Multi-UAV Cooperative Operation Management Method, System and Readable Storage Medium

Abstract

A method, system, and readable storage medium for multi-drone collaborative operation management disclosed by the present invention, wherein the method includes: numbering the drones in the hangar based on the operation area of the target plan to obtain a drone demand grouping; extracting the attribute information of the target plan as the input of the out-of-warehouse neural network model to obtain a drone out-of-warehouse pre-plan, and combining the drone demand grouping with the drone out-of-warehouse pre-plan to obtain a drone out-of-warehouse plan for scheduling and managing the drones; and real-time obtaining the status information of each working drone and visually displaying it for synchronous management. The present invention can adjust and optimize the out-of-warehouse plan of the drones according to the specific requirements of the target operation area and the training of the CNN neural network model, making the drone collaborative scheduling more perfect. In addition, when the drones are working, the statuses of multiple drones can be obtained and displayed, and synchronous management can be carried out according to the specific situation, effectively improving the operation ability and the rationality of multi-drone collaborative scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV operations, and more specifically, to a method, system, and readable storage medium for multi-UAV collaborative operation management. Background Art

[0002] An unmanned aerial vehicle, abbreviated as "UAV", is an unpiloted aircraft controlled by a radio remote control device and a self-contained program control device. In fact, UAVs are a general term for unpiloted aircraft. From a technical perspective, they can be classified into: unmanned fixed-wing aircraft, unmanned vertical takeoff and landing aircraft, unmanned airships, unmanned helicopters, unmanned multi-rotor aircraft, unmanned parafoil aircraft, etc. Their applications cover various industries such as police, urban management, agriculture, geology, meteorology, electricity, emergency rescue, and video shooting, and they have a wide range of uses.

[0003] Today, both cities and rural areas are being transformed by the emergence of UAVs. If you live in a rural area, you will be amazed by the outstanding performance of UAVs in enhancing transportation capabilities in remote areas and improving agricultural production efficiency; if you are a city dweller, you will also be pleasantly surprised to see the extraordinary role of UAVs in solving slow logistics and urban planning, construction, and management. In this era of technology engulfing the world, UAVs are infiltrating people's daily lives like air, in crowded big cities and rural areas at the other end.

[0004] Currently, for some fields suitable for collective UAV operations, such as fertilizing or spraying pesticides on large areas of agricultural land, there are still gaps, and it is not possible to coordinate and dispatch each UAV well. The same problem also exists in the field of high-altitude maintenance. Since traditional manual maintenance is dangerous, and using UAVs for maintenance requires reasonably dispatching a UAV fleet according to different maintenance requirements, the research on multi-UAV collaboration is urgently needed. Summary of the Invention

[0005] In view of the above problems, the purpose of the present invention is to provide a method, system, and readable storage medium for multi-UAV collaborative operation management, which can rationally configure a large UAV fleet, and can perform collaborative management according to the independent status of each UAV, so as to maximize the operation benefits of UAVs.

[0006] The first aspect of the present invention provides a method for multi-UAV collaborative operation management, including the following steps:

[0007] Number the UAVs in the hangar based on the operation area of the target plan to obtain a UAV demand grouping;

[0008] Extract the attribute information of the target plan as the input of the outbound neural network model to obtain a UAV outbound plan;

[0009] Based on the drone demand grouping and combining with the drone outbound plan, a drone outbound plan is obtained to schedule and manage the drones;

[0010] The status information of each working drone is obtained in real time and visually displayed for synchronous management.

[0011] In this solution, the drone demand grouping is obtained by numbering the hangar drones based on the operation area of the target plan, specifically:

[0012] Based on a preset magnitude algorithm, the fleet quantity matching level required for the operation area is obtained;

[0013] Based on the fleet quantity matching level, threshold judgment is performed to obtain the corresponding fleet expectation value;

[0014] Based on the matching level and the fleet expectation value, the hangar drones are numbered to obtain the drone demand grouping.

[0015] In this solution, the training method of the outbound neural network model is specifically:

[0016] Obtain the attribute information and outbound plan of the target plan of historical outbound data;

[0017] Preprocess the attribute information and outbound plan of the target plan of the historical outbound data to obtain a training sample set;

[0018] Input the training sample set into the initialized outbound neural network model for training;

[0019] Obtain the accuracy rate of the output result;

[0020] If the accuracy rate is greater than the preset accuracy threshold, stop training to obtain the outbound neural network model.

[0021] In this solution, the real-time acquisition of the status information of each working drone and visual display specifically includes:

[0022] Establish a wireless communication connection with each working drone;

[0023] Real-time obtain the working status information of each drone, where the working status information includes the overall machine status information and the operation trajectory data information;

[0024] Visualize the overall machine status information and the operation trajectory data information according to a preset display mechanism, where the overall machine status information is displayed in a list form and the operation trajectory data information is displayed as a three-dimensional coordinate animated graph.

[0025] In this solution, it also includes judging whether to trigger the automatic matching and replacement mechanism, specifically:

[0026] Extract the overall machine status information and the operation trajectory data information as recognition factors;

[0027] Judge whether the drone in the working state meets the remaining operation requirements through the recognition factors, where

[0028] If the remaining operation requirements are met, the automatic matching and replacement mechanism is not triggered;

[0029] If the remaining operation requirements cannot be met, trigger the automatic matching and replacement mechanism, output a takeoff replacement instruction to the drone in the hangar, and when the replacement drone reaches the predetermined operation position, output a recall instruction to recall the replaced drone.

[0030] In this solution, the method further includes performing a self-check operation on the drone, specifically:

[0031] Before outputting the drone out-of-hangar plan, perform a self-check on the drone;

[0032] Obtain the power information and the overall machine status information of the drone, where

[0033] Judge whether the power information meets the operation requirements of the target plan. If not, exclude the corresponding drone from the drone out-of-hangar plan;

[0034] Judge whether the overall machine status information meets the out-of-hangar requirements. If not, exclude the corresponding drone from the drone out-of-hangar plan.

