Unmanned aerial vehicle cluster task decomposition sending method and system

Abstract

The invention relates to a divisional application of a patent number of 201810407291.4. The invention relates to the technical field of unmanned aerial vehicle control, and particularly provides an unmanned aerial vehicle cluster task decomposition sending method and system, and the method specifically comprises the steps that a master control unmanned aerial vehicle determines the number of sub-waypoint tasks needing to be decomposed according to a cluster waypoint task; and the master control unmanned aerial vehicle decomposes the cluster waypoint task into a plurality of sub-waypoint tasks according to an optimal decomposition principle and the number of the sub-waypoint tasks, and the master control unmanned aerial vehicle allocates and sends the sub-waypoint tasks to the corresponding slave unmanned aerial vehicles. According to the method, the main control unmanned aerial vehicle is responsible for decomposition, distribution and sending of cluster tasks, so that the unmanned aerial vehicle shares part of work of the central control system, the information processing pressure of the central control system is reduced, greater degree of freedom is given to the unmanned aerial vehicle cluster, management of mass data of the unmanned aerial vehicle is also facilitated, and the total calculated amount of formation task distribution is optimized.

Description

[0001] The patent application for this invention is a divisional application. The application number of the original application is 201810407291.4, the application date is April 28, 2018, and the application name is: UAV cluster scheduling method and system. Technical Field

[0002] The present invention relates to the field of unmanned aerial vehicle control technology, and in particular to a method and system for decomposing and sending tasks of an unmanned aerial vehicle cluster. Background Art

[0003] A drone swarm consists of multiple drones and can be used to perform tasks such as formation transportation and aerial performances. The currently commonly used method of decomposing and sending swarm tasks is that the swarm control system communicates with all drones, and the swarm control system is responsible for decomposing and distributing the tasks and sending them to each drone. Each drone is only responsible for receiving and executing the tasks. This method requires a high level of information processing capability of the swarm control system. When there are many drone swarms, the swarm control system is under great pressure for information processing and communication. Once the system crashes or information processing is not timely, the task may not be executed, or even a safety accident may occur. Summary of the invention

[0004] (I) Purpose of the invention

[0005] In order to solve at least one of the defects in the above-mentioned prior art solutions and reduce the information processing pressure of the cluster control system, the following technical solution is provided.

[0006] (II) Technical solution

[0007] As a first aspect of the present invention, the present invention provides a method for decomposing and sending tasks of a drone cluster, wherein the drone cluster includes at least one master drone and a plurality of slave drones that are communicatively connected to the master drone;

[0008] The method for decomposing and sending the UAV cluster task includes:

[0009] The master control drone determines the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint tasks;

[0010] The master control drone decomposes the cluster waypoint task into a plurality of sub-waypoint tasks according to the optimal decomposition principle and the number of the sub-waypoint tasks;

[0011] The master drone allocates and sends the sub-waypoint tasks to the corresponding slave drones; wherein,

[0012] The optimal decomposition principle includes one or more of the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate;

[0013] Among them, the drone cluster includes at least two master-controlled drones, one of the at least two master-controlled drones is a central master-controlled drone, and the other master-controlled drones in the same cluster are all collaborative master-controlled drones. The collaborative master-controlled drones are controlled by the central master-controlled drone and assist the central master-controlled drone in decomposing and sending cluster waypoint tasks.

[0014] The master control drone decomposes the cluster waypoint task into multiple sub-waypoint tasks according to the optimal decomposition principle and the number of sub-waypoint tasks. Specifically:

[0015] The central master control drone decomposes the cluster waypoint mission into multiple hierarchical tasks;

[0016] The central master drone sends part or all of the hierarchical tasks to each collaborative master drone for processing;

[0017] Each master drone processes hierarchical tasks according to the optimal decomposition principle;

[0018] The collaborative master control drone sends the hierarchical task processing result to the central master control drone;

[0019] The central master control drone further processes the processing results of each level of tasks to obtain the multiple sub-waypoint tasks.

[0020] The cluster waypoint mission includes information about the number of drones, and the master control drone determines the number of sub-waypoint missions that need to be decomposed according to the information about the number of drones.

[0021] The master control drone allocates and sends the sub-waypoint tasks to the corresponding drones as follows:

[0022] The central master drone divides all or part of the sub-waypoint tasks into a plurality of sub-waypoint task sets according to the number of master drones;

[0023] The central master control drone sends all or part of the sub-waypoint task sets to all or part of the collaborative master control drones respectively;

[0024] The master drone sends the sub-waypoint tasks contained in the sub-waypoint task set to the corresponding slave drones.

[0025] As a specific implementation of the above technical solution, the drone quantity information includes the number of the master drones and the number of the slave drones.

[0026] As a specific implementation of the above technical solution, all drones in the drone cluster have the same cluster identifier, which is different from the cluster identifier of any other drone cluster.

[0027] As a second aspect of the present invention, the present invention provides a drone cluster task decomposition and transmission system, wherein the drone cluster includes at least one master drone and multiple slave drones that are connected to the master drone in communication. The master drone always maintains a communication connection with the central control system responsible for controlling the drone cluster, and is responsible for communicating with the central control system at the task level. At the same time, the master drone also sends control instructions to the slave drones, and the slave drones are controlled by the master drone and complete a cluster waypoint task for the drone cluster together with the master drone.

