Dual planning method for swarm UAV missions

Abstract

The present invention relates to the technical field of unmanned aerial vehicles, and specifically to a dual planning method for cluster unmanned aerial vehicle tasks. The method comprises constructing a hierarchical task planning architecture including a global planning subsystem and a local planning subsystem, so that the global planning subsystem performs task sub-area division and task allocation; based on the local planning subsystem, the task is refined and the path is planned for the unmanned aerial vehicles in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; based on the feedback information, the task planning strategy of the cluster unmanned aerial vehicles is adjusted again through the global planning subsystem, and the hierarchical task planning architecture is adopted to get rid of the excessive dependence on a single central control node, and a complex task can be decomposed into a plurality of relatively independent sub-tasks, thereby reducing the complexity of task planning, reducing the task calculation amount and planning time, ensuring that the resources of each unmanned aerial vehicle are fully utilized, and improving the resource utilization rate, information processing capability and decision-making scientificity of the cluster unmanned aerial vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a dual planning method for cluster unmanned aerial vehicle tasks. Background Art

[0002] With the advantage of multi-machine collaborative operation, swarm drones have shown great potential in complex mission scenarios such as emergency rescue, geographic surveying and mapping, agricultural plant protection, and logistics distribution. The core of swarm drone mission planning and allocation is to reasonably plan the flight path and mission sequence of each drone according to the mission requirements and the performance characteristics of swarm drones, so as to maximize the overall mission execution efficiency and optimize the allocation of resources. At present, the more common swarm drone mission planning is the centralized planning method, that is, a central control node is responsible for collecting the status information of all drones in the cluster and the location distribution and mission type of the mission target, and uses the preset optimization algorithm to comprehensively process this information to determine the task allocation plan and flight path planning for each drone.

[0003] However, the centralized planning method for swarm drones is heavily dependent on the computing power and reliability of the central control node. Once the central control node fails, the entire swarm drone will be paralyzed and unable to continue to perform tasks. In addition, as the scale of swarm drones continues to expand, the amount of data that the central control node needs to process increases exponentially, which will lead to a significant increase in the calculation time of task planning and task response delays, making it difficult to meet some task scenarios with high real-time requirements, and reducing the task planning efficiency of swarm drones. Summary of the invention

[0004] In view of the defects of the prior art, the present invention provides a dual planning method for cluster UAV tasks, constructs a hierarchical task planning architecture including a global planning subsystem and a local planning subsystem, so that the global planning subsystem executes task sub-area division and task allocation; based on the local planning subsystem, the task is refined and the path planning is performed for the UAVs in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; based on the feedback information, the task planning strategy of the cluster UAVs is adjusted again through the global planning subsystem, and the hierarchical task planning architecture is adopted to get rid of the excessive dependence on a single central control node, and reduce the risk of the entire cluster being paralyzed due to the failure of a single central control node. Through the collaborative working mechanism of the global planning subsystem and the local planning subsystem, the cluster UAVs are adjusted in real time and dynamically, and flexibly adapt to changing task scenarios. Complex tasks can be decomposed into multiple relatively independent sub-tasks, reducing the complexity of task planning, reducing the task calculation amount and planning time, ensuring that the resources of each UAV are fully utilized, avoiding idle or excessive resource consumption, and improving the resource utilization rate, information processing capability and decision-making scientificity of the cluster UAVs.

[0005] The present invention provides a dual planning method for swarm UAV tasks, comprising the following steps:

[0006] Constructing a hierarchical task planning architecture, the architecture comprising a global planning subsystem and a local planning subsystem; wherein the global planning subsystem performs task sub-area division and task allocation;

[0007] Based on the local planning subsystem, the task refinement and path planning are performed for the UAVs in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem;

[0008] Based on the feedback information of the local planning subsystem, the mission planning strategy of the swarm UAVs is adjusted again through the global planning subsystem.

[0009] In one embodiment disclosed in the present application, a hierarchical task planning architecture is constructed, including:

[0010] Acquire hardware feature information and mission area feature information of all drones in the cluster of drones; wherein the hardware feature information includes wireless communication performance feature information of the drones; and the mission area feature information includes physical space feature information of the mission area;

[0011] Based on the wireless communication performance characteristic information and the physical space characteristic information of the task area, a global planning subsystem and a local planning subsystem of a hierarchical task planning architecture are constructed; wherein the planning priority of the global planning subsystem is higher than the planning priority of the local planning subsystem.

[0012] In one embodiment disclosed in the present application, based on the wireless communication performance characteristic information and the physical space characteristic information of the task area, a global planning subsystem and a local planning subsystem of a hierarchical task planning architecture are constructed, including:

[0013] Determine the wireless communication coverage area of ​​the task area based on the size of the task area space included in the physical space feature information of the task area; and construct a global planning subsystem of a hierarchical task planning architecture based on the wireless communication coverage area.