[0035] The second aspect of the present invention also provides a multi-drone collaborative operation management system, including a memory and a processor. The memory includes a multi-drone collaborative operation management method program. When the multi-drone collaborative operation management method program is executed by the processor, the following steps are implemented:

[0036] Number the drones in the hangar based on the operation area of the target plan to obtain a drone demand grouping;

[0037] Extract the attribute information of the target plan as the input of the out-of-hangar neural network model to obtain a drone out-of-hangar pre-plan;

[0038] Based on the drone demand grouping and the drone out-of-hangar pre-plan, obtain a drone out-of-hangar plan to schedule and manage the drones;

[0039] Real-time obtain the status information of each working drone and visually display it for synchronous management.

[0040] In this solution, the hangar drones are numbered based on the operation area of the target solution to obtain the drone demand grouping, specifically as follows:

[0041] Based on a preset magnitude algorithm, obtain the matching level of the number of aircraft groups required for the operation area;

[0042] Based on the matching level of the number of aircraft groups, perform a threshold judgment to obtain the corresponding expected value of the aircraft group;

[0043] Based on the matching level and the expected value of the aircraft group, number the hangar drones to obtain the drone demand grouping.

[0044] In this solution, the training method of the outbound neural network model is specifically as follows:

[0045] Obtain the attribute information and outbound plan of the target solution of the historical outbound data;

[0046] Preprocess the attribute information and outbound plan of the target solution of the historical outbound data to obtain a training sample set;

[0047] Input the training sample set into the initialized outbound neural network model for training;

[0048] Obtain the accuracy rate of the output result;

[0049] If the accuracy rate is greater than the preset accuracy threshold, stop training to obtain the outbound neural network model.

[0050] In this solution, the real-time acquisition of the status information of each working drone and visual display specifically includes:

[0051] Establish a wireless communication connection with each working drone;

[0052] Real-time acquire the working status information of each drone, where the working status information includes the overall machine status information and the operation trajectory data information;

[0053] Visualize the overall machine status information and the operation trajectory data information according to a preset display mechanism, where the overall machine status information is displayed in a list form, and the operation trajectory data information is displayed as a three-dimensional coordinate dynamic graph.

[0054] In this solution, it also includes judging whether to trigger the automatic matching and replacement mechanism, specifically as follows:

[0055] Extract the overall machine status information and the operation trajectory data information as recognition factors;

[0056] Judge whether the drones in the working state meet the remaining operation requirements through the recognition factors, where

[0057] If the remaining operation requirements are met, the automatic matching replacement mechanism is not triggered;

[0058] If the remaining operation requirements cannot be met, the automatic matching replacement mechanism is triggered, and a take-off replacement instruction is output to the hangar UAV. When the replacement UAV reaches the predetermined operation position, a recall instruction is output to recall the replaced UAV.

[0059] In this solution, the method further includes performing a self-check operation on the UAV, specifically:

[0060] Before outputting the UAV out-of-hangar plan, perform a self-check on the UAV;

[0061] Obtain the power information and the overall machine status information of the UAV, where

[0062] Judge whether the power information meets the operation requirements of the target plan. If not, exclude the corresponding UAV from the UAV out-of-hangar plan;

[0063] Judge whether the overall machine status information meets the out-of-hangar requirements. If not, exclude the corresponding UAV from the UAV out-of-hangar plan.

[0064] A third aspect of the present invention provides a computer-readable storage medium, which includes a program for a method for managing multi-UAV cooperative operations of a machine. When the program for the method for managing multi-UAV cooperative operations is executed by a processor, the steps of a method for managing multi-UAV cooperative operations as described in any one of the above are implemented.

[0065] A method, system and readable storage medium for managing multi-UAV cooperative operations disclosed by the present invention can adjust and optimize the UAV out-of-hangar plan according to the specific requirements of the target operation area and the training of the CNN neural network model, making the UAV cooperative scheduling more perfect. In addition, when the UAVs are working, the statuses of multiple UAVs can be obtained and displayed, and synchronous management can be performed according to specific situations, which can effectively improve the operation ability and the rationality of multi-UAV cooperative scheduling. Description of the Drawings

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0067] Figure 1Flowchart of a method for managing multi - drone collaborative operations provided by an embodiment of the present invention;

[0068] Figure 2 Block diagram of a system for managing multi - drone collaborative operations provided by an embodiment of the present invention. Detailed implementation manners

[0069] In order to more clearly understand the above - mentioned objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0070] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0071] Figure 1 Shows a flowchart of a method for managing multi - drone collaborative operations of the present application.

[0072] As Figure 1 shown, the present application discloses a method for managing multi - drone collaborative operations, including the following steps:

[0073] S101, number the drones in the hangar based on the operation area of the target plan to obtain a drone demand grouping;

[0074] S102, extract the attribute information of the target plan as the input of the out - of - warehouse neural network model to obtain a drone out - of - warehouse pre - plan;

[0075] S103, obtain a drone out - of - warehouse plan based on the drone demand grouping in combination with the drone out - of - warehouse pre - plan to schedule and manage the drones;

[0076] S104, real - time obtain the status information of each working drone and visually display it for synchronous management.

[0077] It should be noted that first, the target solution is obtained, and then the operation area is obtained. Based on the preset magnitude algorithm, the order of magnitude of the required drones can be obtained, and then the drones in the hangar are numbered to obtain the drone demand grouping. Further, the attribute information of the target solution is extracted as the input of the outbound neural network model, and is handed over to the trained outbound neural network model for simulation to obtain the drone outbound plan. Then, according to the drone outbound plan and the drone demand grouping, the final drone outbound plan is obtained, and the status information of the drones sent for operation is obtained in real time, monitored and visually displayed, and collaborative management is carried out according to the status of different drones.

[0078] According to an embodiment of the present invention, numbering the hangar drones based on the operation area of the target solution to obtain the drone demand grouping specifically includes:

[0079] Obtaining the matching level of the number of aircraft groups required for the operation area based on a preset magnitude algorithm;

[0080] Performing threshold judgment based on the matching level of the number of aircraft groups to obtain the corresponding expected value of the aircraft group;

[0081] Numbering the hangar drones based on the matching level and the expected value of the aircraft group to obtain the drone demand grouping.