[0028] The UAV cluster task decomposition and sending system includes:

[0029] A quantity determination module, used to enable the master control drone to determine the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task;

[0030] A task decomposition module, used for decomposing the cluster waypoint task into a plurality of sub-waypoint tasks according to the optimal decomposition principle and the number of the sub-waypoint tasks;

[0031] The task distribution module is used to distribute and send the sub-waypoint tasks to the corresponding slave drones;

[0032] The optimal decomposition principle includes the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate;

[0033] The task decomposition module includes a first decomposition submodule provided in the central master control UAV and a second decomposition submodule provided in the collaborative master control UAV;

[0034] The first decomposition submodule includes:

[0035] A first decomposition unit, used for decomposing the cluster waypoint task into multiple hierarchical tasks according to the hierarchical level;

[0036] The first allocation unit is used to send part or all of the hierarchical tasks to each collaborative master control drone for processing;

[0037] A first processing unit is used to enable the central master control drone to process the hierarchical tasks according to the optimal decomposition principle; and

[0038] A first integration unit is used to further process the processing result sent by the collaborative master control drone to obtain the multiple sub-waypoint tasks;

[0039] The second decomposition submodule includes:

[0040] A second processing unit is used to enable the collaborative master control drone to process the hierarchical tasks according to the optimal decomposition principle; and

[0041] The second reply unit is used to reply the hierarchical task processing result to the central master control drone.

[0042] The task distribution module includes: a first distribution submodule provided in the central master control drone and a second distribution submodule provided in the cooperative master control drone;

[0043] The first distribution submodule includes:

[0044] A first division unit is used to divide all or part of the sub-waypoint tasks into a plurality of sub-waypoint task sets according to the number of master control drones;

[0045] A first distribution unit is used to send all or part of the sub-waypoint task sets to all or part of the coordinated master control drones respectively; and

[0046] A first sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the central master drone is responsible for to the corresponding slave drones;

[0047] The second distribution submodule includes:

[0048] The second sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the collaborative master drone is responsible for to the corresponding slave drone.

[0049] The cluster waypoint mission includes information about the number of drones, and the number determination module determines the number of sub-waypoint missions that need to be decomposed according to the information about the number of drones.

[0050] Among them, the drone cluster includes at least two master-controlled drones, one of the at least two master-controlled drones is a central master-controlled drone, and the other master-controlled drones in the same cluster are all collaborative master-controlled drones. The collaborative master-controlled drones are controlled by the central master-controlled drone and assist the central master-controlled drone in decomposing and sending cluster waypoint tasks.

[0051] As a specific implementation of the above technical solution, the drone quantity information includes the number of the master drones and the number of the slave drones.

[0052] As a specific implementation of the above technical solution, each drone in the drone cluster includes:

[0053] The cluster identification module is used to store the cluster identification of the drone cluster to which the drone belongs, and the cluster identification is different from the cluster identification of any other drone cluster.

[0054] (III) Beneficial effects

[0055] 1. The master drone is responsible for the decomposition, allocation and sending of cluster tasks, which allows the drone to share part of the work of the central control system, reduces the information processing pressure of the central control system, gives the drone cluster greater freedom, facilitates the management of massive data of drones, and optimizes the total computing amount of formation task allocation;

[0056] 2. The central master drone decomposes the cluster waypoint tasks according to hierarchical tasks and uses the data processing capabilities of the collaborative master drone to help the central master drone process tasks synchronously, thereby speeding up the decomposition of cluster waypoint tasks into sub-waypoint tasks. At the same time, the collaborative master drone distributes each sub-waypoint task to each subordinate drone, thereby improving the drone cluster's processing speed for the received cluster waypoint tasks and reducing the drone cluster's response time. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain and illustrate the present invention, but should not be construed as limiting the protection scope of the present invention.

[0058] Figure 1 It is a flowchart of one embodiment of the method for decomposing and sending unmanned aerial vehicle cluster tasks provided by the present invention;

[0059] Figure 2 It is a flowchart of another embodiment of the method for decomposing and sending unmanned aerial vehicle cluster tasks provided by the present invention;

[0060] Figure 3 It is a structural block diagram of one embodiment of the UAV cluster task decomposition and sending system provided by the present invention;

[0061] Figure 4 It is a structural block diagram of another embodiment of the UAV cluster task decomposition and sending system provided by the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the implementation of the present invention clearer, the technical solution in the embodiment of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiment of the present invention.

[0063] It should be noted that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] The following is one embodiment of the method for decomposing and sending a drone cluster task provided by the present invention, which is the first embodiment. Through the method for decomposing and sending a drone cluster task, the cluster waypoint task sent from the central control system controlling the drone cluster is decomposed to obtain multiple sub-waypoint tasks, and the sub-waypoint tasks are sent to the drones, so that each drone performs its own sub-waypoint task, and finally realizes that the drones perform tasks such as formation flying, aerial performances, and even terrain reconnaissance in a cluster unit. Among them, a drone cluster includes at least one master drone and multiple slave drones that maintain communication connection with the master drone. The master drone always maintains a communication connection with the central control system responsible for controlling the drone cluster, and is responsible for communicating with the central control system at the task level. At the same time, the master drone also sends control instructions to the slave drones. The master drone has higher computing performance and data processing capabilities (for example, equipped with a more advanced master control chip). Therefore, the master drone can be regarded as the brain of the drone cluster, and the slave drones are similar to the limbs of the drone cluster. The slave drones are controlled by the master drone, and the slave drones and the master drone jointly complete a cluster waypoint task for the drone cluster.

[0065] As a specific implementation of the above technical solution, each drone cluster has its own cluster identification, and all drones in the same drone cluster have the same cluster identification, which is different from the cluster identification of any other drone cluster. In the drone communication process and other behavioral activities, when selecting a drone cluster or when drones in a cluster communicate with each other, the cluster identification can be used to identify whether the communication object is a member of the same cluster, so as to prevent interference with drones in other clusters and facilitate the central control system to control only a specific cluster.

[0066] Figure 1 It is a schematic diagram of the process of this embodiment, as shown in Figure 1 As shown, the UAV cluster task decomposition and sending method includes the following steps:

[0067] Step 100: The master drone determines the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task.