[0014] Based on the wireless communication bandwidth of the UAV included in the wireless communication performance characteristic information, a local planning subsystem of a hierarchical mission planning architecture is constructed.

[0015] In one embodiment disclosed in the present application, the global planning subsystem performs task sub-area division and task allocation, including:

[0016] The global planning subsystem performs task sub-area division and task allocation based on external task instructions and the initial state information of all UAVs in the cluster.

[0017] In one embodiment disclosed in the present application, the global planning subsystem performs task sub-area division and task allocation based on external task instructions and initial state information of all drones in the cluster, including:

[0018] The global planning subsystem parses and processes the external task instructions to obtain the task type and the estimated completion time of the task that the cluster drones need to perform;

[0019] The global planning subsystem collects the initial position, maximum range, maximum load, and flight speed of all drones in the cluster as the initial state information;

[0020] The global planning subsystem divides the task area into several task sub-areas based on the task type and estimated completion time of the task that the cluster drones need to perform, as well as the initial state information, and determines the task volume of each task sub-area and the performance requirement range of the required drones, thereby assigning tasks to all drones in the cluster.

[0021] In one embodiment disclosed in the present application, based on the local planning subsystem, task refinement and path planning are performed on the UAV in each task sub-area divided by the global planning subsystem, including:

[0022] Based on the local planning subsystem, collect the task execution status information and location information of all UAVs in each task sub-area;

[0023] Based on the task execution status information and the location information, task refinement and path planning are performed for all UAVs in each task sub-area.

[0024] In one embodiment disclosed in the present application, based on the task execution status information and the location information, task refinement and path planning are performed for all drones in each task sub-area, including:

[0025] Based on the mission execution status information and the location information, identifying a drone in an abnormal state within the mission sub-area;

[0026] The task execution status information of other drones around the drone in the abnormal state in the task sub-area is obtained, and based on the task execution status information of other drones around, other drones around are instructed to replace the task originally executed by the drone in the abnormal state.

[0027] In one embodiment disclosed in the present application, the local planning subsystem feeds back information to the global planning subsystem, including:

[0028] The local planning subsystem feeds back the task execution status information and self-resource status information of all drones in the task sub-area to the global planning subsystem; wherein the task execution status information includes the flight path and the amount of completed tasks during the drone's task execution; and the self-resource status information includes the remaining power of all drones in the task sub-area and the number of drones in normal state.

[0029] In one embodiment disclosed in the present application, an information fusion and arbitration subsystem is also provided;

[0030] The information fusion and arbitration subsystem screens the feedback information and the information collected by the global planning subsystem, and removes redundant information parts and conflicting information parts between the feedback information and the collected information.

[0031] In one embodiment disclosed in the present application, based on the feedback information of the local planning subsystem, the mission planning strategy of the swarm drone is adjusted again through the global planning subsystem, including:

[0032] Based on the feedback information of the local planning subsystem, task sub-areas with and without task execution obstacles are determined; based on the working status of all drones in the task sub-area without task execution obstacles, the drones in the task sub-area without task execution obstacles are dispatched to the task sub-area with task execution obstacles.

[0033] Compared with the existing technology, the dual planning method for cluster UAV tasks constructs a hierarchical task planning architecture including a global planning subsystem and a local planning subsystem, so that the global planning subsystem executes task sub-area division and task allocation; based on the local planning subsystem, the task is refined and the path is planned for the UAVs in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; based on the feedback information, the task planning strategy of the cluster UAVs is adjusted again through the global planning subsystem, and a hierarchical task planning architecture is adopted to get rid of excessive dependence on a single central control node, and reduce the risk of the entire cluster being paralyzed due to the failure of a single central control node. Through the collaborative working mechanism of the global planning subsystem and the local planning subsystem, the cluster UAVs are adjusted in real time and dynamically, and flexibly adapt to changing task scenarios. Complex tasks can be decomposed into multiple relatively independent sub-tasks, reducing the complexity of task planning, reducing the task calculation amount and planning time, ensuring that the resources of each UAV are fully utilized, avoiding idle or excessive resource consumption, and improving the resource utilization rate, information processing capability and decision-making scientificity of cluster UAVs.

[0034] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0035] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A schematic flow chart of a dual planning method for swarm UAV missions provided by the present invention.

[0038] Figure 2 A schematic diagram of the relationship between subsystems under a hierarchical task planning architecture in a dual planning method for swarm UAV tasks provided by the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0040] See also Figure 1 , is a flow chart of a dual planning method for a swarm drone mission provided by an embodiment of the present invention. The dual planning method for a swarm drone mission includes:

[0041] Construct a hierarchical task planning architecture, which includes a global planning subsystem and a local planning subsystem. The global planning subsystem performs task sub-area division and task allocation.

