Cluster scheduling method and system for multi-unmanned aerial vehicle cooperative detection of hoisting equipment

Abstract

The invention relates to a cluster scheduling method and system for multi-unmanned aerial vehicle cooperative detection of hoisting equipment, and the method comprises the steps: obtaining detection object information from a detection request when the detection request of a user side is received; calling a preset equipment three-dimensional model based on the detection object information, and generating a detection task based on the equipment three-dimensional model; performing demand analysis on the detection task, and outputting task demand information; a cooperative detection strategy is generated based on the task demand to generate a cooperative flight route, the unmanned aerial vehicles of the corresponding model and number are called, and the multiple unmanned aerial vehicles are controlled to cooperatively execute the flight task according to the generated cooperative flight route; and based on a preset communication link, receiving detection data acquired by the unmanned aerial vehicle, mapping the detection data to a corresponding detection point location in the equipment three-dimensional model, and sending the detection data to a user side. The method and the device have the effect of improving the efficiency and the accuracy of large-scale equipment detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device detection, and in particular to a cluster scheduling method and system for cooperative detection of hoisting equipment by multiple unmanned aerial vehicles. BACKGROUND

[0002] With the rapid development of global logistics and manufacturing, large hoisting equipment (such as port gantry cranes, gantry cranes, nuclear power plant ring cranes, etc.) as key infrastructure, its safe and efficient operation is crucial. The traditional manual periodic detection method has been difficult to meet the requirements of modern industry for safety, efficiency and data.

[0003] The existing method is to replace manual detection with remote control of a single unmanned aerial vehicle, but the single unmanned aerial vehicle takes a long time to complete the full equipment detection, and the single machine has limited endurance, which makes it difficult to complete all inspection items of the equipment in one take-off and landing, and easy to miss detection. Even if the number of unmanned aerial vehicles is increased, the logic is still based on single machine independent detection, the unmanned aerial vehicles are independently controlled from each other, the cooperation is poor, and the detection efficiency is not obviously improved, so it needs to be improved. SUMMARY

[0004] In order to improve the efficiency and accuracy of large equipment detection, the present application provides a cluster scheduling method and system for cooperative detection of hoisting equipment by multiple unmanned aerial vehicles.

[0005] The above invention object of the present application is achieved by the following technical scheme:

[0006] A cluster scheduling method for cooperative detection of hoisting equipment by multiple unmanned aerial vehicles, comprising the steps of:

[0007] When receiving a detection request from a user terminal, obtaining detection object information from the detection request;

[0008] Based on the detection object information, a preset device three-dimensional model is called, and a detection task is generated based on the device three-dimensional model;

[0009] The detection task is analyzed, and the task requirement information is output, including detection point information, unmanned aerial vehicle quantity information and unmanned aerial vehicle performance information;

[0010] Based on the task requirement, a cooperative detection strategy is generated, a cooperative flight route is generated, a corresponding model and quantity of unmanned aerial vehicles are called, and the multiple unmanned aerial vehicles are controlled to cooperatively execute the flight task according to the generated cooperative flight route;

[0011] Based on the preset communication link, the detection data obtained by the unmanned aerial vehicle is received, the detection data is mapped to the corresponding detection point of the device three-dimensional model, and the detection data is sent to the user terminal.

[0012] By adopting the above technical solution, when drone inspection of lifting equipment is required, the user first selects the object to be inspected, i.e., the lifting equipment to be inspected. A 3D model of the corresponding equipment is generated through digital twin. By pre-setting multiple different inspection points on the equipment's 3D model, inspection tasks for different points can be generated. Further, a requirements analysis of the inspection tasks is conducted to determine the inspection points, the required drone models, performance, and quantity to cover different types of inspection points, resulting in more accurate inspection results. After determining the drone requirements, a collaborative inspection route for multiple drones is generated. The collaborative inspection route, i.e., the route of each drone, will not meet at the same trajectory point in either space or time; each route is unique. Multiple drones conduct simultaneous inspections after route planning. Compared to a single drone or multiple manually controlled drones, this shortens the inspection time and improves inspection efficiency. At the same time, the inspection data acquired by the drones, including images, videos, laser data, temperature data, and sound waves, is transmitted back to the backend through a pre-set communication link. This data is then displayed on the corresponding inspection points on the equipment's 3D model through data mapping relationships, improving the efficiency and display effect of inspection data summarization.

[0013] Optionally, the device 3D model has multiple preset detection viewpoints. The step of retrieving the preset device 3D model based on the detection object information and generating a detection task based on the device 3D model includes:

[0014] The user selects the viewpoint to be inspected for this inspection task based on the inspection viewpoint of the device's 3D model;

[0015] Obtain the task list associated with the viewpoint to be detected, and the user selects the viewpoint task to be executed for the current viewpoint based on the task list;

[0016] The 3D model generates the detection task for this device inspection based on all viewpoint tasks associated with the selected viewpoints to be inspected.