[0068] When dispatching drones in clusters, it is necessary to issue instructions to the drone cluster so that each drone in the cluster performs corresponding actions according to the instructions to achieve the purpose of controlling the drone cluster. The above instructions are cluster waypoint tasks. The cluster waypoint tasks are generated by the central control system responsible for controlling the drone cluster. After the operator of the central control system plans the tasks that the drone cluster needs to perform, the cluster waypoint tasks are generated in a way that the drone cluster can parse, and the cluster waypoint tasks are sent to the master drone of the cluster. After the master drone receives the cluster waypoint task sent by the central control system, it needs to decompose the cluster waypoint task into multiple sub-waypoint tasks for individual drones after determining the number of sub-waypoint tasks, so that the drone cluster can perform the tasks of the entire cluster as a single drone. For example, the cluster waypoint task is a small aerial performance, which requires the formation of numbers within 10 in the air. The master drone can determine how many slave drones are required to participate in the execution of the task based on the content of the task, and then determine how many sub-waypoint tasks the cluster waypoint task needs to be decomposed into.

[0069] As a specific implementation of the above technical solution, the cluster waypoint task includes information on the number of drones, and the master drone determines the number of sub-waypoint tasks that need to be decomposed based on the information on the number of drones. The number of sub-waypoint tasks can be determined by the master drone according to the task content, or the cluster waypoint task can directly include the requirements for the number of drones, and the master drone can determine how many slave drones are required to participate in the execution of the task according to the number requirements, and then determine how many sub-waypoint tasks the cluster waypoint task needs to be decomposed into. Therefore, it can be understood that before the master drone determines the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task, the master drone first obtains the information on the number of drones in the cluster waypoint task, and then the master drone determines the number of sub-waypoint tasks according to the information on the number of drones. Different tasks require different numbers of drones to be performed by the drone cluster. Large and complex tasks may require more than one master drone and a large number of slave drones, while medium and small tasks and simple tasks may only require one master drone and a small number of slave drones.

[0070] As a specific implementation of the above technical solution, the drone quantity information includes the number of master drones and the number of subordinate drones. Further, the requirement for the number of drones included in the cluster waypoint task can be specific to the number of master drones and the number of subordinate drones, so that the master drone can directly determine the number of sub-waypoint tasks that need to be decomposed.

[0071] It should be noted that if the cluster waypoint mission is to make the drone cluster perform multiple actions in an animation, the central control system will generate an animation script file, the master drone will analyze the script file, and then perform the subsequent decomposition, distribution, and sending to the slave drones. If the master drone is not capable enough, the central control system will process it to the extent that the master drone can analyze it before issuing the cluster waypoint mission.

[0072] Step 200, the master drone decomposes the cluster waypoint task into multiple sub-waypoint tasks according to the optimal decomposition principle and the number of sub-waypoint tasks. The optimal decomposition principle includes one or more of the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate.

[0073] Specifically, assuming that the number of sub-waypoint tasks to be decomposed is 20 drones, one of which is the master drone. The master drone decomposes the cluster waypoint task into 20 different sub-waypoint tasks through the built-in task decomposition program based on the three measurement factors of the minimum overall energy consumption of the cluster, the minimum loss cost of drones, and the lowest safety accident rate. Taking the aerial performance task as an example, assuming that the cluster waypoint task requires the drone cluster to pose the shape of the number 8 at time T, and then pose the shape of the number 1 at time T+20s, then the content of each sub-waypoint task includes moving to a certain position range at time T to pose the number 8, and then moving to another position range at time T+20s to pose the number 1. In the process of changing the action from the shape of the number 8 to the shape of the number 1, which position of the drone in the shape of the number 8 needs to move to which position in the shape of the number 1, there are many different solutions, and through the above decomposition optimal principle, a most preferred solution can be obtained, which makes the overall energy consumption of the cluster that performs the task the minimum, the loss cost of drones the minimum, and the safety accident rate the lowest. It is understandable that the above-mentioned optimal decomposition principle can be implemented by weighing the above-mentioned three measurement factors, such as assigning weights according to the importance of each measurement factor, and taking a weighted average of the three factors to determine how to decompose to achieve the best result, or using some other intelligent algorithms to calculate how to decompose to achieve the best task execution effect.

[0074] Advantageously, the task decomposition program built into the master drone processes cluster waypoint tasks based on a large amount of early manually planned data sets and multiple simulated flight data sets on computers (as training sets for machine learning, using technologies such as reinforcement learning, RNN, and intelligent algorithms of multi-agent systems), and uses the routines derived from machine learning to handle macro tasks, such as processing animation scripts to derive the motion trajectory and color changes of each point, and the corresponding flight routes and lighting change requirements of each drone, that is, the route planning files of each drone.

[0075] Step 300: The master drone allocates and sends sub-waypoint tasks to corresponding slave drones.

[0076] After the master drone decomposes the cluster waypoint task into multiple sub-waypoint tasks, the master drone also needs to assign each sub-waypoint task to the corresponding slave drones, which is equivalent to pairing each sub-waypoint task with each slave drone. Then, based on the pairing results, the master drone distributes each sub-waypoint task to other drones except the master drone.

[0077] The following is another embodiment of the method for decomposing and sending drone cluster tasks provided by the present invention, which is the second embodiment. When a drone cluster with higher data processing capabilities is required to perform a cluster waypoint task, a drone cluster with multiple master drones will be designated in the cluster waypoint task to be responsible for the execution of the task. At this time, there is more than one master drone in the drone cluster that performs the task. It can be understood that the more slave drones are included in the drone cluster, the higher the data processing capabilities required for the master drone, and the more likely it is that more than one master drone will be needed in the cluster. Figure 2 The flowchart of this embodiment is shown in FIG. Figure 2 As shown, in this embodiment, the drone cluster includes at least two master drones, one of which is a central master drone, and the other master drones in the same cluster are all cooperative master drones. The cooperative master drone is controlled by the central master drone and assists the central master drone in decomposing and sending cluster waypoint tasks, and assists the central master drone in data processing and other tasks. It should be noted that the central master drone receives the cluster waypoint tasks sent by the central control system, and the cooperative master drone is not responsible for communicating with the central control system at the task level. The other steps and implementation methods in this embodiment are the same as those in the first embodiment, and will not be repeated here.