[0042] Based on the local planning subsystem, the task is refined and the path is planned for the UAV in each task sub-area divided by the global planning subsystem, and the information is fed back to the global planning subsystem;

[0043] Based on the feedback information of the local planning subsystem, the mission planning strategy of the cluster UAV is adjusted again through the global planning subsystem.

[0044] The beneficial effects of the above technical solution are as follows: the dual planning method for cluster UAV tasks constructs a hierarchical task planning architecture including a global planning subsystem and a local planning subsystem, so that the global planning subsystem performs task sub-area division and task allocation; based on the local planning subsystem, the task is refined and the path is planned for the UAVs in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; based on the feedback information, the task planning strategy of the cluster UAV is adjusted again through the global planning subsystem, and a hierarchical task planning architecture is adopted to get rid of excessive dependence on a single central control node, and reduce the risk of paralysis of the entire cluster due to failure of a single central control node. Through the collaborative working mechanism of the global planning subsystem and the local planning subsystem, the cluster UAVs are adjusted in real time and dynamically, and flexibly adapt to changing task scenarios. Complex tasks can be decomposed into multiple relatively independent sub-tasks, reducing the complexity of task planning, reducing the amount of task calculation and planning time, ensuring full utilization of the resources of each UAV, avoiding idle or excessive resource consumption, and improving the resource utilization rate, information processing capability and decision-making scientificity of cluster UAVs.

[0045] Preferably, a hierarchical task planning architecture is constructed, including:

[0046] Obtaining hardware feature information and mission area feature information of all drones in the cluster; wherein the hardware feature information includes wireless communication performance feature information of the drones; and the mission area feature information includes physical space feature information of the mission area;

[0047] Based on the wireless communication performance characteristic information and the physical space characteristic information of the mission area, a global planning subsystem and a local planning subsystem of a hierarchical mission planning architecture are constructed; among which, the planning priority of the global planning subsystem is higher than the planning priority of the local planning subsystem.

[0048] The beneficial effects of the above technical solution are as follows: the existing technology cluster drones are highly dependent on the central control node. Once the central control node fails to work due to electromagnetic interference, hardware damage or software failure, the entire cluster drone will be paralyzed and seriously affect the task execution. In addition, the expansion of the scale of cluster drones will increase the data processing volume of the central control node, resulting in a decrease in the efficiency of task planning and failure to meet the stringent requirements of real-time task execution. In view of the above defects, a hierarchical task planning architecture is constructed for cluster drones. On the one hand, it ensures the accuracy and rationality of task planning, and on the other hand, it enhances the reliability of the cluster and the real-time task execution, so that cluster drones can still perform tasks in an orderly manner in complex task environments even if they encounter large-scale data processing challenges, electromagnetic interference or equipment hardware and software failures, and comprehensively improve the performance and practical value of cluster drones. In practical applications, the global planning subsystem and the local planning subsystem of the hierarchical task planning architecture can be but are not limited to two different servers, which have the function of wireless communication with drones and the function of independent calculation. In view of the fact that the global planning subsystem and the local planning subsystem respectively plan and allocate tasks to drones within the global scope and local scope of the task area, different performance requirements are imposed on the global planning subsystem and the local planning subsystem. In order to ensure that the global planning subsystem and the local planning subsystem can accurately plan the corresponding range of the mission area, the hardware feature information and mission area feature information of all drones in the cluster are first obtained, and the wireless communication performance characteristics of the drones and the spatial physical characteristics of the mission area are identified. This is to facilitate the subsequent selection of suitable servers as the global planning subsystem and the local planning subsystem, and to make the planning priority of the global planning subsystem higher than that of the local planning subsystem, ensuring that the global planning subsystem takes priority over the local planning subsystem in task planning.

[0049] Preferably, based on the wireless communication performance characteristic information and the physical space characteristic information of the mission area, a global planning subsystem and a local planning subsystem of a hierarchical mission planning architecture are constructed, including:

[0050] Based on the size of the task area space included in the physical space feature information of the task area, determine the wireless communication coverage area of ​​the task area; based on the wireless communication coverage area, construct a global planning subsystem of the hierarchical task planning architecture;

[0051] Based on the wireless communication bandwidth of the UAV contained in the wireless communication performance characteristic information, a local planning subsystem of the hierarchical mission planning architecture is constructed.