[0082] It should be noted that the matching level of the number of aircraft groups is calculated based on the operation area of the target solution. Among them, the calculation method of the magnitude algorithm is as follows:

[0083]

[0084] Among them, N t represents the order of magnitude of the required drones, represents the area of the operation area of the target solution, S v represents the theoretical operation area range of a single drone, N s represents the quantity increment, a and b are set constants. After obtaining the order of magnitude of the required drones, the drones in the hangar can be numbered to obtain the drone demand grouping. For example, If the calculation result is "10", take N s as a fixed value of "2", and obtain N t as 12. Take the threshold as six drones, and then obtain the expected value of the aircraft group as "2". Then the drone demand grouping is two groups, and the hangar drones are numbered as "1, 2, ···, 12".

[0085] According to an embodiment of the present invention, the training method of the outbound neural network model specifically includes:

[0086] Obtain the attribute information and the outbound plan of the target plan for historical outbound data;

[0087] Preprocess the attribute information and the outbound plan of the target plan for the historical outbound data to obtain a training sample set;

[0088] Input the training sample set into the initialized outbound neural network model for training;

[0089] Obtain the accuracy rate of the output result;

[0090] If the accuracy rate is greater than a preset accuracy rate threshold, stop training to obtain the outbound neural network model.

[0091] It should be noted that the outbound neural network model requires a large amount of historical data for training. The larger the amount of data, the more accurate the result. The outbound neural network model in this application can be trained with the attribute information and the outbound plan of the target plan for historical outbound data as the input. Of course, when training the neural network model, it is not only necessary to train through the attribute information and the outbound plan of the target plan for historical outbound data, but also to combine the determined outbound plan for training. By comparing a large amount of test data with real data, the obtained result will be more accurate, and thus the output result of the outbound neural network will be more accurate. Preferably, the accuracy rate threshold is generally set to 90%.

[0092] According to an embodiment of the present invention, the real-time acquisition and visual display of the state information of each working drone specifically includes:

[0093] Establish a wireless communication connection with each working drone;

[0094] Real-time acquire the working state information of each drone, where the working state information includes the overall machine state information and the operation trajectory data information;

[0095] Visualize the overall machine state information and the operation trajectory data information according to a preset display mechanism, where the overall machine state information is displayed in a list form, and the operation trajectory data information is displayed as a three-dimensional coordinate moving graph.

[0096] It should be noted that during the process of the UAV leaving the hangar and operating in the designated area, a wireless communication connection is established with the UAV to obtain and monitor the working status information of each UAV. Among them, the working status information includes the overall machine status information and the operation trajectory data information. The overall machine status information includes the working status of each component of the UAV and is visually displayed in the form of a list; the operation trajectory represents the flight path of the corresponding UAV. Since the path is three-dimensional, the operation trajectory data information is displayed as an animated three-dimensional coordinate graph.

[0097] According to an embodiment of the present invention, the method further includes determining whether to trigger an automatic matching replacement mechanism, specifically:

[0098] Extract the overall machine status information and the operation trajectory data information as identification factors;

[0099] Determine whether the UAV in the working state meets the remaining operation requirements through the identification factors, where

[0100] If the remaining operation requirements are met, the automatic matching replacement mechanism is not triggered;

[0101] If the remaining operation requirements cannot be met, the automatic matching replacement mechanism is triggered, and a takeoff replacement instruction is output to the hangar UAV. When the replacement UAV reaches the predetermined operation position, a recall instruction is output to recall the replaced UAV.

[0102] It should be noted that the identification factor of the overall machine status information is V 1 , and the identification factor of the operation trajectory data information is V 2 . Determine whether the UAV in the working state meets the remaining operation requirements, where the discrimination formula is as follows:

[0103]

[0104] Among them, α and β are set parameters. If the discrimination result meets the remaining operation requirements, the automatic matching replacement mechanism is not triggered; if the remaining operation requirements cannot be met, the automatic matching replacement mechanism is triggered, and a takeoff replacement instruction is output to the hangar UAV. When the replacement UAV reaches the predetermined operation position, a recall instruction is output to recall the replaced UAV.

[0105] It is worth mentioning that when α and β are taken as "0.5" and "0.8" respectively, the corresponding D r is "0.2", V 1 is the remaining power ratio V 1 ', and when V 2 is taken as the flight amplitude swing ratio V 2', such as V 1 ' is 20%, V 2 ' is 20%, then the discriminant result calculation formula is "0.5 * 0.2 + 0.8 * 0.2", which is equal to "0.26", greater than the set D r , which indicates that replacement is required.

[0106] According to an embodiment of the present invention, the method further includes performing a self-check operation on the drone, specifically:

[0107] Before outputting the drone out-of-warehouse plan, perform a self-check on the drone;

[0108] Obtain the power information and the overall machine status information of the drone, where

[0109] Judge whether the power information meets the operation requirements of the target plan. If not, remove the corresponding drone from the drone out-of-warehouse plan;

[0110] Judge whether the overall machine status information meets the out-of-warehouse requirements. If not, remove the corresponding drone from the drone out-of-warehouse plan.

[0111] It should be noted that before the drone out-of-warehouse plan is shown, it is also necessary to perform a self-check on the power information and the overall machine status information of the drones in the warehouse. For example, drones with a power below 80% are not allowed to be out of the warehouse, and drones with a wear ratio of the whole machine exceeding 50% are not allowed to be out of the warehouse. New drones are re-supplemented to fill the vacancies.

[0112] According to an embodiment of the present invention, the method further includes performing hierarchical display on the visualization data, specifically:

[0113] Read the overall machine status information of the drone;

[0114] Identify the attribute type of the data to be displayed in the overall machine status information;

[0115] Judge the grade factor of the attribute type for hierarchical display.