[0017] By adopting the above technical solution, the detection task is selected by the user on the 3D model of the device, making the generation of detection tasks more flexible. The user first selects the viewpoint to be detected. Each viewpoint includes different detection methods and requirements, such as time, pixel requirements, and detection methods such as thermoforming, image, and video. After selecting in the task list, the viewpoint task with specific detection methods and requirements can be obtained. The detection task can be generated by packaging the viewpoint tasks.

[0018] Optionally, the step of performing requirements analysis on the detection task and outputting task requirements information, including detection point information, number of drones, and drone performance information, includes:

[0019] Obtain the number of viewpoint tasks in this detection mission, the sensor requirements for each viewpoint task, and the observation quality indicators;

[0020] Based on the sensor requirements and observation quality indicators of each viewpoint mission, select the corresponding UAV model;

[0021] Analyze the required drone models for each viewpoint mission and match the corresponding number of drones for each model.

[0022] By adopting the above technical solution, the number, requirements, and indicators of each viewpoint task in this detection mission are first extracted as features. The required drone models are then classified in the form of aircraft model features to ensure that viewpoint tasks for each detection method can be matched. Then, by statistically analyzing the required drone models for each viewpoint task, the workload of each drone model can be known. The demand for each drone model is matched according to the workload of different drone models. Since the detection time of each drone model is different, the number of drones can be matched by first determining the model and then determining the quantity. This can achieve synchronization between different drone models in the detection cycle.

[0023] Optionally, the step of statistically analyzing the required drone models for each current viewpoint task and matching the corresponding number of drones for each model includes:

[0024] Viewpoint tasks with the same sensor requirements and observation quality indicators are labeled as the same type of viewpoint task;

[0025] The number of tasks of each type of viewpoint is counted. Based on the consensus binding algorithm with time window constraints, the total detection time is minimized and the working time of each type of viewpoint task is balanced to match the number of drones required for different types of viewpoint tasks.

[0026] By adopting the above technical solution and using the consensus binding algorithm, the objective of the function is to solve the balance between minimizing the total detection time and the working time of each type of viewpoint task. The corresponding number of drones is allocated to different types of viewpoint tasks. That is, since each type of drone is a heterogeneous drone, the corresponding detection time is different. In order to balance the detection time of each type of drone while minimizing the total detection time, and to achieve the effect of starting and ending detection at the same time as much as possible, it is necessary to consider the allocation of the number of drones of each type to enhance the synchronicity and consistency of the detection process, so as to facilitate the subsequent synchronous analysis of detection data.

[0027] Optionally, the steps of generating a collaborative flight route based on task requirements using a collaborative detection strategy, retrieving the corresponding type and number of UAVs, and controlling multiple UAVs to collaboratively execute the flight mission according to the generated collaborative flight route include:

[0028] Acquire spatial information of similar viewpoint tasks and generate associated node information;

[0029] Based on multiple sets of different associated node information, multiple sets of non-conflicting flight routes are generated according to the fast marching tree algorithm to obtain cooperative flight routes. Each set of flight routes consists of UAVs of the same type, and the flight routes of UAVs of the same type in each set of flight routes do not conflict.

[0030] By adopting the above technical solution, since viewpoint tasks belong to different detection points, the corresponding viewpoint tasks in the device's 3D model have pre-associated spatial information, such as 3D spatial coordinates. By associating the spatial information of viewpoint tasks of the same type, the associated node information is obtained, that is, the distribution of node positions that the same type of aircraft needs to fly. By quickly performing a tree algorithm combined with the time and spatial information dimensions, non-conflicting flight routes are planned for UAVs used for different types of viewpoint tasks. Flight conflicts in time and space are eliminated through communication and coordination between UAVs. Each flight route group contains one or more UAVs. When a single flight route group contains multiple UAVs, the flight routes between UAVs in the group do not conflict, achieving efficient and accurate flight route planning and improving the efficiency of viewpoint task detection.

[0031] Optionally, the step of generating multiple sets of non-conflicting flight paths based on the fast progress tree algorithm to obtain a cooperative flight path further includes:

[0032] When the number of drones of the same model in the same flight route is greater than or equal to two;

[0033] After obtaining the route of each drone in the same flight path using the fast walking tree algorithm, the number of associated node information to be responsible for each drone is allocated according to the battery level and performance of each drone in the same flight path.

[0034] By adopting the above technical solution, when there are two or more drones of the same type in the same flight route, it is necessary to divide the flight trajectory of each drone in the same flight route in order to avoid other drones in the same flight route in the time dimension. Therefore, it is necessary to consider assigning viewpoint tasks to different drones in the same flight route. At this time, it is necessary to consider the battery power and performance of different drones in the same flight route to allocate the number of viewpoint tasks they are responsible for, so as to achieve complete coordination between multiple flight routes.

