Control method and device of unmanned aerial vehicle group and computer equipment

Abstract

The invention relates to an unmanned aerial vehicle group control method and device and computer equipment, and relates to the technical field of unmanned aerial vehicle communication. The method comprises the following steps: acquiring a first historical data set of a historical shooting task of an unmanned aerial vehicle group for a target substation; the first historical data set comprises flight data, shooting object data and shooting environment data of each unmanned aerial vehicle in the unmanned aerial vehicle group; based on the first historical data set, obtaining feature data associated with shooting point locations of the plurality of unmanned aerial vehicles; inputting the current task demand information associated with the target substation and the feature data into a pre-constructed shooting point location optimization model to obtain target shooting point location information associated with the current task demand information; and controlling the unmanned aerial vehicle group to execute the current shooting task based on the target shooting point information. By adopting the method, the quality of the shooting data obtained in the inspection process of the transformer substation can be improved.

Description

Technical Field

[0001] The present application relates to the field of drone communication technology, and in particular to a control method, device and computer equipment for a drone swarm. Background Art

[0002] At present, during the inspection of substations, the intelligent image terminal is mainly used to take photos of the knife switches in the substation, and the operation of the substation is evaluated based on the photographed data. However, there are many types of knife switches in the substation, with different shapes and compact equipment layout, which are prone to obstruction, resulting in the inability of the intelligent image terminal to obtain a complete image of the knife switch. Therefore, how to improve the quality of the photographed data has become a problem to be solved.

[0003] However, the current method of acquiring shooting data is mainly through the shooting of PTZ cameras. However, since the shooting angle of PTZ cameras is fixed and the shooting position cannot be flexibly adjusted, the collected shooting data contains a large amount of repeated and redundant information. Therefore, there is currently a problem of low quality of the shooting data obtained during the inspection of substations. Summary of the invention

[0004] Based on this, it is necessary to provide a control method, device, computer equipment, computer-readable storage medium and computer program product for a drone swarm to address the technical problem of low quality of shooting data obtained during the above-mentioned inspection process of the substation.

[0005] In a first aspect, the present application provides a method for controlling a drone swarm, comprising:

[0006] Acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data, and shooting environment data of each drone in the drone group;

[0007] Based on the first historical data set, acquiring feature data associated with a plurality of shooting points of the drones;

[0008] Inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0009] Based on the target shooting point information, the drone group is controlled to perform the current shooting task.

[0010] In one embodiment, the method of obtaining characteristic data associated with the shooting points of multiple drones based on the first historical data set includes: obtaining first relevant data of multiple drones based on the flight data; the first relevant data includes flight status, flight position and flight speed; obtaining second relevant data of multiple drones based on the shooting object data; the second relevant data includes equipment information and spatial structure information of the target substation; obtaining third relevant data of multiple drones based on the shooting environment data; the third relevant data includes lighting information, field of view information and flight safety information associated with the shooting points of multiple drones; and obtaining characteristic data associated with the shooting points of multiple drones based on the first relevant data, the second relevant data and the third relevant data.

[0011] In one embodiment, the shooting environment data includes weather data, terrain data and electromagnetic interference data; the third related data of the multiple drones is obtained based on the shooting environment data, including: based on the weather data, obtaining the lighting conditions, field of view clarity and weather influencing factors of the shooting point as the first sub-data; based on the terrain data, obtaining the accessibility and safety of the shooting point as the second sub-data; based on the electromagnetic interference data, obtaining the impact data of the shooting point on the communication of the multiple drones as the third sub-data; based on the first sub-data, the second sub-data and the third sub-data, the third related data of the multiple drones is obtained.

[0012] In one of the embodiments, the current task requirement information and the feature data associated with the target substation are input into a pre-constructed shooting point optimization model to obtain the target shooting point information associated with the current task requirement information, including: inputting the current task requirement information and the feature data into the pre-constructed shooting point optimization model to instruct the shooting point optimization model to obtain multiple shooting points of multiple drones; based on the task requirement information, obtaining the pre-set importance level of each device to be detected in the target substation; adjusting the priority of the multiple shooting points according to the importance level; wherein the priority includes shooting order and shooting time; detecting the conflict relationship between the multiple shooting points, and if there is a conflict in the conflict relationship, adjusting the shooting order and / or the shooting time of the multiple shooting points to obtain the adjusted target shooting point information associated with the current task requirement information.

[0013] In one of the embodiments, the detecting of the conflict relationship among the plurality of shooting points and, if a conflict exists among the conflict relationships, adjusting the shooting order and / or the shooting time of the plurality of shooting points to obtain the adjusted target shooting point information associated with the current task requirement information includes: obtaining the conflict type and the impact range corresponding to the conflict relationship among the plurality of shooting points; adjusting the shooting order and / or the shooting time of the plurality of shooting points according to the conflict type, the impact range and the priority to obtain the adjusted target shooting point information associated with the current task requirement information.