[0052] The beneficial effects of the above technical solution are as follows: the global planning subsystem performs the global scope of the task area task planning, which requires stable and continuous monitoring of all drones in the task area. In order to ensure the reliability of the global planning subsystem's monitoring of drones, the wireless communication coverage area of ​​the task area is determined based on the size of the task area space range contained in the physical space feature information of the task area. The larger the task area space range, the larger the wireless communication coverage area of ​​the task area. Based on the wireless communication coverage area range, a server that can match the monitoring coverage of the corresponding area is selected as the global planning subsystem to ensure the reliability of the global planning subsystem's monitoring of the task area. In addition, the local planning subsystem monitors and plans the local range (i.e., the task sub-area) of the task area, which requires the local planning subsystem to maintain continuous and stable communication with all drones in the task sub-area. To this end, based on the drone wireless communication bandwidth contained in the wireless communication performance feature information, the total wireless communication bandwidth demand of all drones in each task sub-area is determined, and a server that can meet the above total wireless communication bandwidth demand is selected as the local planning subsystem.

[0053] Preferably, the global planning subsystem performs task sub-area division and task allocation, including:

[0054] The global planning subsystem performs task sub-area division and task allocation based on external task instructions and the initial state information of all UAVs in the cluster.

[0055] The beneficial effects of the above technical solution are as follows: the global planning subsystem is responsible for receiving external task instructions and determining external specific task execution requirements. In addition, different drones in the cluster have different performance states. In order to rationally utilize the resources of different drones and effectively dispatch different drones to perform different tasks, it is necessary to combine external task instructions and the initial performance state information of all drones in the cluster, divide the sub-areas corresponding to the tasks performed by all drones in the cluster, and assign tasks to ensure that all drones in the cluster are accurately directed to the corresponding sub-areas to perform corresponding tasks.

[0056] Preferably, the global planning subsystem performs task sub-area division and task allocation based on external task instructions and initial state information of all UAVs in the cluster, including:

[0057] The global planning subsystem parses and processes external task instructions to obtain the task type and estimated completion time that the swarm drones need to perform;

[0058] The global planning subsystem collects the initial position, maximum range, maximum payload, and flight speed of all drones in the cluster as initial state information;

[0059] The global planning subsystem divides the mission area into several mission sub-areas based on the mission type and estimated completion time of the mission that the swarm drones need to perform, as well as the initial state information, and determines the mission volume and performance requirement range of the required drones in each mission sub-area, thereby assigning tasks to all drones in the swarm.

[0060] The beneficial effect of the above technical solution is: in actual work, when the global planning subsystem receives an external task instruction, it parses the external task instruction to obtain the overall characteristics of the task target, such as the scope of the task area, the comprehensive requirements of the task type, and at least one of the expected completion time of the task. At the same time, the initial state information of all drones in the cluster, such as the initial position, maximum cruising range, maximum load, and flight speed, is collected to quantitatively characterize the performance state of all drones in the cluster. Based on the above information, the global planning subsystem uses a convolutional neural network model to implement the optimization algorithm calculation, divide the task area into several task sub-areas, and determine the task volume of each task sub-area and the performance requirement range of the required drones, so as to assign tasks to all drones in the cluster; for example, in a large-area geographic mapping task, the mapping area is divided into several sub-areas according to its shape and geographic complexity, and the number and type of drones required for each sub-area are estimated (such as drones suitable for mountain mapping and drones suitable for plain mapping).

[0061] Preferably, based on the local planning subsystem, task refinement and path planning are performed for the UAV in each task sub-area divided by the global planning subsystem, including:

[0062] Based on the local planning subsystem, the task execution status and location information of all UAVs in each task sub-area are collected;

[0063] Based on the mission execution status information and location information, the mission is refined and the path is planned for all UAVs in each mission sub-area.

[0064] The beneficial effects of the above technical solution are as follows: the local planning subsystem is used to refine the tasks and plan the paths of the drones in each task sub-area divided by the global planning subsystem. In this process, a local communication network is established between the drones to exchange information such as positions and task execution progress. The local planning subsystem collects information such as the task execution progress and position of all drones in each task sub-area, and uses a distributed collaborative algorithm to dynamically adjust the task execution of all drones in the same task sub-area, so as to achieve task refinement and path planning for all drones in each task sub-area. For example, when a drone finds that it is low on power or encounters sudden airflow during the execution of a task, it will send a signal to the surrounding drones. The surrounding drones will reallocate the tasks in the sub-area through a distributed collaborative algorithm based on their own task conditions and remaining resources to ensure the continuity and efficiency of task execution.

[0065] Preferably, based on the task execution status information and location information, task refinement and path planning are performed for all drones in each task sub-area, including:

[0066] Based on the mission execution status information and location information, identify the drones in abnormal status within the mission sub-area;

[0067] The task execution status information of other UAVs around the UAV in the abnormal state in the task sub-area is obtained, and based on the task execution status information of other UAVs in the surrounding area, other UAVs in the surrounding area are instructed to replace the task originally performed by the UAV in the abnormal state.