[0078] When decomposing tasks, the central master drone can decompose all the cluster waypoint tasks by itself, but when the amount of data is large or the data processing speed needs to be increased, it is necessary to use the data processing capabilities of the collaborative master drone to decompose the tasks together with the collaborative master drone. Therefore, as a specific implementation of the above technical solution, the master drone in step 200 decomposes the cluster waypoint task into multiple sub-waypoint tasks according to the optimal decomposition principle and the number of sub-waypoint tasks, which specifically includes the following steps:

[0079] Step 210: The central master control drone decomposes the cluster waypoint mission into multiple level tasks.

[0080] After receiving the cluster waypoint task, if the central master drone wants to complete the task decomposition faster, or if the central master drone's own data processing capability cannot meet the huge amount of data contained in the cluster waypoint task, the central master drone will adopt a collaborative decomposition strategy and first decompose the cluster waypoint task into multiple hierarchical tasks. Taking aerial performances as an example, the cluster waypoint task can be divided into multiple hierarchical tasks according to the levels of text demonstration content, total path optimization, etc. For example, it can be divided into 3 hierarchical tasks, with each master drone responsible for 1 hierarchical task, or divided into 5 hierarchical tasks, with 1 master drone responsible for 1 hierarchical task and the other 2 master drones responsible for 2 hierarchical tasks each.

[0081] In step 220, the central master drone sends part or all of the hierarchical tasks to each collaborative master drone for processing.

[0082] Due to the differences in hierarchical tasks, the procedures for decomposing different hierarchical tasks are also different. According to the differences in the decomposition procedures built into each master drone, some hierarchical tasks can only be assigned to the master drone with the corresponding specific task decomposition procedure for decomposition, while other master drones cannot decompose the hierarchical tasks. However, if the master drone can decompose various types of hierarchical tasks, there is no need to assign the hierarchical tasks specifically to a specific master drone for task decomposition. At this time, hierarchical tasks can be freely assigned according to the situation. In addition, in order to speed up the task decomposition, the central master drone will be responsible for part of the hierarchical tasks like the collaborative master drone, so only part of the hierarchical tasks will be sent to the collaborative master drone. If there are other special circumstances, the central master drone is not responsible for the decomposition of any hierarchical tasks. At this time, all hierarchical tasks need to be sent to the collaborative master drone for processing. In general, the central master drone will participate in the decomposition of hierarchical tasks.

[0083] In step 230, each master drone processes the hierarchical task according to the optimal decomposition principle. The cooperative master drone assigned with the hierarchical task processes the received hierarchical task, and the central master drone also processes the hierarchical task assigned to it.

[0084] Step 240: The cooperative master drone sends the hierarchical task processing result to the central master drone. After the cooperative master drone completes the hierarchical task, it feeds back the processing result to the central master drone, which then processes the result uniformly.

[0085] Step 250, the central master drone further processes the processing results of each level task to obtain multiple sub-waypoint tasks. The central master drone collects all the processing results of the level tasks and integrates them to obtain all the sub-waypoint tasks after the cluster waypoint task is decomposed. Each drone is responsible for one sub-waypoint task. Through the assistance of the collaborative master drone, the data processing speed during task decomposition is accelerated, so that the cluster waypoint task can be decomposed into sub-waypoint tasks as soon as possible.

[0086] In a drone cluster that includes multiple master drones, the number of slave drones is likely to be large, so the number of corresponding sub-waypoint tasks is also large, and there is only one central master drone, so the central master drone is likely to need to divide the sub-waypoint tasks into multiple batches and send them to each drone in batches, which may cause a certain delay in the execution of the task. In order to speed up the distribution of sub-waypoint tasks, it is necessary to use a collaborative master drone to distribute sub-waypoint tasks together. Therefore, as a specific implementation method of the above technical solution, the master drone in step 300 distributes and sends the sub-waypoint tasks to the corresponding drones, which specifically includes the following steps:

[0087] Step 310: The central master drone divides all or part of the sub-waypoint tasks into a plurality of sub-waypoint task sets according to the number of master drones.

[0088] After obtaining all the sub-waypoint tasks, each sub-waypoint task needs to be sent to each drone. Specifically, for example, the drone cluster includes 1 central master drone, 2 collaborative master drones, 189 slave drones, and a total of 192 drones. In the process of distributing sub-waypoint tasks, the central master drone can first divide all 192 sub-waypoint tasks into multiple sub-waypoint task sets, and the number of sub-waypoint tasks contained in each sub-waypoint task set can be the same or different. At this time, the three master drones are responsible for sending sub-waypoint tasks; the central master drone can also first divide 191 sub-waypoint tasks into multiple sub-waypoint task sets, and the number of sub-waypoint tasks contained in each sub-waypoint task set can be the same or different. At this time, only two collaborative master drones are responsible for sending sub-waypoint tasks. After the central master drone leaves the remaining 1 sub-waypoint task belonging to itself, it does not participate in sending sub-waypoint tasks of other drones. The central master drone will not participate in sending sub-waypoint tasks under special circumstances. Under normal circumstances, in order to speed up the task sending speed, all master drones, including the central master drone, will participate in the process of sending sub-waypoint tasks. It should be noted that when dividing the sub-waypoint task set, the sub-waypoint task of the collaborative master drone is preferentially divided into the sub-waypoint task set that the collaborative master drone is responsible for, so that the collaborative master drone can save sending its own sub-waypoint task when sending the sub-waypoint task.

[0089] In step 320, the central master drone sends all or part of the sub-waypoint mission sets to all or part of the collaborative master drones.