[0116] It should be noted that the overall machine status information of the drone includes many types of data. For example, remaining power value data, wing rotation speed data, airtightness data, etc. Identify the attribute type of the data to be displayed. Among them, the priority of the remaining power value data is higher than that of the wing rotation speed data, and the wing rotation speed data is higher than the airtightness data. Therefore, the remaining power value data will be displayed first, followed by the wing rotation speed data, and then the airtightness data.

[0117] According to an embodiment of the present invention, it further includes:

[0118] Establish a UAV out-of-warehouse scheduling database according to the UAV out-of-warehouse task and the out-of-warehouse plan;

[0119] The UAV out-of-warehouse scheduling database includes the task levels and the number of aircraft groups of various types of UAVs in different historical tasks, as well as the corresponding historical out-of-warehouse plans and emergency response levels under the out-of-warehouse plan;

[0120] Generate the corresponding emergency response level according to the task risk parameters of each historical out-of-warehouse plan in the UAV out-of-warehouse scheduling database;

[0121] Conduct a similarity comparison in the UAV out-of-warehouse scheduling database according to the type of UAV to be dispatched, the task level and the number of aircraft groups, and obtain the UAV historical out-of-warehouse plan that meets the preset value requirements for the similarity of the type of UAV to be dispatched, the task level and the number of aircraft groups, as the target out-of-warehouse plan for the UAV to be dispatched;

[0122] Mark the emergency response level of the out-of-warehouse plan according to the task risk parameters corresponding to the obtained UAV historical out-of-warehouse plan;

[0123] Revise the target out-of-warehouse plan according to the emergency response level, formulate the revised out-of-warehouse plan for the UAV to be dispatched, and dispatch the UAV for out-of-warehouse.

[0124] It should be noted that in order to prevent the task risk situation of the dispatched UAV and avoid damage to the UAV caused by harsh working environments and conditions, a UAV out-of-warehouse scheduling database is statistically established for the out-of-warehouse plans and emergency measures of different types of UAVs under different dispatch tasks, the number of dispatched aircraft, and the working conditions of the task plans. This database contains the task situations, the number of dispatched aircraft, and the corresponding out-of-warehouse plans and emergency risk response levels in the historical out-of-warehouse tasks of various types of UAVs, which can facilitate obtaining the best task plan and task risk situation of the UAV to be dispatched according to the comparison of the historical database, and revising the task instructions according to the risks of the task plan to reduce the task risk and UAV loss.

[0125] According to the embodiments of the present invention, it further includes:

[0126] Set a UAV status preset threshold according to the UAV status information;

[0127] The UAV status preset threshold is divided into threshold ranges according to the working status of the type of UAV when implementing the out-of-warehouse plan;

[0128] The UAV status preset threshold is divided into three hierarchical ranges;

[0129] Obtain the UAV real-time working status parameters according to the real-time overall status information and the operation trajectory data information of the dispatched UAV;

[0130] Compare the obtained real-time working state parameters of the drone with the preset threshold values of the drone state;

[0131] When the threshold value of the real-time working state parameter of the drone is within the first preset threshold range of the drone state, the drone continues to execute the task;

[0132] When the threshold value of the real-time working state parameter of the drone is within the second preset threshold range of the drone state, interrupt the working state of the drone and change it to the standby state in place, waiting for the maintenance drone to perform maintenance;

[0133] When the threshold value of the real-time working state parameter of the drone is within the third preset threshold range of the drone state, terminate the working task of the drone and recall it.

[0134] It should be noted that according to the real-time state information of various types of drones during the execution of the outbound task, their working states are formulated and divided into different levels of threshold ranges. According to the formulated threshold ranges, the real-time working state information parameters of the drones are monitored for thresholds. When the real-time working state information parameters of the drones are within different preset threshold ranges, the instructions of the drones are corrected according to the preset threshold ranges. According to the drone state preset threshold ranges of this embodiment, they are divided into: (1.0, 0.85], (0.85, 0.6], (0.6, 0]. When the threshold value of the real-time working state parameter of the drone is within the first preset threshold range, it continues to execute the task. When it is within the second preset threshold range, the working state of the drone is interrupted and changed to the standby state in place, waiting for the maintenance drone to perform maintenance. When it is within the third preset threshold range, the working task of the drone is terminated and recalled. Among them, the real-time working state parameters of the drone are obtained according to the real-time overall state information and operation trajectory data information of the dispatched drone.

[0135] According to an embodiment of the present invention, it further includes:

[0136] Real-time monitor the dynamic information of the drone fleet status;

[0137] According to the real-time working dynamic thresholds of each drone in the drone fleet and the power information and operation trajectory data information of each drone, weighted calculation is performed to obtain the weighted dynamic thresholds of each drone;

[0138] According to the weighted dynamic thresholds of each drone in the drone fleet, weighted summation is performed on the emergency response parameters of each drone to obtain a set of weighted emergency response parameters of the fleet;

[0139] Calculate the emergency response coefficient of the fleet according to the set of weighted emergency response parameters of the fleet;

[0140] According to the obtained emergency response coefficient of the drone fleet, correct the outbound plan of the drone fleet in the drone outbound scheduling database.

[0141] It should be noted that after the implementation of the UAV fleet dispatch task, this dispatch task will be added to the UAV out-of-warehouse scheduling database, and a historical UAV fleet out-of-warehouse plan will be generated. To improve the accuracy and safety of the task instructions of this UAV fleet out-of-warehouse plan, it is necessary to correct the task instructions of this out-of-warehouse plan according to the actual working conditions of the UAV fleet. Specifically, the dynamic information of the UAV fleet status is monitored in real time, and the weighted dynamic threshold of each UAV is obtained by weighted calculation based on the real-time working dynamic threshold of each UAV in the UAV fleet, the power information of each UAV, and the running trajectory data information. Then, the weighted sum of the emergency response parameters of each UAV is obtained to obtain the set of weighted emergency response parameters of the fleet and the average value is obtained to obtain the emergency response coefficient of the fleet. Then, the UAV fleet out-of-warehouse plan in the UAV out-of-warehouse scheduling database is corrected according to the emergency response coefficient of the fleet;

[0142] The calculation formula for the emergency response coefficient of the fleet is:

[0143]

[0144] Wherein, is the emergency response coefficient of the fleet, f is the emergency response parameter of each UAV, R is the real-time working dynamic threshold of each UAV, ε is the remaining power coefficient, γ is the trajectory deviation coefficient, n is the number of UAVs in the fleet, and i represents the i-th UAV among the n UAVs.