[0068] The beneficial effects of the above technical solution are: based on the task execution status information and position information of each drone in the task sub-area, the drone in the abnormal state in the task sub-area is identified, that is, the drone that cannot continue to execute the original task in the task sub-area is identified; then the task execution status information of other drones around the drone in the abnormal state in the task sub-area is obtained, and it is judged whether other surrounding drones have completed their original tasks, that is, it is judged whether other surrounding drones are in an idle state, so that the drone in the idle state replaces the drone in the abnormal state to execute the task originally executed by the drone, so that the drone in the abnormal state stops executing the original task, and instructs the drone in the idle state to continue to execute the above task, so as to ensure the continuity and reliability of task execution.

[0069] Preferably, the local planning subsystem feeds back information to the global planning subsystem, including:

[0070] The local planning subsystem feeds back the task execution status information and its own resource status information of all drones in the task sub-area to the global planning subsystem; the task execution status information includes the flight path and the amount of completed tasks during the drone's task execution; the own resource status information includes the remaining power of all drones in the task sub-area and the number of drones in normal state.

[0071] The beneficial effect of the above technical solution is: when the task execution status of a drone in a certain task sub-area changes, in order to fully schedule the task execution status of all drones in the task area, the flight paths and completed tasks of all drones in the task sub-area during the task execution, as well as the remaining power of all drones in the task sub-area and the number of drones in normal status are fed back to the global planning subsystem, so as to facilitate the update of the global planning subsystem on the task execution status of all drones in the task area, and improve the subsequent global planning subsystem to accurately and comprehensively update the drone scheduling status.

[0072] Preferably, an information fusion and arbitration subsystem is also provided;

[0073] The information fusion and arbitration subsystem screens the feedback information and the information collected by the global planning subsystem, and removes the redundant information and conflicting information between the feedback information and the collected information.

[0074] The beneficial effects of the above technical solution are: in order to improve the reliability of information interaction between the global planning subsystem and the local planning subsystem, an information fusion and arbitration subsystem is set between the global planning subsystem and the local planning subsystem. Figure 2 As shown, the information fusion and arbitration subsystem can be, but is not limited to, a server with information processing functions. The information fusion and arbitration subsystem is responsible for the information interaction integration and screening between the global planning subsystem and the local planning subsystem to avoid information redundancy and conflict, such as screening the feedback information and the information collected by the global planning subsystem, and eliminating the redundant information part and the conflicting information part between the feedback information and the collected information. In addition, when the decision of the global planning subsystem and the local planning subsystem diverges, the information fusion and arbitration subsystem makes judgments and coordination based on the preset rules and priority system; for example, in a reconnaissance mission, if the global planning subsystem requires the drone in a certain area to change the reconnaissance route to cooperate with the overall strategy, but the local planning subsystem objects due to the potential threat in the area, the information fusion and arbitration subsystem will comprehensively consider the strategic importance of the mission, the degree of regional threat, and the status of drone resources, and make the final decision.

[0075] Preferably, based on the feedback information of the local planning subsystem, the mission planning strategy of the swarm UAVs is adjusted again through the global planning subsystem, including:

[0076] Based on the feedback information of the local planning subsystem, the task sub-areas with and without task execution obstacles are determined; based on the working status of all UAVs in the task sub-area without task execution obstacles, the UAVs in the task sub-area without task execution obstacles are dispatched to the task sub-area with task execution obstacles.

[0077] The beneficial effects of the above technical solution are: based on the feedback information of the local planning subsystem, the task sub-areas with and without task execution obstacles are determined, that is, the task sub-areas corresponding to the tasks that can be completely and continuously executed and the tasks that cannot be completely and continuously executed are determined, and then based on the working status of all drones in the task sub-area without task execution obstacles, all idle drones in the task sub-area without task execution obstacles are determined, and these idle drones are instructed to fly from the current task sub-area to the task sub-area with task execution obstacles, so as to continue to execute the corresponding tasks in the task sub-area with task execution obstacles, thereby ensuring the integrity and continuity of task execution in the entire task area.

[0078] In summary, the dual planning method for swarm UAV tasks of this application has the following beneficial effects:

[0079] First, compared with the traditional centralized task planning method, the hierarchical task planning architecture gets rid of the excessive reliance on a single central control node; when the drones in the local area or the local planning subsystem fails or is interfered with, the drones in other areas can still continue to perform tasks under the overall planning subsystem, effectively avoiding the risk of the entire cluster of drones being paralyzed due to the failure of the central control node, and greatly improving the reliability of the cluster drone system in performing tasks in complex environments. For example, in a reconnaissance mission, if traditional centralized planning is adopted, once the central control node is attacked by the enemy or fails, the entire cluster of drones will lose its task planning capability and the reconnaissance mission will be forced to be interrupted. Under the hierarchical task planning architecture of this application, even if drones in some areas are interfered with, drones in other areas can continue to perform reconnaissance missions and transmit information back to ensure the continuity of the mission.