[0014] In one of the embodiments, before obtaining a first historical data set of historical shooting tasks of a drone swarm for a target substation, the method includes: obtaining a second historical data set of the drone swarm for a substation group and the current task requirement information; the second historical data set includes historical flight data, historical shooting object data, and historical shooting environment data of each drone in the drone swarm; the substation group includes the target substation; and based on the second historical data set, a shooting point optimization model is trained.

[0015] In a second aspect, the present application also provides a control device for a drone swarm, comprising:

[0016] A first acquisition module is used to acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data and shooting environment data of each drone in the drone group;

[0017] A second acquisition module, configured to acquire feature data associated with a plurality of shooting points of the drone based on the first historical data set;

[0018] A data processing module, used for inputting the current task requirement information associated with the target substation and the characteristic data into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0019] The drone control module is used to control the drone group to perform the current shooting task based on the target shooting point information.

[0020] In a third aspect, the present application further provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0021] Acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data, and shooting environment data of each drone in the drone group;

[0022] Based on the first historical data set, acquiring feature data associated with a plurality of shooting points of the drones;

[0023] Inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0024] Based on the target shooting point information, the drone group is controlled to perform the current shooting task.

[0025] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0026] Acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data, and shooting environment data of each drone in the drone group;

[0027] Based on the first historical data set, acquiring feature data associated with a plurality of shooting points of the drones;

[0028] Inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0029] Based on the target shooting point information, the drone group is controlled to perform the current shooting task.

[0030] In a fifth aspect, the present application further provides a computer program product, the computer program product comprising a computer program, which implements the following steps when executed by a processor:

[0031] Acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data, and shooting environment data of each drone in the drone group;

[0032] Based on the first historical data set, acquiring feature data associated with a plurality of shooting points of the drones;

[0033] Inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0034] Based on the target shooting point information, the drone group is controlled to perform the current shooting task.

[0035] The control method, device, computer equipment, storage medium and computer program product of the drone swarm have the following beneficial effects in the process of controlling the drone swarm: first, a first historical data set of the drone swarm's historical shooting task for the target substation is obtained; the first historical data set includes the flight data, shooting object data and shooting environment data of each drone in the drone swarm; then, based on the first historical data set, characteristic data associated with the shooting points of multiple drones are obtained; then, the current task requirement information and characteristic data associated with the target substation are input into the pre-built shooting point optimization model to obtain the target shooting point information associated with the current task requirement information; finally, based on the target shooting point information, the drone swarm is controlled to perform the current shooting task. In the above process, by obtaining the first historical data set, the characteristic data can be obtained more accurately, so that the pre-built shooting point optimization model can obtain the final target shooting point information based on the characteristic data and the current task requirement, thereby efficiently controlling the drone swarm to perform the current shooting task. Therefore, the above process not only improves the shooting efficiency, image quality and flight safety, but also ensures the efficient, accurate and safe execution of the substation inspection task, so that the drone swarm can obtain high-quality shooting data during the inspection of the substation. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0037] Figure 1 A step environment diagram of a method for controlling a drone swarm in one embodiment;

[0038] Figure 2 A schematic diagram of a process for obtaining characteristic data in one embodiment;

[0039] Figure 3 A schematic diagram of a process for obtaining third related data in an embodiment;

[0040] Figure 4 is a structural block diagram of a control device for a drone swarm in one embodiment;

[0041] Figure 5 is a detailed structural block diagram of a control device for a drone swarm in one embodiment;

[0042] Figure 6 1 is a diagram showing the internal structure of a control device for a drone swarm in one embodiment. DETAILED DESCRIPTION

[0043] In the inspection of substations, traditionally, fixed-angle shooting is relied on by intelligent image terminals such as pan-tilt cameras. Due to the wide variety of knife switches in substations with different shapes and compact equipment layout, there are often obstructions, which makes it difficult for the pan-tilt camera to capture the complete image of the knife switches, thus limiting the comprehensiveness and accuracy of the shooting data. In addition, due to the fixed shooting angle, the pan-tilt camera cannot flexibly adjust the shooting position, resulting in poor quality of the collected shooting data and a large amount of repeated and redundant information, further increasing the difficulty and cost of data processing. Therefore, how to overcome the limitations of traditional inspection methods and improve the quality and comprehensiveness of shooting data has become a technical problem that needs to be solved urgently in the current substation inspection operations.

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] In one embodiment, Figure 1 As shown, a control method for a drone swarm is provided. This embodiment takes the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0046] Step S102, obtaining a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data and shooting environment data of each drone in the drone group.

[0047] Among them, the target substation refers to the substation area that needs to be inspected by the drone swarm at the current time, which can be changed arbitrarily according to different user needs; the first historical data set refers to the flight data, shooting object data and shooting environment data of each drone in the drone swarm for the target substation.