[0145] According to the embodiment of the present invention, it further includes:

[0146] Obtaining the fleet trigger rate of each UAV in the UAV fleet that triggers the automatic matching replacement mechanism by real-time monitoring;

[0147] If the fleet trigger rate is greater than the preset threshold, the entire UAV fleet is recalled;

[0148] Mark the out-of-warehouse plan of the UAV fleet according to the fleet trigger rate, correct the out-of-warehouse plan, and dispatch a suitable UAV fleet for out-of-warehouse operation according to the corrected out-of-warehouse plan.

[0149] It should be noted that if the overall trigger rate of each UAV in the dispatched UAV fleet that triggers the automatic matching replacement mechanism exceeds the preset threshold, it indicates that this fleet is not suitable for this out-of-warehouse task. The entire UAV fleet is recalled, the out-of-warehouse plan is corrected, and then a suitable UAV fleet is dispatched for out-of-warehouse operation according to the new plan.

[0150] Figure 2 The block diagram of a UAV multi-aircraft collaborative operation management system according to the present invention is shown.

[0151] As Figure 2As shown in the figure, the present invention discloses a multi - drone collaborative operation management system, including a memory and a processor. The memory includes a multi - drone collaborative operation management method program. When the multi - drone collaborative operation management method program is executed by the processor, the following steps are implemented:

[0152] Number the drones in the hangar based on the operation area of the target plan to obtain the drone demand grouping;

[0153] Extract the attribute information of the target plan as the input of the outbound neural network model to obtain the drone outbound plan;

[0154] Based on the drone demand grouping and combined with the drone outbound plan, obtain the drone outbound plan to schedule and manage the drones;

[0155] Real - time obtain the status information of each working drone and visually display it for synchronous management.

[0156] It should be noted that first, the target plan is obtained, and then the operation area is obtained. Based on the preset magnitude algorithm, the quantity level of the required drones can be obtained, and then the drones in the hangar are numbered to obtain the drone demand grouping. Further, extract the attribute information of the target plan as the input of the outbound neural network model, hand it over to the trained outbound neural network model for simulation to obtain the drone outbound plan, and then obtain the final drone outbound plan according to the drone outbound plan and the drone demand grouping, and real - time obtain the status information of the drones sent for operation, monitor and visually display it, and perform collaborative management according to the status of different drones.

[0157] According to an embodiment of the present invention, numbering the drones in the hangar based on the operation area of the target plan to obtain the drone demand grouping specifically includes:

[0158] Obtain the matching level of the required drone group quantity for the operation area based on a preset magnitude algorithm;

[0159] Perform threshold judgment based on the matching level of the drone group quantity to obtain the corresponding expected value of the drone group;

[0160] Number the drones in the hangar based on the matching level and the expected value of the drone group to obtain the drone demand grouping.

[0161] It should be noted that the matching level of the drone group quantity is calculated based on the operation area of the target plan. Among them, the calculation method of the magnitude algorithm is as follows:

[0162]

[0163] Where Nt Indicates the order of magnitude of the required UAVs Indicates the area of the operation area for the target solution, S v Indicates the theoretical operation area range of a single UAV, N s Indicates the quantity increment. a and b are set constants. After obtaining the order of magnitude of the required UAVs, the UAVs in the hangar can be numbered to obtain the UAV demand grouping. For example, If the calculation result is "10", take N s Is a fixed value of "2", obtain N t Is 12, take the threshold as six UAVs, and then obtain the expected value of the UAV group as "2", then the UAV demand grouping is two groups, and the UAVs in the hangar are numbered as "1, 2, ···, 12".

[0164] According to an embodiment of the present invention, the method for training the out-of-storage neural network model is specifically as follows:

[0165] Obtain the attribute information and out-of-storage plan of the target solution of the historical out-of-storage data;

[0166] Preprocess the attribute information and out-of-storage plan of the target solution of the historical out-of-storage data to obtain a training sample set;

[0167] Input the training sample set into the initialized out-of-storage neural network model for training;

[0168] Obtain the accuracy rate of the output result;

[0169] If the accuracy rate is greater than the preset accuracy rate threshold, stop training to obtain the out-of-storage neural network model.

[0170] It should be noted that the out-of-storage neural network model requires a large amount of historical data for training. The larger the data volume, the more accurate the result. The out-of-storage neural network model in this application can be trained by using the attribute information and out-of-storage plan of the target solution of the historical out-of-storage data as the input. Of course, when training the neural network model, it is not only necessary to train through the attribute information and out-of-storage plan of the target solution of the historical out-of-storage data, but also to combine the determined out-of-storage plan for training. By comparing a large amount of test data with real data, the obtained result will be more accurate, and thus the output result of the out-of-storage neural network will be more accurate. Preferably, the accuracy rate threshold is generally set to 90%.

[0171] According to an embodiment of the present invention, the real-time acquisition of the status information of each working UAV and visual display specifically includes:

[0172] Establish a wireless communication connection with each working UAV;

[0173] Obtain the working status information of each of the drones in real time, where the working status information includes the overall machine status information and the operation trajectory data information;

[0174] Visualize and display the overall machine status information and the operation trajectory data information according to a preset display mechanism, where the overall machine status information is displayed in the form of a list, and the operation trajectory data information is displayed as an animated 3D coordinate graph.