[0080] Second, based on the collaborative working mechanism of the global planning subsystem and the local planning subsystem, dynamic adjustments can be made according to the real-time situation during task execution; the local planning subsystem can quickly respond to emergencies in the local area, such as environmental changes, drone failures, etc., and adjust task allocation and path planning in a timely manner; the global planning subsystem starts from an overall perspective and flexibly allocates resources and adjusts task priorities based on local feedback. This flexibility enables cluster drones to better adapt to changing mission scenarios.

[0081] Third, through hierarchical task area division and task type allocation, the global planning subsystem can decompose complex tasks into multiple relatively independent subtasks, reducing the complexity of task planning, reducing the amount of calculation and planning time. At the same time, the distributed collaborative algorithm of the local planning subsystem enables drones to collaborate efficiently and autonomously within the sub-area, avoiding the communication bottleneck and delay problems caused by centralized processing; for example, in large-scale geographic surveying and mapping tasks, compared with traditional centralized planning methods, it can significantly shorten the overall task completion time and improve task execution efficiency.

[0082] Fourth, the performance characteristics and resource status of each drone in the cluster are fully considered during the task planning process. The global planning subsystem makes reasonable arrangements based on the drone's range, payload and other parameters when initially allocating tasks. The local planning subsystem can make dynamic adjustments according to actual conditions during execution to ensure that the resources of each drone are fully utilized, avoid idle or excessive resource consumption, and improve the resource utilization rate of the entire cluster of drones.

[0083] Fifth, by introducing the information fusion and arbitration module subsystem, the problems of information inconsistency and decision conflicts that may occur in the hierarchical task planning architecture are effectively solved. The information transmitted by the global planning subsystem and the local planning subsystem is integrated and screened to ensure the accuracy and effectiveness of the information. The global planning subsystem and the local planning subsystem can make decisions and plans based on high-quality information, thereby improving information processing capabilities and the scientific nature of decision-making.

[0084] The dual planning method of swarm UAV missions in this application can be used in scenarios such as agricultural plant protection, urban environmental monitoring, mountain rescue, military reconnaissance, etc. The specific application implementation examples are as follows:

[0085] Example 1

[0086] In a simulated agricultural plant protection mission, there are 20 swarm drones. The global planning subsystem first receives the mission instructions and determines that the farmland area that needs plant protection is 500 mu, the crop type is wheat, and the pest situation is mild aphid infestation. Collect the initial information of 20 drones, of which 10 drones have a maximum range of 30 kilometers, a maximum payload of 10 kilograms, and a flight speed range of 10-15 meters per second; the other 10 drones have a maximum range of 25 kilometers, a maximum payload of 8 kilograms, and a flight speed range of 8-12 meters per second.

[0087] The global planning subsystem divides 500 mu of farmland into 4 areas according to the shape of the farmland and the performance of the drone, assigns 5 drones to each area, and determines the task type to spray a specific concentration of aphid control agent. In each area of ​​the local planning subsystem, the drone establishes a local communication network, plans a specific flight path based on its own position and task requirements, and ensures uniform spraying of the agent. For example, when one of the drones finds that the remaining agent is insufficient during the mission, it informs the surrounding drones through the local communication network. The surrounding drones adjust the task allocation according to their own mission progress and the remaining agent, and a drone with sufficient remaining agent and closer distance takes over its unfinished spraying task. During the execution of the task, the local planning subsystem feeds back the task completion progress of each area (such as the proportion of the sprayed area), the remaining power of the drone, and the remaining agent to the global planning subsystem in real time. Based on the feedback information, if the global planning subsystem finds that the task execution progress of a certain area is lagging due to complex terrain, it can adjust the flight path of drones in other areas to give priority to supporting this area to ensure that the plant protection task of the entire farmland is completed within the expected time.

[0088] Example 2

[0089] In the scenario of urban environmental monitoring, there are 30 swarm drones. After receiving the task, the global planning subsystem specifies that the monitoring area is a city center area with an area of ​​100 square kilometers. The monitoring tasks include air quality monitoring, noise monitoring, and thermal imaging monitoring of some areas. Information from 30 drones is collected, of which 15 drones are equipped with high-precision air quality sensors and noise sensors, with a maximum range of 20 kilometers and a flight speed of 10-18 meters per second; the other 15 drones are equipped with thermal imagers, with a maximum range of 18 kilometers and a flight speed of 8-15 meters per second. The global planning subsystem divides the city center area into 6 monitoring sub-areas, assigns 5 drones to each sub-area, and determines the monitoring focus of different areas. In the local planning subsystem, drones perform task refinement and path planning in their respective sub-areas. For example, in a sub-area, a drone performing air quality monitoring finds that the air quality data in a certain area is abnormal, transmits the information to other drones through the local communication network, and adjusts its own flight path for more detailed monitoring. Drones equipped with thermal imagers conduct thermal imaging monitoring of power facilities in the city during specific time periods to check for overheating hazards according to the instructions of the global planning subsystem. The local planning subsystem continuously feeds back monitoring data and drone status information to the global planning subsystem. Based on the feedback, if the global planning subsystem finds that the noise monitoring data in a certain area is missing or abnormal, drones in other areas can be deployed to carry noise sensors for supplementary monitoring to ensure the data integrity and accuracy of the entire urban environmental monitoring mission.