[0048] As an example, flight data can be collected through the built-in sensors of the drone, which include but are not limited to GPS, altimeters, gyroscopes and other sensor devices that can reflect the drone's position, speed, altitude, heading and other key information in real time; the captured object data includes real-life images directly related to the target substation and its equipment captured by the drone's camera during the process of shooting the target substation; the shooting environment data can be collected through the environmental sensors carried by the drone or the ground meteorological station, which can reflect the current weather conditions, lighting conditions and possible electromagnetic interference and other information in real time, and the environmental sensors include but are not limited to anemometers and light sensors.

[0049] Furthermore, shooting environment data can also be obtained by remote monitoring methods such as remote sensing satellites and drone aerial photography. When obtaining the first historical data set, it is necessary to comprehensively consider the source, accuracy, real-time and reliability of the data. For example, the data provided by meteorological monitoring stations and remote sensing satellites can be used to monitor and predict weather conditions in real time. Among them, meteorological monitoring stations can collect ground meteorological information in real time, such as temperature, humidity, wind speed, wind direction and air pressure; remote sensing satellites can provide large-scale, high-precision cloud maps, precipitation distribution and surface temperature and other spatial information. After obtaining the above data, the acquired data can be analyzed and integrated through professional meteorological data processing software to generate a weather forecast model to predict weather trends in the future, such as extreme weather events that may affect drone flights, such as strong winds, heavy rains and lightning.

[0050] Furthermore, the light sensors and electromagnetic field detectors carried by drones can be used to measure the light intensity and electromagnetic interference level in real time during the flight of drones. The light sensors can accurately measure the natural light intensity and possible shadow areas in different time periods and weather conditions; the electromagnetic field detectors can be used to monitor the electromagnetic interference on the flight path of drones, such as interference from strong electromagnetic sources such as high-voltage transmission lines and substations.

[0051] Step S104: acquiring feature data associated with the shooting points of the plurality of drones based on the first historical data set.

[0052] The feature data refers to data that is highly correlated with the shooting points of the drone, and can be selected from the first historical data set according to a predefined correlation degree.

[0053] As an example, time series analysis technology can be used to identify key nodes and possible abnormal behaviors during the flight of drones by building a time series model. Key nodes can be used as shooting points. Among them, key nodes include but are not limited to the take-off, landing and turning of drones; abnormal behaviors include but are not limited to sudden acceleration, deceleration or abnormal height; clustering algorithms can be used to group flight data, and all drones can be divided into different groups according to the similarity of flight modes, so as to obtain the differences in flight modes between different drones; then statistical analysis is performed on the clustered flight data, the performance indicators of the drone group are calculated, and the distribution of performance indicators in different groups is analyzed. The performance indicators include but are not limited to the average flight speed, maximum altitude and minimum turning radius; based on the results of time series analysis, clustering algorithm application and statistical analysis, the flight status and performance parameter characteristics of the drone are extracted as feature data.

[0054] Furthermore, the real-life images acquired by drones can be preprocessed, including denoising, contrast enhancement and other processing methods. The shape, size, color and microscopic features of substation-related equipment can be extracted from the real-life images using image segmentation, shape analysis, texture analysis and color recognition techniques. The images can be matched with the preset equipment model library to determine the type of equipment. Combined with the Geographic Information System (GIS) layout information, the identified equipment can be spatially labeled according to its actual location to form a complete substation layout and equipment type distribution information as feature data.

[0055] Furthermore, meteorological monitoring data and remote sensing satellite data can be preprocessed, including data cleaning, missing value filling and outlier detection; feature selection algorithms can be used to extract environmental features that have the greatest impact on drone flight and shooting performance as feature data; environmental features include the stability of wind speed and direction, cloud thickness and height, uniformity of light intensity and intensity changes, etc.

[0056] Step S106, inputting the current task requirement information and characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information.

[0057] Among them, the target point information is optimized point information, which can be used to control the drone group to perform the current shooting task.

[0058] As an example, a target optimization algorithm can be determined according to the current mission requirement information, and the shooting points of each drone in the drone group can be optimized based on the optimization algorithm and feature data to obtain target shooting point information. The optimization purpose can be to improve shooting efficiency, ensure image quality, and ensure the safety of the shooting process.

[0059] More importantly, the choice of target optimization algorithm can be to select multiple optimization objective functions according to the current task requirement information, evaluate the performance of various multi-objective optimization algorithms in processing the objective function, the performance includes but is not limited to convergence speed, diversity of solutions and global search capability, and the multi-objective optimization algorithm includes but is not limited to the target genetic algorithm and the multi-objective particle swarm algorithm; finally, through comparative analysis, select an algorithm that performs best in balancing multiple objectives and satisfying constraints as the target optimization algorithm.

[0060] Furthermore, in the optimization process, the feature data is used as input information, and the shooting points of each UAV are iteratively adjusted through the target optimization algorithm. A variety of feature data can be used as optimization variables, and shooting efficiency, image quality, flight safety, etc. can be used as current task requirement information. Through the optimization algorithm, a set of optimal shooting point parameters are found and obtained as target shooting point information. In the optimization process, some constraints need to be considered, such as the flight range, height limit, shooting time window, etc. of the UAV. These constraints can be achieved by setting corresponding constraint functions in the optimization algorithm. After multiple iterative optimizations, the target shooting point information is obtained. The target shooting point information includes but is not limited to the best shooting point, shooting angle and shooting time of each UAV, as well as the corresponding optimization target value. The optimization target value includes but is not limited to shooting efficiency and image quality.