[0090] When the central master drone participates in sending sub-waypoint tasks, the central master drone divides 192 sub-waypoint tasks into multiple sub-waypoint task sets, and sends part of the sub-waypoint task sets to two collaborative master drones, and the remaining unsent sub-waypoint task sets are used as their own sending tasks; when the central master drone does not participate in sending sub-waypoint tasks, the central master drone sends all sub-waypoint task sets to the collaborative master drone. It can be understood that the allocation between the sub-waypoint task set and the master drone can be allocated according to the physical distance. It should be noted that there are special cases where the collaborative master drone does not participate in sending sub-waypoint tasks. Therefore, when there are collaborative master drones that do not participate in sending sub-waypoint tasks, the central master drone sends the sub-waypoint task sets to some of the collaborative master drones respectively. When all collaborative master drones participate in sending sub-waypoint tasks, the central master drone sends the sub-waypoint task sets to all collaborative master drones respectively.

[0091] In step 330, the master drone sends the sub-waypoint tasks included in the sub-waypoint task set to the corresponding slave drones. The master drones participating in the sending of sub-waypoint tasks, whether they are central master drones or collaborative master drones, send the sub-waypoint tasks included in the sub-waypoint task set that they are responsible for to the corresponding slave drones. After all sub-waypoint tasks are sent to each drone, each drone accepts its own sub-waypoint task and executes the task content.

[0092] The cluster waypoint tasks are decomposed into hierarchical tasks and the data processing capabilities of the collaborative master drone are used to help the central master drone to synchronously process the tasks. The cluster waypoint tasks are decomposed into sub-waypoint tasks, and the collaborative master drone is used to distribute each sub-waypoint task to each slave drone. This improves the processing speed of the drone cluster for the received cluster waypoint tasks and reduces the reaction time of the drone cluster.

[0093] The following is one embodiment of the unmanned aerial vehicle cluster task decomposition and transmission system provided by the present invention, which is the third embodiment. Through the unmanned aerial vehicle cluster task decomposition and transmission system, the cluster waypoint task sent from the central control system controlling the unmanned aerial vehicle cluster is decomposed to obtain multiple sub-waypoint tasks, and the sub-waypoint tasks are sent to the unmanned aerial vehicles, so that each unmanned aerial vehicle performs its own sub-waypoint task, and finally realizes the unmanned aerial vehicle to perform formation flying, aerial performances and even terrain reconnaissance tasks as a cluster unit. Among them, a unmanned aerial vehicle cluster includes at least one master control unmanned aerial vehicle and multiple slave unmanned aerial vehicles that maintain communication connection with the master control unmanned aerial vehicle. The master control unmanned aerial vehicle always maintains communication connection with the central control system responsible for controlling the unmanned aerial vehicle cluster, and is responsible for communicating with the central control system at the task level. At the same time, the master control unmanned aerial vehicle also issues control instructions to the slave unmanned aerial vehicle. The master control unmanned aerial vehicle has higher computing performance and data processing capabilities (for example, equipped with a more advanced master control chip). Therefore, the master control unmanned aerial vehicle can be regarded as the brain of the unmanned aerial vehicle cluster, and the slave unmanned aerial vehicle is similar to the limbs of the unmanned aerial vehicle cluster. The slave unmanned aerial vehicle is controlled by the master control unmanned aerial vehicle, and the slave unmanned aerial vehicle and the master control unmanned aerial vehicle jointly complete a cluster waypoint task for the unmanned aerial vehicle cluster.

[0094] As a specific implementation of the above technical solution, each drone in the drone cluster includes a cluster identification module. The cluster identification module is used to store the cluster identification of the drone cluster to which the drone belongs, and the cluster identification is different from the cluster identification of any other drone cluster. In the drone communication process and other behavioral activities, when selecting a drone cluster or when drones in a cluster communicate with each other, the cluster identification can be used to identify whether the communication object is a member of the same cluster, so as to prevent interference with drones in other clusters and facilitate the central control system to control only a specific cluster.

[0095] Figure 3 is a structural block diagram of this embodiment, such as Figure 3 As shown, the UAV cluster task decomposition and sending system includes a quantity determination module, a task decomposition module and a task distribution module.

[0096] The quantity determination module is used to determine the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task.

[0097] When dispatching drones in clusters, it is necessary to issue instructions to the drone cluster so that each drone in the cluster performs corresponding actions according to the instructions to achieve the purpose of controlling the drone cluster. The above instructions are cluster waypoint tasks. The cluster waypoint tasks are generated by the central control system responsible for controlling the drone cluster. After the operator of the central control system plans the tasks that the drone cluster needs to perform, the cluster waypoint tasks are generated in a way that the drone cluster can parse, and the cluster waypoint tasks are sent to the master drone of the cluster. After the master drone receives the cluster waypoint task sent by the central control system, it needs to determine the number of sub-waypoint tasks through the quantity determination module, and then decompose the cluster waypoint task into multiple sub-waypoint tasks for individual drones, so that the drone cluster can perform the tasks of the entire cluster as a single drone. For example, the cluster waypoint task is a small aerial performance, which requires the formation of numbers within 10 in the air. The master drone can determine how many slave drones are required to participate in the execution of the task based on the content of the task, and then determine how many sub-waypoint tasks the cluster waypoint task needs to be decomposed into.

[0098] As a specific implementation of the above technical solution, the cluster waypoint task includes the number of drones information, and the number determination module determines the number of sub-waypoint tasks that need to be decomposed according to the number of drones information. The number of sub-waypoint tasks can be determined by the master drone according to the task content, or the cluster waypoint task can directly include the requirements for the number of drones, and the master drone can determine how many slave drones are required to participate in the execution of the task according to the number requirements, and then determine how many sub-waypoint tasks the cluster waypoint task needs to be decomposed into. Therefore, it can be understood that before the master drone determines the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task, the master drone first obtains the number of drones in the cluster waypoint task, and then the master drone determines the number of sub-waypoint tasks according to the number of drones information. According to the different tasks that the drone cluster needs to perform, the number of drones required is also different. The execution of large and complex tasks may require more than one master drone and a large number of slave drones, while medium and small tasks and simple tasks may only require one master drone and a small number of slave drones.

