether each node in the drone cluster meets the execution of multiple target subtasks; the multiple target subtasks are tasks obtained by executing distributed task processing on the target task; if so, sending the multiple target subtasks to the corresponding second nodes respectively; if not, repeating the above distributed task processing steps until each node in the drone cluster meets the execution of the multiple target subtasks obtained by re-dividing; Planning a flight path of the drone cluster in the target task based on the processing result of the first node or the distributed task processing result; The determining whether the computing resources required for the target task exceed the computing threshold includes: Obtaining pixel number information of image data in the target task; Based on the pixel number information of the image data and the preset pixel number threshold, determine whether the computing resources required for the target task exceed the computing threshold, wherein the preset pixel number threshold is pre-set in the master control node, and when setting the preset pixel number threshold, the upper limit of the preset pixel number threshold is the upper limit value of the image data that can be processed by a single node in the drone cluster, and the lower limit of the preset pixel number threshold is the value of the quotient of the pixel number information of the target task and the number of nodes in the drone cluster; Determining whether the computing resources required for the target task exceed the computing threshold based on the pixel number information of the image data and a preset pixel number threshold includes: If the number of pixels of the target task is greater than or equal to the preset pixel number threshold, the computing resources required for the target task exceed the computing threshold; If the number of pixels of the target task is less than the preset pixel number threshold, the computing resources required for the target task do not exceed the computing threshold.

2. A distributed collaborative task processing method according to claim 1, characterized in that: The determining whether each node in the drone cluster satisfies the execution of the multiple target subtasks based on the real-time occupancy rate of computing resources of each node in the drone cluster includes: Constructing and training a computing resource occupancy rate database, wherein the computing resource occupancy rate database is used to characterize the computing resource occupancy of each node in the drone cluster when calculating various forest fire fighting tasks; Based on the computing resource occupancy rate database, calculating the estimated computing resource occupancy rate of the target task; Determine whether there are multiple nodes in the drone cluster whose real-time computing resource occupancy rate exceeds the estimated computing resource occupancy rate; If so, each node in the drone cluster satisfies the execution of the multiple target subtasks; If not, each node in the drone cluster does not meet the execution of the multiple target subtasks.

3. A distributed collaborative task processing method according to claim 2, characterized in that: The method further comprises: Obtaining the real-time occupancy rate and attribute information of computing resources of each node in the drone cluster, wherein the attribute information includes at least: state information, decision target type and location information of each node; In response to an online decision task, the online decision task is used to determine a target location of the drone cluster; The online decision-making task and the attribute information of each node are sent to a third node with the lowest real-time computing resource occupancy rate, so as to output the target location to be headed by each node in the drone cluster at the next moment through the third node.

4. A distributed collaborative task processing method according to claim 3, characterized in that: Each node in the drone cluster is also in communication connection with a ground auxiliary node, and the method further comprises: Acquire the communication status with the ground auxiliary node and the timeliness of the target task; If the target task is of the first timeliness and the communication state satisfies the target communication state, sending the target task to the ground auxiliary node for execution; Obtain the execution result returned by the ground auxiliary node after executing the target task.

5. A distributed collaborative task processing method according to claim 4, characterized in that: The method further comprises: Regularly obtain the communication connection between the nodes in the drone cluster and the master control node; If multiple nodes in the drone cluster detect that the communication connection with the master control node is interrupted, one of the nodes in the drone cluster other than the master control node is determined as a new master control node.

6. A distributed collaborative task processing device, used to execute a distributed collaborative task processing method according to any one of claims 1 to 5, characterized in that: The method is applied to a drone cluster composed of multiple drones, the drone cluster is used to perform forest fire fighting tasks and / or emergency rescue tasks, the drone cluster includes a master control node, the method is applied to the master control node, and the device includes: a first judgment module, a distributed processing module and a planning module, wherein: A first judgment module is used to judge whether the computing resources required for the target task to be executed by the drone cluster exceed the computing threshold when the target task to be executed by the drone cluster is obtained; wherein the target task is a task of performing path planning for the fire fighting task of the multiple drones in the forest fire fighting and / or emergency rescue task; if it does not exceed, the target task is assigned to the first node for processing; if it exceeds, distributed task processing is performed; A distributed processing module is used to determine whether each node in the drone cluster meets the execution of the multiple target subtasks based on the real-time occupancy rate of the computing resources of each node in the drone cluster; the target subtask is a task obtained by executing distributed task processing on the target task; if so, the multiple target subtasks are sent to the corresponding second nodes respectively; if not, the above distributed task processing steps are repeated until each node in the drone cluster meets the execution of the multiple target subtasks obtained by re-dividing; A planning module, configured to plan a flight path of the drone cluster in the target task based on the processing result of the first node or the distributed task processing result; The ground auxiliary module, each node in the drone cluster is also connected to the ground auxiliary node in communication, and the method further includes: obtaining the communication status between the ground auxiliary node and the timeliness of the target task; if the target task is of the first timeliness and the communication status meets the target communication status, sending the target task to the ground auxiliary node for execution; obtaining the execution result returned by the ground auxiliary node after executing the target task; The master control node determination module regularly obtains the communication connection between the nodes in the drone cluster and the master control node; if multiple nodes in the drone cluster detect that the communication connection with the master control node is interrupted, one of the nodes in the drone cluster is determined as the new master control node.

7. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the method, a distributed collaborative task processing method as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer program stored therein enables the processor to execute a distributed collaborative task processing method as described in any one of claims 1-5.

Images