[0061] Step S108, based on the target shooting point information, control the drone group to execute the current shooting task.

[0062] Among them, the target shooting point information is converted into control instructions for each drone in the drone group, and each drone is controlled to make real-time adjustments according to the control instructions, which may include adjustments to parameters such as flight speed, altitude, heading, and shooting parameters.

[0063] Moreover, during the adjustment process, the flight status and shooting quality of the drone can be monitored, and dynamic adjustments can be made according to environmental changes and current mission requirements. At the same time, the obstacle avoidance and self-protection mechanisms of the drone can be enhanced to ensure that the mission can be performed safely and stably in complex environments and high-risk scenarios; and the flight status and shooting progress of the drone can be displayed in real time through a visual monitoring platform.

[0064] In the control method of the above drone swarm, firstly, the first historical data set of the historical shooting task of the drone swarm for the target substation is obtained; the first historical data set includes the flight data, shooting object data and shooting environment data of each drone in the drone swarm; then, based on the first historical data set, the feature data associated with the shooting points of multiple drones are obtained; then, the current task requirement information and feature data associated with the target substation are input into the pre-built shooting point optimization model to obtain the target shooting point information associated with the current task requirement information; finally, based on the target shooting point information, the drone swarm is controlled to perform the current shooting task. In the above process, by obtaining the first historical data set, the feature data can be obtained more accurately, so that the pre-built shooting point optimization model can obtain the final target shooting point information based on the feature data and the current task requirement, so as to efficiently control the drone swarm to perform the current shooting task. Therefore, the above process not only improves the shooting efficiency, image quality and flight safety, but also ensures the efficient, accurate and safe execution of the substation inspection task, so that the drone swarm can obtain high-quality shooting data during the inspection process of the substation.

[0065] In an exemplary embodiment, Figure 2 As shown, step S104, based on the first historical data set, obtains characteristic data associated with the shooting points of multiple drones, including: step S202, based on the flight data, obtains first relevant data of multiple drones; the first relevant data includes flight status, flight position and flight speed; step S204, based on the shooting object data, obtains second relevant data of multiple drones; the second relevant data includes equipment information and spatial structure information of the target substation; step S206, based on the shooting environment data, obtains third relevant data of multiple drones; the third relevant data includes lighting information, field of view information and flight safety information associated with the shooting points of multiple drones; step S208, according to the first relevant data, the second relevant data and the third relevant data, obtains characteristic data associated with the shooting points of multiple drones.

[0066] The flight data can be preprocessed to obtain preprocessed flight data; the preprocessing process can include steps such as data cleaning, data smoothing and data normalization, and the process removes noise and outliers. In addition, the preprocessed flight data includes but is not limited to flight altitude, speed, attitude angle and GPS location information; the attitude angle includes pitch angle, yaw angle and roll angle, etc.

[0067] As an example, the altitude and speed data can be used, combined with the physical characteristics of the drone and the flight environment, to make a preliminary judgment on the current flight state of the drone. For example, when the altitude data continues to increase and the speed data is relatively stable, it can be judged that the drone is in a climbing state; when both the altitude and speed data remain relatively stable, it can be judged that the drone is in a cruising state; and when the altitude data continues to decrease and the speed data may increase, it can be judged that the drone is in a descending state. The GPS location information is used to determine the precise location of the drone. By parsing the longitude, latitude and altitude information in the GPS data, the precise coordinates of the drone under the global positioning system can be calculated, and the location information of the drone can be updated in real time. The speed data is used to accurately calculate the flight speed of the drone. By calculating the average value, rate of change and change trend of the speed data, the current flight speed of the drone and possible acceleration or deceleration information can be obtained.

[0068] Furthermore, the real scene images in the photographed object data can be preprocessed to obtain preprocessed images, and the preprocessed images can be analyzed using image recognition technology to identify key equipment and structures of the substation.

[0069] As an example, the edge algorithm can be used to identify the outline of objects in the real-life image to obtain an edge image; based on the edge image, the feature extraction technology is used to capture the unique visual features of the equipment and structure, and the captured features include but are not limited to shape, size, color and texture; the captured features are converted into feature vectors to obtain a feature vector set; pattern recognition processing is performed on the feature vector set, the feature vector set is compared with a known device feature library, a matching strategy is used to find the device that best matches the input features, and the recognition result is output to achieve accurate identification of key equipment in the substation, and the recognition result is further processed for spatial relationship verification, considering the spatial relationship in the image, such as the relative position between the equipment and the structure, to verify the rationality of the recognition result, and the recognition result is fine-tuned or optimized based on the spatial relationship information to output the final recognition result.

[0070] In this embodiment, by acquiring and further preprocessing the flight data, the shooting object data and the shooting environment data, and finally based on the first related data, the second related data and the third related data, the feature data associated with the shooting points of multiple drones can be more accurately acquired.