[0090] Example 3

[0091] In the rescue material transportation mission in the mountainous area, there are 15 swarm drones. The global planning subsystem obtains the mission information, that is, to transport relief materials, including food, medicine and tents, to three disaster-stricken villages in the mountainous area, with a total transportation volume of 500 kg. The performance parameters of 15 drones are collected, of which 5 drones have a maximum load of 50 kg, a maximum range of 40 km, and a flight speed of 12-18 m / s; the other 10 drones have a maximum load of 30 kg, a maximum range of 30 km, and a flight speed of 10-15 m / s. The global planning subsystem plans the material distribution and transportation routes according to the village location and drone performance, and distributes the tasks to 3 local areas, corresponding to 3 villages. In the local planning subsystem, during the flight of drones in the mountainous area, due to the complex terrain and changeable airflow in the mountainous area, drones cooperate with each other through local communication networks. For example, a drone encounters strong airflow during flight and cannot fly as planned. It sends a distress signal to the surrounding drones and shares the information of the materials it carries. The surrounding drones re-plan their missions based on their own positions, remaining power and load conditions, and a drone that is closer and has spare load will take over the transportation mission to ensure that the relief supplies can be smoothly delivered to each village. The local planning subsystem feeds back the progress of material transportation, drone status and other information to the global planning subsystem in real time. Based on the feedback, if the global planning subsystem finds that the material transportation progress of a village is slow, it can adjust the transportation mission priority of drones in other areas to give priority to ensuring the material supply of the village.

[0092] Example 4

[0093] In the military reconnaissance and strike scenario, the number of cluster drones is set to 50. The global planning subsystem receives the task instructions issued by the superior, with the goal of conducting reconnaissance on a mountainous military target area of ​​about 1,000 square kilometers and striking high-value targets. The initial data of 50 drones are collected, of which 20 are reconnaissance drones equipped with high-resolution optical cameras, infrared thermal imagers and electronic signal detection equipment, with a maximum range of 80 kilometers and a flight speed of 15-25 meters per second; the other 30 are attack drones carrying small precision-guided bombs, with a maximum range of 60 kilometers and a flight speed of 18-30 meters per second. The global planning subsystem first divides the military target area into 5 reconnaissance sub-areas and 3 strike areas, and assigns tasks according to the performance characteristics of the drones. The reconnaissance drones are dispersed to each reconnaissance sub-area, fly according to the predetermined search route, use a variety of reconnaissance equipment to conduct comprehensive reconnaissance of military facilities and troop deployment in the area, and transmit the reconnaissance data to the global planning subsystem in real time. For example, in a certain reconnaissance sub-area, a reconnaissance drone discovers the signal characteristics of a suspected high-value target, and immediately shares the target location, characteristic information, etc. with other reconnaissance drones through the local communication network for joint confirmation and more accurate positioning. After the global planning subsystem determines the location of the high-value target based on the reconnaissance information, it issues a strike command to the strike area where the attack drone is located. With the coordination of the local planning subsystem, the attack drone quickly plans the attack route, adjusts the flight attitude and speed, and takes into account the threat of enemy air defense firepower and its own concealment. During the attack, if an attack drone encounters enemy air defense firepower interception or fails, other attack drones in the local planning subsystem can reallocate the strike mission according to the real-time battlefield situation to ensure that the high-value target is successfully destroyed. The local planning subsystem continuously feeds back information such as the execution progress of the strike mission, the remaining ammunition of the drone, and the damage to the global planning subsystem. Based on the feedback, the global planning subsystem dynamically adjusts the entire military operation, such as dispatching reinforcement drones or changing the order of strikes, so as to achieve efficient completion of military reconnaissance and strike missions.