[0175] It should be noted that during the process of the drone leaving the hangar to the designated area for operation, establish a wireless communication connection with the drone to obtain and monitor the working status information of each of the drones, where the working status information includes the overall machine status information and the operation trajectory data information, the overall machine status information includes the working status of each component of the drone, and is visually displayed in the form of a list; the operation trajectory represents the flight path of the corresponding drone. Since the path is three-dimensional, the operation trajectory data information is displayed as an animated 3D coordinate graph.

[0176] According to an embodiment of the present invention, the method further includes determining whether to trigger an automatic matching and replacement mechanism, specifically:

[0177] Extract the overall machine status information and the operation trajectory data information as recognition factors;

[0178] Judge whether the drone in the working state meets the remaining operation requirements through the recognition factors, where

[0179] If it meets the remaining operation requirements, do not trigger the automatic matching and replacement mechanism;

[0180] If it cannot meet the remaining operation requirements, trigger the automatic matching and replacement mechanism, output a takeoff replacement instruction to the hangar drone, and when the replacement drone reaches the predetermined operation position, output a recall instruction to recall the replaced drone.

[0181] It should be noted that the recognition factor of the overall machine status information is V 1 , and the recognition factor of the operation trajectory data information is V 2 , judge whether the drone in the working state meets the remaining operation requirements, where the discrimination formula is as follows:

[0182]

[0183] Among them, α and β are set parameters. If the discrimination result meets the remaining operation requirements, the automatic matching and replacement mechanism is not triggered; if the remaining operation requirements cannot be met, the automatic matching and replacement mechanism is triggered, and a take-off replacement instruction is output to the hangar UAV. When the UAV to be replaced reaches the predetermined operation position, a recall instruction is output to recall the replaced UAV.

[0184] It is worth mentioning that when α and β are taken as "0.5" and "0.8" respectively, the corresponding D r is "0.2", and V 1 is the remaining power ratio V 1 ', and when V 2 is taken as the flight amplitude ratio V 2 ', for example, V 1 ' is 20%, and V 2 ' is 20%, then the discrimination result calculation formula is "0.5 * 0.2 + 0.8 * 0.2", which is equal to "0.26", greater than the set D r , indicating that replacement is required.

[0185] According to an embodiment of the present invention, the method further includes performing a self-check operation on the UAV, specifically:

[0186] Before outputting the UAV out-of-hangar plan, perform a self-check on the UAV;

[0187] Obtain the power information and the overall machine status information of the UAV, where

[0188] Judge whether the power information meets the operation requirements of the target plan. If not, remove the corresponding UAV from the UAV out-of-hangar plan;

[0189] Judge whether the overall machine status information meets the out-of-hangar requirements. If not, remove the corresponding UAV from the UAV out-of-hangar plan.

[0190] It should be noted that before the UAV out-of-hangar plan is shown, it is also necessary to perform a self-check on the power information and the overall machine status information of the UAVs in the hangar. For example, UAVs with a power below 80% are not allowed to leave the hangar, and those with a wear ratio of the whole machine exceeding 50% are not allowed to leave the hangar. New UAVs are re-supplemented to fill the vacancies.

[0191] According to an embodiment of the present invention, the method further includes hierarchical display of visualization data, specifically:

[0192] Read the overall machine status information of the UAV;

[0193] Identify the attribute type of the data to be displayed in the overall machine status information;

[0194] Judge the grading factor of the attribute type for hierarchical display.

[0195] It should be noted that the overall status information of the UAV includes many types of data, such as the remaining battery level data, wing rotation speed data, airtightness data, etc. Identify the attribute type of the data to be displayed. Among them, the priority of the remaining battery level data is higher than that of the wing rotation speed data, and the wing rotation speed data is higher than the airtightness data. Therefore, the remaining battery level data will be displayed first, followed by the wing rotation speed data, and then the airtightness data.

[0196] The invention embodiment further includes:

[0197] Establish a UAV outbound dispatch database according to the UAV outbound task and the outbound plan;

[0198] The UAV outbound dispatch database includes the task levels and the number of aircraft groups of various types of UAVs in different historical tasks, as well as the corresponding historical outbound plans and emergency response levels under the outbound plan;

[0199] Generate the corresponding emergency response level according to the task risk parameters of each historical outbound plan in the UAV outbound dispatch database;

[0200] Perform similarity comparison in the UAV outbound dispatch database according to the type, task level and number of aircraft groups of the UAV to be dispatched, and obtain the UAV historical outbound plan whose similarity with the type, task level and number of aircraft groups of the UAV to be dispatched meets the preset value requirements as the target outbound plan of the UAV to be dispatched;

[0201] Mark the emergency response level of the outbound plan according to the task risk parameters corresponding to the obtained UAV historical outbound plan;

[0202] Revise the target outbound plan according to the emergency response level, formulate the revised outbound plan for the UAV to be dispatched, and dispatch the UAV for outbound.

[0203] It should be noted that in order to prevent the task risk situation of the dispatched UAV and avoid damage to the UAV caused by harsh working environments and conditions, a UAV outbound dispatch database is statistically established for the outbound plans and emergency measures of different types of UAVs under different dispatch tasks, dispatched quantities and working conditions of the task plan. This database contains the task situations, dispatched quantities, corresponding outbound plans and emergency risk response levels in the historical outbound tasks of various types of UAVs, which can facilitate obtaining the best task plan and task risk situation of the UAV to be dispatched according to the comparison of the historical database, and revising the task instructions according to the risk of the task plan to reduce the task risk and UAV loss.

[0204] According to an embodiment of the present invention, it further includes:

[0205] Set a preset threshold for the UAV state according to the UAV state information;

[0206] The preset threshold for the UAV state divides the threshold range according to the working state of the UAV of the type when executing the outbound plan;

[0207] The preset threshold for the UAV state is divided into three hierarchical ranges;

[0208] Obtain the real-time working state parameters of the UAV according to the real-time overall state information and operation trajectory data information of the dispatched UAV;

[0209] Compare the threshold of the obtained real-time working state parameters of the UAV with the preset threshold of the UAV state;

[0210] When the threshold of the real-time working state parameters of the UAV is within the first preset threshold range of the UAV state, the UAV continues to execute the task;

[0211] When the threshold of the real-time working state parameters of the UAV is within the second preset threshold range of the UAV state, interrupt the working state of the UAV and change it to the standby state in place, waiting for the maintenance UAV to perform maintenance;

[0212] When the threshold of the real-time working state parameters of the UAV is within the third preset threshold range of the UAV state, terminate the working task of the UAV and recall it.