[0071] Further, in one embodiment, if Figure 3As shown, the shooting environment data includes weather data, terrain data and electromagnetic interference data; based on the shooting environment data, third related data of multiple drones are obtained, including: step S302, based on the weather data, the lighting conditions, field of view clarity and weather influencing factors of the shooting point are obtained as the first sub-data; step S304, based on the terrain data, the accessibility and safety of the shooting point are obtained as the second sub-data; step S306, based on the electromagnetic interference data, the impact data of the shooting point on the communication of multiple drones is obtained as the third sub-data; step S308, based on the first sub-data, the second sub-data and the third sub-data, the third related data of multiple drones are obtained.

[0072] Among them, according to the weather data in the shooting environment data, the lighting conditions, field of view clarity and possible weather influences of each shooting point are evaluated; according to the terrain data in the shooting environment data, the accessibility and safety of each shooting point are analyzed. Accessibility mainly examines whether the shooting point is convenient for the drone to approach and take off and land, including whether the terrain is flat and whether there is enough space for take-off and landing. Safety involves whether the shooting point is in a dangerous area, such as under high-voltage lines or on the edge of a steep mountain. According to the electromagnetic interference data in the shooting environment data, the impact of each shooting point on the drone communication is evaluated. Electromagnetic interference information may interfere with the drone's communication link, resulting in unstable or lost data transmission. Therefore, the intensity and frequency of electromagnetic interference need to be considered to ensure that the drone can maintain a stable communication connection during flight.

[0073] In this embodiment, by shooting the first sub-data, the second sub-data and the third sub-data included in the environment data, the third related data of multiple drones can be obtained more accurately, that is, the lighting information, field of view information and flight safety information associated with the shooting points of the multiple drones, so as to more accurately detect the flight status of the drone and improve the efficiency of the drone operation.

[0074] In an exemplary embodiment, current task requirement information and feature data associated with a target substation are input into a pre-constructed shooting point optimization model to obtain target shooting point information associated with the current task requirement information, including: inputting the current task requirement information and feature data into a pre-constructed shooting point optimization model to instruct the shooting point optimization model to obtain multiple shooting points of multiple drones; based on the task requirement information, obtaining the pre-set importance level of each device to be detected in the target substation; adjusting the priority of multiple shooting points according to the importance level; wherein the priority includes shooting order and shooting time; detecting the conflict relationship between multiple shooting points, and when there is a conflict in the conflict relationship, adjusting the shooting order and / or shooting time of the multiple shooting points to obtain the adjusted target shooting point information associated with the current task requirement information.

[0075] As an example, after the current mission requirement information and feature data are input into the pre-built shooting point optimization model, the shooting point optimization model will start the target optimization algorithm, and then deeply analyze and process these key features while considering the current mission requirement information. The target optimization algorithm will try multiple possible shooting point combinations and evaluate the effect of each combination. Through continuous iteration and adjustment, it will gradually approach the optimal shooting point. According to the processing results of the target optimization algorithm, the shooting point optimization model will generate shooting point optimization information, including parameters such as the best shooting position, angle, and possible flight path.

[0076] Furthermore, since there are many types of equipment in the target substation, some equipment may be more critical or important, and therefore need to be given higher priority when shooting; therefore, after calculating the preliminary best shooting points for each drone, it is also necessary to adjust the priority of each point according to the pre-set importance level of the equipment to be tested.

[0077] Among them, the pre-set importance level of the equipment to be detected includes influencing factors of multiple dimensions, including the structural complexity of the object to be photographed, the criticality of the operating status, the potential risk of failure, the analysis results of historical inspection data, and the specific requirements and goals of the inspection task. Based on the influencing factors of the above multiple dimensions, a reasonable importance level is assigned to each device to be detected, so as to divide it into different priority categories. After completing the importance level evaluation, the previously calculated optimal shooting points of each drone are adjusted accordingly according to the level of each device to be detected. For devices to be detected with a higher importance level, the priority of their shooting points is increased to ensure that these devices to be detected can be photographed first within the limited inspection time. While increasing the priority of the shooting points, the shooting quality must also be taken into account. Therefore, when adjusting the priority of the shooting points, it is necessary to ensure that the drone can capture the detailed information of these devices to be detected at the best shooting distance and angle, so as to minimize the possibility of missed detection and false detection. In addition, in order to ensure the smooth progress of the inspection task, it is also necessary to comprehensively consider factors such as the drone's flight capability, battery capacity, weather conditions, and possible air traffic restrictions.

[0078] In this embodiment, by adjusting the priority of the shooting points, it can be ensured that the drone group can give priority to more important equipment during the shooting process, thereby improving the efficiency and accuracy of the inspection.