[0099] As a specific implementation of the above technical solution, the drone quantity information includes the number of master drones and the number of subordinate drones. Further, the requirement for the number of drones included in the cluster waypoint task can be specific to the number of master drones and the number of subordinate drones, so that the master drone can directly determine the number of sub-waypoint tasks that need to be decomposed.

[0100] It should be noted that if the cluster waypoint mission is to make the drone cluster perform multiple actions in an animation, the central control system will generate an animation script file, the master drone will analyze the script file, and then perform the subsequent decomposition, distribution, and sending to the slave drones. If the master drone is not capable enough, the central control system will process it to the extent that the master drone can analyze it before issuing the cluster waypoint mission.

[0101] The task decomposition module is used to decompose the cluster waypoint task into multiple sub-waypoint tasks according to the optimal decomposition principle and the number of sub-waypoint tasks. Among them, the optimal decomposition principle includes the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate.

[0102] Specifically, assuming that the number of sub-waypoint tasks to be decomposed is 20 drones, one of which is the master drone. The master drone decomposes the cluster waypoint task into 20 different sub-waypoint tasks through the built-in task decomposition program of the task decomposition module based on the three measurement factors of the minimum overall energy consumption of the cluster, the minimum loss cost of drones, and the lowest safety accident rate. Taking the aerial performance task as an example, assuming that the cluster waypoint task requires the drone cluster to pose the shape of the number 8 at time T, and then pose the shape of the number 1 at time T+20s, then the content of each sub-waypoint task includes moving to a certain position range at time T to pose the number 8, and then moving to another position range at time T+20s to pose the number 1. In the process of changing the action from the shape of the number 8 to the shape of the number 1, which position of the drone in the shape of the number 8 needs to move to which position in the shape of the number 1, there are many different solutions, and through the above decomposition optimal principle, a most preferred solution can be obtained, which makes the overall energy consumption of the cluster that performs the task the minimum, the loss cost of drones the minimum, and the safety accident rate the lowest. It is understandable that the above-mentioned optimal decomposition principle can be implemented by weighing the above-mentioned three measurement factors, such as assigning weights according to the importance of each measurement factor, and taking a weighted average of the three factors to determine how to decompose to achieve the best result, or using some other intelligent algorithms to calculate how to decompose to achieve the best task execution effect.

[0103] Advantageously, the task decomposition program built into the master drone processes cluster waypoint tasks based on a large amount of early manually planned data sets and multiple simulated flight data sets on computers (as training sets for machine learning, using technologies such as reinforcement learning, RNN, and intelligent algorithms of multi-agent systems), and uses the routines derived from machine learning to handle macro tasks, such as processing animation scripts to derive the motion trajectory and color changes of each point, and the corresponding flight routes and lighting change requirements of each drone, that is, the route planning files of each drone.

[0104] The task distribution module is used to assign and send sub-waypoint tasks to the corresponding slave drones.

[0105] After the master drone decomposes the cluster waypoint task into multiple sub-waypoint tasks, the master drone also needs to assign each sub-waypoint task to the corresponding slave drones through the task distribution module, which is equivalent to pairing each sub-waypoint task with each slave drone. Then, based on the pairing results, the master drone distributes each sub-waypoint task to other drones except the master drone.

[0106] The following is another embodiment of the drone cluster task decomposition and transmission system provided by the present invention, which is the fourth embodiment. When a drone cluster with higher data processing capabilities is required to perform a cluster waypoint task, a drone cluster with multiple master drones will be designated in the cluster waypoint task to be responsible for the execution of the task. At this time, there is more than one master drone in the drone cluster that performs the task. It can be understood that the more slave drones are included in the drone cluster, the higher the data processing capabilities required for the master drone, and the more likely it is that more than one master drone will be needed in the cluster. Figure 4 The structural block diagram of this embodiment is shown as follows. Figure 2 As shown, in this embodiment, the drone cluster includes at least two master drones, one of which is a central master drone, and the other master drones in the same cluster are all collaborative master drones. The collaborative master drone is controlled by the central master drone and assists the central master drone in decomposing and sending cluster waypoint tasks, and assists the central master drone in data processing and other tasks. It should be noted that the central master drone receives the cluster waypoint tasks sent by the central control system, and the collaborative master drone is not responsible for communicating with the central control system at the task level. The other steps and implementation methods in this embodiment are the same as those in the third embodiment, and will not be repeated here.

[0107] When decomposing tasks, the central master drone can decompose all the cluster waypoint tasks by itself, but when the amount of data is large or the data processing speed needs to be increased, it is necessary to use the data processing capabilities of the cooperative master drone and decompose the tasks together with the cooperative master drone. Therefore, as a specific implementation of the above technical solution, the task decomposition module includes a first decomposition submodule set in the central master drone and a second decomposition submodule set in the cooperative master drone.

[0108] The first decomposition submodule includes a first decomposition unit, a first allocation unit, a first processing unit and a first integration unit. The second decomposition submodule includes a second processing unit and a second recovery unit.

[0109] The first decomposition unit is used to decompose the cluster waypoint task into multiple hierarchical tasks. After the central master drone receives the cluster waypoint task, if it wants to complete the task decomposition faster, or the central master drone's own data processing capability cannot meet the huge amount of data contained in the cluster waypoint task, the central master drone will adopt a collaborative decomposition strategy, and first decompose the cluster waypoint task into multiple hierarchical tasks through the first decomposition unit. Taking aerial performances as an example, the cluster waypoint task can be divided into multiple hierarchical tasks according to the levels of text demonstration content, total path optimization, etc. For example, it can be divided into 3 hierarchical tasks, each master drone is responsible for 1 hierarchical task, or divided into 5 hierarchical tasks, of which 1 master drone is responsible for 1 hierarchical task, and the other 2 master drones are responsible for 2 hierarchical tasks each.