[0213] It should be noted that according to the real-time state information of each type of UAV when performing the outbound task, its working state is divided into different hierarchical threshold ranges. According to the formulated threshold ranges, the real-time working state information parameters of the UAV are monitored for thresholds. When the real-time working state information parameters of the UAV are within different preset threshold ranges, the instructions for the UAV are corrected according to the preset threshold ranges. According to this embodiment, the preset threshold range of the UAV state is divided into: (1.0, 0.85], (0.85, 0.6], (0.6, 0]. When the threshold of the real-time working state parameters of the UAV is within the first preset threshold range, it continues to execute the task. When it is within the second preset threshold range, the working state of the UAV is interrupted and changed to the standby state in place waiting for the maintenance UAV to perform maintenance. When it is within the third preset threshold range, the working task of the UAV is terminated and recalled. Among them, the real-time working state parameters of the UAV are obtained according to the real-time overall state information and operation trajectory data information of the dispatched UAV.

[0214] According to an embodiment of the present invention, it further includes:

[0215] Real-time monitor the dynamic information of the UAV fleet status;

[0216] The weighted dynamic threshold of each UAV is obtained by weighted calculation based on the real-time working dynamic threshold of each UAV in the UAV fleet, the power information of each UAV, and the operation trajectory data information;

[0217] The weighted sum of the emergency response parameters of each UAV is obtained according to the weighted dynamic threshold of each UAV in the UAV fleet to obtain the weighted emergency response parameter set of the fleet;

[0218] The emergency response coefficient of the fleet is calculated according to the weighted emergency response parameter set of the fleet;

[0219] The outbound scheduling plan of the UAV fleet in the UAV outbound scheduling database is corrected according to the obtained emergency response coefficient of the UAV fleet.

[0220] It should be noted that after the implementation of the UAV fleet dispatch task, this dispatch task is added to the UAV outbound scheduling database, and a historical UAV fleet outbound plan is generated. To improve the accuracy and safety of the task instructions of this UAV fleet outbound plan, it is necessary to correct the task instructions of this outbound plan according to the actual working conditions of the UAV fleet. Specifically, the dynamic information of the UAV fleet status is monitored in real time, the weighted dynamic threshold of each UAV is obtained by weighted calculation based on the real-time working dynamic threshold of each UAV in the UAV fleet, the power information of each UAV, and the operation trajectory data information, then the weighted sum of the emergency response parameters of each UAV is obtained to obtain the weighted emergency response parameter set of the fleet and the average is obtained to obtain the emergency response coefficient of the fleet, and then the outbound plan of this UAV fleet in the UAV outbound scheduling database is corrected according to the emergency response coefficient of the fleet;

[0221] The calculation formula of the emergency response coefficient of the fleet is:

[0222]

[0223] Among them, is the emergency response coefficient of the fleet, f is the emergency response parameter of each UAV, R is the real-time working dynamic threshold of each UAV, ε is the remaining power coefficient, γ is the trajectory deviation coefficient, n is the number of UAVs in the fleet, and i represents the i-th UAV among the n UAVs.

[0224] According to the embodiment of the present invention, it further includes:

[0225] The fleet trigger rate at which each UAV in the UAV fleet triggers the automatic matching replacement mechanism is obtained by real-time monitoring;

[0226] If the fleet trigger rate is greater than the preset threshold, the entire UAV fleet is recalled;

[0227] Mark the outbound plan of the UAV fleet according to the fleet trigger rate, correct the outbound plan, and dispatch a suitable UAV fleet for outbound operations according to the corrected outbound plan.

[0228] It should be noted that if the overall trigger rate of the UAV fleet where each UAV triggers the automatic matching replacement mechanism exceeds the preset threshold, it indicates that this fleet is not suitable for this outbound task. Recall the entire UAV fleet, correct the outbound plan, and then dispatch a suitable UAV fleet for outbound operations according to the new plan.

[0229] The third aspect of the present invention provides a computer-readable storage medium, which includes a program for a method of managing multi-UAV collaborative operations of a machine. When the program for the method of managing multi-UAV collaborative operations is executed by a processor, it realizes the steps of a method of managing multi-UAV collaborative operations as described in any one of the above.

[0230] A method, system, and readable storage medium for managing multi-UAV collaborative operations disclosed by the present invention can adjust and optimize the outbound plan of UAVs according to the specific requirements of the target operation area and the training of the CNN neural network model, making the collaborative scheduling of UAVs more perfect. In addition, when the UAVs are working, the states of multiple UAVs can be obtained and displayed, and synchronous management can be carried out according to specific situations, which can effectively improve the operation ability and the rationality of multi-UAV collaborative scheduling.

[0231] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0232] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0233] In addition, in each embodiment of the present invention, each functional unit can be entirely integrated into one processing unit, or each unit can be separately regarded as one unit, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0234] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including the above method embodiments. The foregoing storage medium includes various media that can store program codes, such as removable storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0235] Alternatively, if the above integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The foregoing storage medium includes various media that can store program codes, such as removable storage devices, ROM, RAM, magnetic disks, or optical discs.