[0079] Furthermore, in one embodiment, the conflict relationship among multiple shooting points is detected, and when there is a conflict in the conflict relationship, the shooting order and / or shooting time of the multiple shooting points are adjusted to obtain the adjusted target shooting point information associated with the current task requirement information, including: obtaining the conflict type and impact range corresponding to the conflict relationship among the multiple shooting points; adjusting the shooting order and / or shooting time of the multiple shooting points according to the conflict type, impact range and priority to obtain the adjusted target shooting point information associated with the current task requirement information.

[0080] Among them, since the drone swarm needs to maintain a certain safety distance and avoid mutual interference when performing tasks, it is necessary to ensure that the shooting points of each drone do not overlap in space; at the same time, since the shooting task may need to be completed within a specific time period, it is also necessary to check whether there is any time conflict in the shooting plans of each drone.

[0081] Specifically, spatial analysis technology can be used to check whether there is any spatial overlap among the best shooting points of each drone. If there is any overlap, spatial conflicts can be avoided by adjusting the flight trajectory or shooting angle of the drone. Time planning algorithms can also be used to check whether there is any temporal conflict among the shooting plans of each drone. If there is any conflict, temporal conflicts can be avoided by adjusting the shooting order or time window of the drone. Based on the spatial and temporal adjustment results, the target shooting point information can be obtained as the shooting plan actually executed by each drone.

[0082] In this embodiment, by comprehensively processing the flight data, shooting object information and environmental data of the drone group when performing substation inspection and shooting tasks, the shooting points of each drone are calculated and optimized, and the shooting priority is flexibly adjusted according to the importance level of the equipment to be inspected, effectively avoiding spatial overlap and time conflict, significantly improving the inspection efficiency and accuracy, and improving the quality of substation inspection shooting data.

[0083] In one embodiment, before obtaining a first historical data set of historical shooting tasks of a drone swarm for a target substation, the process includes: obtaining a second historical data set of the drone swarm for the substation group and current task requirement information; the second historical data set includes historical flight data, historical shooting object data, and historical shooting environment data of each drone in the drone swarm; the substation group includes a target substation; based on the second historical data set, a shooting point optimization model is trained.

[0084] Among them, the second historical data set includes UAV flight data, shooting object information and environmental data in multiple past shooting missions, and the current task requirement information indicates the specific requirements for the current shooting task; historical data such as UAV flight data, shooting object information, environmental data, etc. are widely collected from the second historical shooting mission, and these historical data constitute the basis for training the shooting point optimization model. At the same time, combined with the specific requirements of the corresponding shooting tasks, such as shooting targets, resolution requirements, shooting time windows and other specific requirements, the core purpose of optimization is clarified to ensure that the shooting point optimization model can improve the shooting efficiency and quality in a targeted manner; according to the optimization purpose and the second historical data set, a suitable optimization algorithm and modeling method can be selected; the shooting point optimization model is trained to obtain, that is, according to the input current task requirement information and feature data associated with the target substation, a pre-constructed shooting point optimization model of the target shooting point information associated with the current task requirement information is obtained.

[0085] As an example, the selected optimization algorithm and modeling method can be used to build a shooting point optimization model. The model is trained using historical data. During the training process, a variety of training strategies are adopted, such as data enhancement and learning rate adjustment, to improve the generalization ability and performance of the model. The data enhancement strategy can generate more training samples by transforming and expanding the data in the second historical data set, increasing the diversity of the shooting point optimization model. The learning rate adjustment strategy dynamically adjusts the learning rate according to the change of the loss function during the training process, so that the shooting point optimization model converges quickly in the early stage of training and stably optimizes in the later stage of training; at the same time, pay attention to the loss function change, accuracy and other indicators during the training process, and adjust the training strategy in time to ensure that the shooting point optimization model can fully learn the useful information in the data. After the training is completed, the shooting point optimization model can also be verified.

[0086] More importantly, the validation methods of the shooting point optimization model can be cross-validation, holdout method, etc., and the remaining historical data can be used to validate the model and evaluate its generalization ability and performance. During the validation process, attention can be paid to the performance of the shooting point optimization model in different scenarios, such as different weather conditions, different shooting objects, etc., and the shooting point optimization model can be tuned according to the validation results, such as adjusting feature weights, optimizing model structure, etc. The trained and validated model can receive input features, including features extracted from flight data, shooting object information, environmental data, etc., and output shooting point optimization information, including but not limited to the best position, shooting angle, and flight path.

[0087] In the above embodiment, the second historical data set provides the flight data and shooting information accumulated by the drone group in multiple missions, so that a more comprehensive understanding of the shooting requirements of the substation group can be achieved, so as to better set the goals of the current task; by combining the second historical data set with the current task requirements, the optimization model can make more accurate judgments based on the rich data background, thereby improving the scientificity and rationality of the selection of shooting points.

[0088] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0089] Based on the same inventive concept, the embodiment of the present application also provides a control device for a drone swarm for implementing the control method for a drone swarm involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the control device for one or more drone swarms provided below can refer to the limitations of the control method for a drone swarm above, and will not be repeated here.