[0110] The first allocation unit is used to send part or all of the hierarchical tasks to each collaborative master drone for processing. Due to the different hierarchical tasks, the procedures for decomposing different hierarchical tasks are also different. According to the different decomposition programs built into each master drone, some hierarchical tasks can only be assigned to the master drone with the corresponding specific task decomposition program for decomposition, while other master drones cannot decompose the hierarchical tasks. However, if the master drone can decompose various types of hierarchical tasks, there is no need to specifically assign the hierarchical tasks to a specific master drone for task decomposition. At this time, the hierarchical tasks can be freely allocated according to the situation. In addition, in order to speed up the task decomposition, the central master drone will be responsible for part of the hierarchical tasks like the collaborative master drone, so only part of the hierarchical tasks will be sent to the collaborative master drone through the first allocation unit. If there are other special circumstances, the central master drone is not responsible for the decomposition of any hierarchical tasks. At this time, it is necessary to send all hierarchical tasks to the collaborative master drone for processing through the first allocation unit. In general, the central master drone will participate in the decomposition of hierarchical tasks.

[0111] The first processing unit is used to enable the central master drone to process the hierarchical tasks according to the optimal decomposition principle. The second processing unit is used to enable the cooperative master drone to process the hierarchical tasks according to the optimal decomposition principle. The cooperative master drone assigned with the hierarchical task processes the received hierarchical task through the second processing unit, and the central master drone also processes the hierarchical task assigned to it through the first processing unit.

[0112] The second reply unit is used to reply the hierarchical task processing result to the central master drone. After the collaborative master drone completes the hierarchical task, it feeds back the processing result to the central master drone through the second reply unit, and the central master drone processes it uniformly.

[0113] The first integration unit is used to further process the processing results sent by the collaborative master drone to obtain multiple sub-waypoint tasks. The central master drone collects all the hierarchical task processing results and integrates them through the first integration unit to obtain all the sub-waypoint tasks after the cluster waypoint task is decomposed. Each drone is responsible for 1 sub-waypoint task. With the assistance of the collaborative master drone, the data processing speed during task decomposition is accelerated, so that the cluster waypoint task can be decomposed into sub-waypoint tasks as soon as possible.

[0114] After obtaining all the sub-waypoint tasks, each sub-waypoint task needs to be sent to each drone. In a drone cluster that includes multiple master drones, the number of slave drones is likely to be large, so the corresponding number of sub-waypoint tasks is also large, and there is only one central master drone, so the central master drone is likely to need to divide the sub-waypoint tasks into multiple batches and send them to each drone in batches, which may cause a certain delay in the execution of the task. In order to speed up the distribution of sub-waypoint tasks, it is necessary to use a collaborative master drone to distribute sub-waypoint tasks together. Therefore, as a specific implementation method of the above technical solution, the task distribution module includes a first distribution sub-module and a second distribution sub-module, the first distribution sub-module is set in the central master drone, and the second distribution sub-module is set inside the collaborative master drone.

[0115] The first distribution submodule includes a first dividing unit, a first distribution unit and a first sending unit. The second distribution submodule includes a second sending unit.

[0116] The first division unit is used to divide all or part of the sub-waypoint tasks into multiple sub-waypoint task sets according to the number of master control drones. Specifically, for example, the drone cluster includes 1 central master control drone, 2 collaborative master control drones, 189 slave drones, a total of 192 drones. In the process of distributing sub-waypoint tasks, the central master control drone can first divide all 192 sub-waypoint tasks into multiple sub-waypoint task sets through the first division unit. The number of sub-waypoint tasks contained in each sub-waypoint task set can be the same or different. At this time, the three master control drones are responsible for sending sub-waypoint tasks; the central master control drone can also first divide 191 sub-waypoint tasks into multiple sub-waypoint task sets through the first division unit. It is divided into multiple sub-waypoint task sets. The number of sub-waypoint tasks contained in each sub-waypoint task set can be the same or different. At this time, only two collaborative master drones are responsible for sending sub-waypoint tasks. After the central master drone leaves the remaining 1 sub-waypoint task belonging to itself, it does not participate in sending sub-waypoint tasks of other drones. The central master drone will not participate in sending sub-waypoint tasks under special circumstances. Under normal circumstances, in order to speed up the task sending speed, all master drones, including the central master drone, will participate in the process of sending sub-waypoint tasks. It should be noted that when dividing the sub-waypoint task set, the sub-waypoint tasks of the collaborative master drone are preferentially divided into the sub-waypoint task set that the collaborative master drone is responsible for, so that the collaborative master drone can save sending its own sub-waypoint tasks when sending sub-waypoint tasks.

[0117] The first distribution unit is used to send all or part of the sub-waypoint task sets to all or part of the collaborative master drones. When the central master drone participates in the sending of sub-waypoint tasks, the central master drone divides the 192 sub-waypoint tasks into multiple sub-waypoint task sets, and then sends part of the sub-waypoint task sets to the two collaborative master drones through the first distribution unit, and the remaining unsent sub-waypoint task sets are used as their own sending tasks; when the central master drone does not participate in the sending of sub-waypoint tasks, the central master drone sends all sub-waypoint task sets to the collaborative master drone through the first distribution unit. It can be understood that the distribution between the sub-waypoint task sets and the master drone can be allocated according to the physical distance. It should be noted that there are special cases where the collaborative master-control drone does not participate in sending sub-waypoint tasks. Therefore, when a collaborative master-control drone does not participate in sending sub-waypoint tasks, the central master-control drone will send the sub-waypoint task set to some collaborative master-control drones respectively. When all collaborative master-control drones participate in sending sub-waypoint tasks, the central master-control drone will send the sub-waypoint task set to all collaborative master-control drones respectively.

[0118] The first sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the central master drone is responsible for to the corresponding slave drone. The second sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the collaborative master drone is responsible for to the corresponding slave drone. The master drone participating in the sending of sub-waypoint tasks, whether it is the central master drone or the collaborative master drone, sends the sub-waypoint tasks included in the part of the sub-waypoint task set that it is responsible for to the corresponding slave drone through the first sending unit or the second sending unit. After all sub-waypoint tasks are sent to each drone, each drone accepts its own sub-waypoint task and executes the task content.