Claims

1. A method for managing multi - drone collaborative operations, characterized in that, it includes the following steps: Number the drones in the hangar based on the operation area of the target plan to obtain a drone demand grouping; Extract the attribute information of the target plan as the input of the out - warehouse neural network model to obtain a drone out - warehouse pre - plan; Based on the drone demand grouping and combined with the drone out - warehouse pre - plan, obtain a drone out - warehouse plan to schedule and manage the drones; Real - time obtain the status information of each working drone and visually display it for synchronous management; Before obtaining the drone out - warehouse plan, it further includes: Modify the drone fleet out - warehouse plan in the drone out - warehouse scheduling database according to the fleet emergency response coefficient of the drones; the drone out - warehouse scheduling database includes the task levels and fleet numbers of different types of drones in different historical tasks, as well as the corresponding historical out - warehouse plans and emergency response levels under the out - warehouse pre - plan; the drone out - warehouse scheduling database is used to determine the drone fleet out - warehouse plan; The calculation formula of the fleet emergency response coefficient is: In the formula, is the emergency response coefficient of the drone fleet, f is the emergency response parameter of each drone, R is the real-time working dynamic threshold of each drone, ε is the remaining power coefficient, γ is the trajectory deviation coefficient, n is the number of drones in the fleet, and the subscript i is the drone number.

2. A method for managing multi - drone collaborative operations according to claim 1, characterized in that, The step of numbering the drones in the hangar based on the operation area of the target plan to obtain a drone demand grouping is specifically: Based on a preset magnitude algorithm, obtain the fleet number matching level required for the operation area; Based on the fleet number matching level, perform threshold judgment to obtain the corresponding fleet expected value; Based on the matching level and the fleet expected value, number the drones in the hangar to obtain the drone demand grouping.

3. A method for managing multi - drone collaborative operations according to claim 2, characterized in that, The training method of the out - warehouse neural network model is: Obtain the attribute information and out - warehouse pre - plan of the target plan of historical out - warehouse data; Pre - process the attribute information and out - warehouse pre - plan of the target plan of the historical out - warehouse data to obtain a training sample set; Input the training sample set into the initialized out - warehouse neural network model for training; Obtain the accuracy of the output result; If the accuracy is greater than the preset accuracy threshold, stop training to obtain the out - warehouse neural network model.

4. A method for managing multi - drone collaborative operations according to claim 1, characterized in that, The step of real - time obtaining the status information of each working drone and visually displaying it specifically includes: Establish a wireless communication connection with each working drone; Real - time obtain the working status information of each drone, where the working status information includes the overall machine status information and the operation trajectory data information; Visually display the overall machine status information and the operation trajectory data information according to a preset display mechanism, where the overall machine status information is displayed in a list form, and the operation trajectory data information is displayed as a 3D coordinate animated graph.

5. A method for managing multi - drone collaborative operations according to claim 4, characterized in that, It further includes judging whether to trigger an automatic matching and replacement mechanism, specifically: Extract the overall machine status information and the operation trajectory data information as recognition factors; Determine whether the drone in the working state meets the remaining operation requirements through the recognition factor, where if it meets the remaining operation requirements, the automatic matching and replacement mechanism is not triggered; if it cannot meet the remaining operation requirements, the automatic matching and replacement mechanism is triggered, an takeoff replacement instruction is output to the hangar drone, and when the drone to be replaced reaches the predetermined operation position, a recall instruction is output to recall the replaced drone.

6. A multi-drone collaborative operation management method according to claim 1, characterized in that the method further includes performing a self-check operation on the drone, specifically: performing a self-check on the drone before outputting the drone out-of-hangar plan; obtaining the power information and the overall machine status information of the drone, where judging whether the power information meets the operation requirements of the target plan, and if not, removing the corresponding drone from the drone out-of-hangar plan; judging whether the overall machine status information meets the out-of-hangar requirements, and if not, removing the corresponding drone from the drone out-of-hangar plan.

7. A multi-drone collaborative operation management system, characterized in that it includes a memory and a processor. The memory includes a multi-drone collaborative operation management method program. When the multi-drone collaborative operation management method program is executed by the processor, the following steps are implemented: numbering the drones based on the operation area of the target plan to obtain a drone demand grouping; extracting the attribute information of the target plan as the input of the out-of-hangar neural network model to obtain a drone out-of-hangar pre-plan, obtaining a drone out-of-hangar plan based on the drone demand grouping in combination with the drone out-of-hangar pre-plan to schedule and manage the drones; real-time obtaining the status information of each working drone and visualizing it for synchronous management; before obtaining the drone out-of-hangar plan, it further includes: correcting the drone fleet out-of-hangar plan in the drone out-of-hangar scheduling database according to the fleet emergency response coefficient of the drone; the drone out-of-hangar scheduling database includes the task levels and fleet numbers of various types of drones in different historical tasks, as well as the corresponding historical out-of-hangar plans and emergency response levels under the out-of-hangar pre-plan; the drone out-of-hangar scheduling database is used to determine the drone fleet out-of-hangar plan; the calculation formula of the fleet emergency response coefficient is: In the formula, is the emergency response coefficient of the drone fleet, f is the emergency response parameter of each drone, R is the real-time working dynamic threshold of each drone, ε is the remaining power coefficient, γ is the trajectory deviation coefficient, n is the number of drones in the fleet, and the subscript i is the drone number.

8. A multi-drone collaborative operation management system according to claim 7, characterized in that the numbering the drones based on the operation area of the target plan to obtain a drone demand grouping is specifically: obtaining the fleet number matching level required for the operation area based on a preset magnitude algorithm; performing a threshold judgment based on the fleet number matching level to obtain the corresponding fleet expected value; numbering the drones based on the matching level and the fleet expected value to obtain the drone demand grouping.

9. A multi-drone collaborative operation management system according to claim 7, characterized in that the real-time obtaining the status information of each working drone and visualizing it specifically includes: Establish a wireless communication connection with the drones described in each job; Obtain the working status information of each of the drones in real time, where the working status information includes the overall machine status information and the operation trajectory data information; Visualize and display the overall machine status information and the operation trajectory data information according to a preset display mechanism, where the overall machine status information is displayed in a list form and the operation trajectory data information is displayed as an animated 3D coordinate graph.

10. A computer-readable storage medium, characterized in that, the computer-readable storage medium includes a program for a multi-drone collaborative operation management method. When the program for the multi-drone collaborative operation management method is executed by a processor, it realizes the steps of a multi-drone collaborative operation management method as described in any one of claims 1 to 6.

Images