[0090] In an exemplary embodiment, Figure 4 As shown, a control device for a drone group is provided, comprising: a first acquisition module 401, a second acquisition module 402, a data processing module 403 and a drone control module 404, wherein:

[0091] A first acquisition module 401 is used to acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data and shooting environment data of each drone in the drone group;

[0092] A second acquisition module 402 is used to acquire feature data associated with a plurality of shooting points of the drone based on the first historical data set;

[0093] The data processing module 403 is used to input the current task requirement information associated with the target substation and the characteristic data into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information;

[0094] The drone control module 404 is used to control the drone group to perform the current shooting task based on the target shooting point information.

[0095] Furthermore, in one embodiment, the feature data acquisition module 402 is also used to acquire first relevant data of multiple drones based on flight data; the first relevant data includes flight status, flight position and flight speed; based on the photographed object data, second relevant data of multiple drones is acquired; the second relevant data includes equipment information and spatial structure information of the target substation; based on the shooting environment data, third relevant data of multiple drones is acquired; the third relevant data includes lighting information, field of view information and flight safety information associated with the shooting points of multiple drones; based on the first relevant data, the second relevant data and the third relevant data, feature data associated with the shooting points of multiple drones is acquired.

[0096] Furthermore, in one embodiment, the feature data acquisition module 402 is also used to obtain the lighting conditions, field of view clarity and weather influencing factors of the shooting point based on the weather data as the first sub-data; to obtain the accessibility and safety of the shooting point based on the terrain data as the second sub-data; to obtain the impact data of the shooting point on the communication of multiple drones based on the electromagnetic interference data as the third sub-data; and to obtain third related data of multiple drones based on the first sub-data, the second sub-data and the third sub-data.

[0097] Furthermore, in one embodiment, the data processing module 403 is also used to input the current task requirement information and feature data into a pre-built shooting point optimization model to instruct the shooting point optimization model to obtain multiple shooting points of multiple drones; based on the task requirement information, obtain the pre-set importance level of each device to be detected in the target substation; adjust the priority of multiple shooting points according to the importance level; detect the conflict relationship between multiple shooting points, and when there is a conflict in the conflict relationship, adjust the shooting order or shooting time of the multiple shooting points to obtain the adjusted target shooting point information associated with the current task requirement information.

[0098] Furthermore, in one embodiment, the data processing module 403 is also used to obtain and analyze the conflict relationship between multiple shooting points to obtain analysis results; based on the analysis results, determine the type, impact range and priority of the conflict relationship; according to the type, impact range and priority of the conflict relationship, adjust the shooting order or shooting time of the multiple shooting points to obtain adjusted target shooting point information associated with the current task requirement information.

[0099] Furthermore, in one embodiment, the first historical data acquisition module 401 is also used to obtain a second historical data set and current task requirement information of the drone group for the substation group; the second historical data set includes historical flight data, historical shooting object data and historical shooting environment data of each drone in the drone group; the substation group includes the target substation, based on the second historical data set and current task requirement information, to obtain a pre-built shooting point optimization model.

[0100] More, such as Figure 5 As shown, the feature data acquisition module 402 may include: a first analysis unit 4021, which is used to analyze the flight data to determine the flight status, position and speed of each drone; a second analysis unit 4022, which is used to analyze the information of the equipment to be detected to identify the equipment information and spatial structure information of the substation; a third analysis unit 4023, which is used to analyze the environmental data and evaluate the lighting conditions, shading conditions and safety factors of the shooting points; an integration unit 4024, which is used to integrate the information related to the shooting points based on the analysis results to obtain feature data; the processing module 403 includes an input unit 4031, which is used to integrate the key features and the current The preset shooting point optimization model is input according to the previous task requirements; the processing unit 4032 is used to process the key features using the target optimization algorithm based on the shooting point optimization model; the generation unit 4033 is used to generate shooting point optimization information according to the optimization results; the calculation unit 4034 is used to calculate the best shooting point of each UAV according to the key features; the adjustment unit 4035 is used to adjust the priority of the best shooting point according to the importance level of the shooting object; the inspection unit 4036 is used to check whether the best shooting points of each UAV have spatial overlap or temporal conflict, and make corresponding adjustments to obtain the shooting point optimization information.

[0101] Optionally, the third analysis unit 4023 can be specifically used to: evaluate the lighting conditions, field of view clarity and possible weather impacts of each shooting point based on the weather information in the environmental data; analyze the accessibility and safety of each shooting point based on the terrain information in the environmental data; evaluate the impact of each shooting point on the drone communication based on the electromagnetic interference information in the environmental data; the inspection unit 4036 can be specifically used to: check whether there is spatial overlap or temporal conflict between the shooting points of each drone; analyze the detected conflict to determine the type, scope of impact and priority of the conflict; formulate an adjustment strategy based on the conflict analysis results, and correct the shooting points with conflicts based on the adjustment strategy to obtain the target shooting points.