[0119] The cluster waypoint tasks are decomposed into hierarchical tasks and the data processing capabilities of the collaborative master drone are used to help the central master drone to synchronously process the tasks. The cluster waypoint tasks are decomposed into sub-waypoint tasks, and the collaborative master drone is used to distribute each sub-waypoint task to each slave drone. This improves the processing speed of the drone cluster for the received cluster waypoint tasks and reduces the reaction time of the drone cluster.

[0120] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for decomposing and sending UAV cluster tasks. It is characterized in that The drone cluster includes at least one master drone and a plurality of slave drones that are communicatively connected to the master drone; The method for decomposing and sending the UAV cluster task includes: The master control drone determines the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint tasks; The master control drone decomposes the cluster waypoint task into a plurality of sub-waypoint tasks according to the optimal decomposition principle and the number of the sub-waypoint tasks; The master drone allocates and sends the sub-waypoint tasks to the corresponding slave drones; wherein, The optimal decomposition principle includes one or more of the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate; The drone cluster includes at least two master drones, one of which is a central master drone, and the other master drones in the same cluster are all cooperative master drones, which are controlled by the central master drone and assist the central master drone in decomposing and sending the cluster waypoint tasks; The master control drone decomposes the cluster waypoint task into multiple sub-waypoint tasks according to the optimal decomposition principle and the number of sub-waypoint tasks. Specifically: The central master control drone decomposes the cluster waypoint mission into multiple hierarchical tasks; The central master drone sends part or all of the hierarchical tasks to each collaborative master drone for processing; Each master drone processes hierarchical tasks according to the optimal decomposition principle; The collaborative master control drone sends the hierarchical task processing result to the central master control drone; The central master control drone further processes the processing results of each level of tasks to obtain the multiple sub-waypoint tasks; The cluster waypoint task includes the number of drones, and the master drone determines the number of sub-waypoint tasks to be decomposed according to the number of drones; the master drone allocates and sends the sub-waypoint tasks to the corresponding drones as follows: The central master drone divides all or part of the sub-waypoint tasks into a plurality of sub-waypoint task sets according to the number of master drones; The central master control drone sends all or part of the sub-waypoint task sets to all or part of the collaborative master control drones respectively; The master drone sends the sub-waypoint tasks contained in the sub-waypoint task set to the corresponding slave drones.

2. The method for decomposing and sending UAV cluster tasks according to claim 1, It is characterized in that The drone quantity information includes the number of the master drones and the number of the slave drones.

3. The method for decomposing and sending UAV cluster tasks according to claim 1, It is characterized in that All drones in the drone cluster have the same cluster identifier, which is different from the cluster identifier of any other drone cluster.

4. A UAV swarm task decomposition and delivery system, It is characterized in that The drone cluster includes at least one master drone and a plurality of slave drones that are communicatively connected to the master drone; The UAV swarm task decomposition and sending system includes: A quantity determination module, used to enable the master control drone to determine the number of sub-waypoint tasks that need to be decomposed according to the cluster waypoint task; A task decomposition module, used for decomposing the cluster waypoint task into a plurality of sub-waypoint tasks according to the optimal decomposition principle and the number of the sub-waypoint tasks; The task distribution module is used to distribute and send the sub-waypoint tasks to the corresponding slave drones; The optimal decomposition principle includes the following measurement factors: minimum overall energy consumption of the cluster, minimum loss cost of drones, and minimum safety accident rate; The task decomposition module includes a first decomposition submodule provided in the central master control UAV and a second decomposition submodule provided in the collaborative master control UAV; The first decomposition submodule includes: A first decomposition unit, used for decomposing the cluster waypoint task into multiple hierarchical tasks according to the hierarchical level; The first allocation unit is used to send part or all of the hierarchical tasks to each collaborative master control drone for processing; A first processing unit is used to enable the central master control drone to process the hierarchical tasks according to the optimal decomposition principle; and A first integration unit is used to further process the processing result sent by the collaborative master control drone to obtain the multiple sub-waypoint tasks; The second decomposition submodule includes: A second processing unit is used to enable the collaborative master control drone to process the hierarchical tasks according to the optimal decomposition principle; and A second reply unit, used to reply the hierarchical task processing result to the central master control drone; The task distribution module includes: a first distribution submodule provided in the central master control drone and a second distribution submodule provided in the cooperative master control drone; The first distribution submodule includes: A first division unit is used to divide all or part of the sub-waypoint tasks into a plurality of sub-waypoint task sets according to the number of master control drones; A first distribution unit is used to send all or part of the sub-waypoint task sets to all or part of the coordinated master control drones respectively; and A first sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the central master drone is responsible for to the corresponding slave drones; The second distribution submodule includes: A second sending unit is used to send the sub-waypoint tasks included in the sub-waypoint task set that the collaborative master control drone is responsible for to the corresponding slave drone; Wherein, the cluster waypoint task includes the number information of drones, and the number determination module determines the number of sub-waypoint tasks to be decomposed according to the number information of drones; Among them, the drone cluster includes at least two master-controlled drones, one of the at least two master-controlled drones is a central master-controlled drone, and the other master-controlled drones in the same cluster are all collaborative master-controlled drones. The collaborative master-controlled drones are controlled by the central master-controlled drone and assist the central master-controlled drone in decomposing and sending cluster waypoint tasks.

5. The UAV cluster task decomposition and transmission system according to claim 4, It is characterized in that The drone quantity information includes the number of the master drones and the number of the slave drones.

6. The UAV cluster task decomposition and transmission system according to claim 4, It is characterized in that Each drone in the drone cluster includes: The cluster identification module is used to store the cluster identification of the drone cluster to which the drone belongs, and the cluster identification is different from the cluster identification of any other drone cluster.

Images