[0094] It can be seen from the contents of the above embodiments that the dual planning method for cluster UAV tasks constructs a hierarchical task planning architecture including a global planning subsystem and a local planning subsystem, so that the global planning subsystem performs task sub-area division and task allocation; based on the local planning subsystem, the task is refined and the path is planned for the UAV in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; based on the feedback information, the task planning strategy of the cluster UAV is adjusted again through the global planning subsystem, and a hierarchical task planning architecture is adopted to get rid of excessive dependence on a single central control node, and reduce the risk of paralysis of the entire cluster due to failure of a single central control node. Through the collaborative working mechanism of the global planning subsystem and the local planning subsystem, the cluster UAV is adjusted in real time and dynamically, and flexibly adapts to changing task scenarios. Complex tasks can be decomposed into multiple relatively independent sub-tasks, reducing the complexity of task planning, reducing the amount of task calculation and planning time, ensuring full utilization of the resources of each UAV, avoiding idle or excessive resource consumption, and improving the resource utilization rate, information processing capability and decision-making scientificity of the cluster UAV.

[0095] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dual planning method for swarm UAV missions, characterized in that: It includes the following steps: Construct a hierarchical task planning architecture, the architecture includes a global planning subsystem and a local planning subsystem; wherein the global planning subsystem performs task sub-area division and task allocation, including: The global planning subsystem performs task sub-area division and task allocation based on external task instructions and the initial state information of all drones in the cluster, which is specifically as follows: The global planning subsystem parses and processes the external task instructions to obtain the task type and the estimated completion time of the task that the cluster drones need to perform; The global planning subsystem collects the initial position, maximum range, maximum load, and flight speed of all drones in the cluster as the initial state information; The global planning subsystem divides the task area into several task sub-areas based on the task type and estimated task completion time that the swarm drones need to perform, as well as the initial state information, and determines the task amount and performance requirement range of the required drones in each task sub-area, thereby assigning tasks to all drones in the swarm; Based on the local planning subsystem, the task is refined and the path is planned for the UAV in each task sub-area divided by the global planning subsystem, and information is fed back to the global planning subsystem; The local planning subsystem feeds back information to the global planning subsystem, including: The local planning subsystem feeds back the task execution status information and self-resource status information of all drones in the task sub-area to the global planning subsystem; wherein the task execution status information includes the flight path and the amount of completed tasks of the drones during the task execution; the self-resource status information includes the remaining power of all drones in the task sub-area and the number of drones in normal state; Based on the feedback information of the local planning subsystem, the mission planning strategy of the swarm UAV is adjusted again through the global planning subsystem, including: Based on the feedback information of the local planning subsystem, task sub-areas with and without task execution obstacles are determined; based on the working status of all drones in the task sub-area without task execution obstacles, the drones in the task sub-area without task execution obstacles are dispatched to the task sub-area with task execution obstacles.

2. The dual planning method for swarm UAV missions as claimed in claim 1, characterized in that: Build a hierarchical task planning architecture, including: Acquire hardware feature information and mission area feature information of all drones in the cluster of drones; wherein the hardware feature information includes wireless communication performance feature information of the drones; and the mission area feature information includes physical space feature information of the mission area; Based on the wireless communication performance characteristic information and the physical space characteristic information of the task area, a global planning subsystem and a local planning subsystem of a hierarchical task planning architecture are constructed; wherein the planning priority of the global planning subsystem is higher than the planning priority of the local planning subsystem.

3. The dual planning method for swarm UAV missions as claimed in claim 2, characterized in that: Based on the wireless communication performance characteristic information and the physical space characteristic information of the mission area, a global planning subsystem and a local planning subsystem of a hierarchical mission planning architecture are constructed, including: Determine the wireless communication coverage area of ​​the task area based on the size of the task area space included in the physical space feature information of the task area; and construct a global planning subsystem of a hierarchical task planning architecture based on the wireless communication coverage area. Based on the wireless communication bandwidth of the UAV included in the wireless communication performance characteristic information, a local planning subsystem of a hierarchical mission planning architecture is constructed.

4. The dual planning method for swarm UAV missions as claimed in claim 1, characterized in that: Based on the local planning subsystem, task refinement and path planning are performed for the UAV in each task sub-area divided by the global planning subsystem, including: Based on the local planning subsystem, collect the task execution status information and location information of all UAVs in each task sub-area; Based on the task execution status information and the location information, task refinement and path planning are performed for all UAVs in each task sub-area.

5. The dual planning method for swarm UAV missions as claimed in claim 4, characterized in that: Based on the task execution status information and the location information, task refinement and path planning are performed for all UAVs in each task sub-area, including: Based on the mission execution status information and the location information, identifying a drone in an abnormal state within the mission sub-area; The task execution status information of other drones around the drone in the abnormal state in the task sub-area is obtained, and based on the task execution status information of other drones around, other drones around are instructed to replace the task originally executed by the drone in the abnormal state.

6. The dual planning method for swarm UAV missions as claimed in claim 1, characterized in that: An information fusion and arbitration subsystem is also provided; The information fusion and arbitration subsystem screens the feedback information and the information collected by the global planning subsystem, and removes redundant information parts and conflicting information parts between the feedback information and the collected information.

Images