[0102] Each module in the control device of the drone group can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0103] In an exemplary embodiment, a control device for a drone group is provided. Figure 6 As shown, the drone swarm control device includes a processor 600 and a memory 601, wherein the memory 601 stores machine executable instructions that can be executed by the processor 600, and the processor 600 executes the machine executable instructions to implement the above-mentioned drone swarm collaborative control method.

[0104] Further, Figure 6 The drone swarm control device shown also includes a bus 602 and a communication interface 603, and the processor 600, the communication interface 603 and the memory 601 are connected through the bus 602. Among them, the memory 601 may include a high-speed random access memory (Random Access Memory, RAM), and may also include a non-volatile memory (non-volatile memory), for example, at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 603 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. Bus 602 can be an ISA bus, a PCI bus or an EISA bus, etc.; the bus can be divided into an address bus, a data bus, a control bus, etc.; for ease of representation, Figure 6Only one bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus; the processor 600 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 600. The above processor 600 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 601, and the processor 600 reads the information in the memory 601 and completes the method steps of the aforementioned embodiment in combination with its hardware. The memory 601 of the computer device is used to store the control data of the drone group. The communication interface 603 of the computer device is used to exchange information between the processor 600 and an external device. The communication interface 603 of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a control method for a drone group is implemented.

[0105] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0106] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0107] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0108] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0109] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0110] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0111] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0112] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for controlling a drone swarm, characterized in that: The method comprises: Acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data, and shooting environment data of each drone in the drone group; Based on the first historical data set, acquiring feature data associated with a plurality of shooting points of the drones; Inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information; Based on the target shooting point information, the drone group is controlled to perform the current shooting task.

2. The method according to claim 1, characterized in that The acquiring, based on the first historical data set, feature data associated with the plurality of shooting points of the drones comprises: Based on the flight data, first relevant data of the plurality of drones is acquired; the first relevant data includes flight status, flight position and flight speed; Based on the photographed object data, second related data of the plurality of drones is acquired; the second related data includes equipment information and spatial structure information of the target substation; Based on the shooting environment data, third related data of the plurality of drones is acquired; the third related data includes illumination information, field of view information, and flight safety information associated with shooting points of the plurality of drones; According to the first relevant data, the second relevant data and the third relevant data, feature data associated with the shooting points of the plurality of drones are acquired.

3. The method according to claim 2, characterized in that The shooting environment data includes weather data, terrain data and electromagnetic interference data; The acquiring, based on the shooting environment data, third related data of the plurality of drones comprises: Based on the weather data, obtaining the lighting conditions, field of view clarity and weather influencing factors of the shooting point as first sub-data; Based on the terrain data, acquiring the accessibility and safety of the shooting point as second sub-data; Based on the electromagnetic interference data, acquiring the influence data of the shooting point on the communication of the plurality of drones as the third sub-data; Based on the first sub-data, the second sub-data and the third sub-data, the third related data of the plurality of drones is acquired.

4. The method according to claim 1, characterized in that: The step of inputting the current task requirement information and the characteristic data associated with the target substation into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information includes: Inputting the current task requirement information and the feature data into the pre-built shooting point optimization model to instruct the shooting point optimization model to obtain the multiple shooting points of the multiple drones; Based on the task requirement information, obtaining a preset importance level of each device to be detected in the target substation; According to the importance level, adjusting the priority of the plurality of shooting points; wherein the priority includes shooting order and shooting time; Detect conflict relationships among the plurality of shooting points, and if there is a conflict among the conflict relationships, adjust the shooting order and / or the shooting time of the plurality of shooting points to obtain adjusted target shooting point information associated with the current task requirement information.

5. The method according to claim 4, characterized in that The detecting the conflict relationship of the plurality of shooting points, and adjusting the shooting order and / or the shooting time of the plurality of shooting points when there is a conflict in the conflict relationship, to obtain the adjusted target shooting point information associated with the current task requirement information, includes: Obtaining conflict types and impact ranges corresponding to conflict relationships between the plurality of shooting points; According to the conflict type, the impact range and the priority, the shooting order and / or shooting time of the plurality of shooting points are adjusted to obtain the adjusted target shooting point information associated with the current task requirement information.

6. The method according to claim 1, characterized in that Before obtaining the first historical data set of the historical shooting task of the drone group for the target substation, the method includes: Acquire a second historical data set of the drone group for the substation group and the current task requirement information; the second historical data set includes historical flight data, historical shooting object data, and historical shooting environment data of each drone in the drone group; the substation group includes the target substation; Based on the second historical data set, a shooting point optimization model is trained.

7. A control device for a drone group, characterized in that: The device comprises: A first acquisition module is used to acquire a first historical data set of historical shooting tasks of the drone group for the target substation; the first historical data set includes flight data, shooting object data and shooting environment data of each drone in the drone group; A second acquisition module, configured to acquire feature data associated with a plurality of shooting points of the drone based on the first historical data set; A data processing module, used for inputting the current task requirement information associated with the target substation and the characteristic data into a pre-built shooting point optimization model to obtain target shooting point information associated with the current task requirement information; The drone control module is used to control the drone group to perform the current shooting task based on the target shooting point information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images