[0035] Optionally, the step of receiving detection data acquired by the UAV based on a preset communication link, mapping the detection data to corresponding detection points in the device's 3D model, and sending it to the user terminal includes:

[0036] Identify the spatiotemporal stamp identifier corresponding to the detection data acquired by the drone. The spatiotemporal stamp identifier includes spatial information, time information, and the data type of the detection data.

[0037] The task of identifying viewpoints corresponding to spatiotemporal stamp identifiers;

[0038] Based on the viewpoint to be detected to which the viewpoint task belongs, the detection data is mapped to the matching detection points in the device's 3D model.

[0039] By adopting the above technical solution, the spatial information in the spatiotemporal stamp identifier, the time information when the data was acquired, and the data type of the detection data can be used to accurately determine the viewpoint task to which the current detection data belongs. Even if the detection data comes from the same detection point, it can be distinguished which viewpoint task of that detection point it is. Finally, based on the source of the viewpoint task, the viewpoint to be detected previously selected by the user terminal is found, and the monitoring data is automatically mapped to the corresponding detection point in the equipment's 3D model, realizing intelligent archiving of the detection data, which facilitates the subsequent analysis of the overall detection data of the equipment by the user terminal.

[0040] The second objective of this invention is achieved through the following technical solution:

[0041] A cluster scheduling system for collaborative detection of lifting equipment using multiple unmanned aerial vehicles (UAVs) includes:

[0042] The detection object acquisition module is used to obtain detection object information from the detection request when a detection request is received from the user.

[0043] The inspection task generation module is used to retrieve a preset 3D model of the equipment based on the inspection object information, and generate an inspection task based on the 3D model of the equipment.

[0044] The requirements analysis module is used to perform requirements analysis on the detection task and output task requirements information, including detection point information, number of drones, and drone performance information.

[0045] The strategy generation module is used to generate collaborative detection strategies and collaborative flight routes based on task requirements, retrieve the corresponding drone models and quantities, and control multiple drones to collaboratively execute flight tasks according to the generated collaborative flight routes.

[0046] The data mapping module is used to receive detection data acquired by the UAV based on a preset communication link, map the detection data to the corresponding detection points in the three-dimensional model of the device, and send it to the user terminal.

[0047] The above-mentioned objective three of this application is achieved through the following technical solution:

[0048] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment.

[0049] The fourth objective of this application is achieved through the following technical solution:

[0050] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment.

[0051] In summary, this application includes at least one of the following beneficial technical effects:

[0052] 1. After determining the drone requirements, generate a collaborative detection route for multiple drones. The collaborative detection route is the route of each drone, and they will not meet at the same trajectory point in the spatial and temporal dimensions. They are all unique. After the route planning, multiple drones can detect simultaneously. Compared with a single drone or multiple manually controlled drones, this shortens the detection time and improves the detection efficiency. At the same time, the detection data acquired by the drones, including images, videos, lasers, temperature, and sound waves, are transmitted back to the background through a preset communication link. The data is then displayed on the corresponding detection points of the equipment's 3D model through data mapping relationships, which also improves the efficiency and display effect of detection data summarization.

[0053] 2. The detection task is selected by the user on the device's 3D model, making the generation of detection tasks more flexible. The user first selects the viewpoint to be detected. Each viewpoint includes different detection methods and requirements, such as time, pixel requirements, and detection methods such as thermoforming, image, and video. After selecting in the task list, the viewpoint task with specific detection methods and requirements can be obtained. The detection task can be generated by packaging the viewpoint tasks.

[0054] 3. By first extracting features from the quantity, requirements, and indicators of each viewpoint task in this detection mission, and classifying them by aircraft type features, the required drone types are determined, ensuring that viewpoint tasks for each detection method can be matched. Then, by statistically analyzing the required drone types for each viewpoint task, the workload of each drone type can be determined. The demand for each drone type is matched according to the workload of different drone types. Since the detection time of each drone type is different, matching the number of drones by first determining the drone type and then determining the quantity can achieve synchronization between different drone types in the detection cycle.

[0055] 4. By using the spatial information in the spatiotemporal stamp, the time information when the data was acquired, and the type of the detected data, the viewpoint task to which the current detected data belongs can be accurately determined. Even if the detected data comes from the same detection point, it can be distinguished which viewpoint task it belongs to. Finally, based on the origin of the viewpoint task, the viewpoint to be detected previously selected by the user terminal is found, and the monitoring data is automatically mapped to the corresponding detection point in the equipment's 3D model, realizing intelligent archiving of the detection data, which facilitates the subsequent analysis of the overall detection data of the equipment by the user terminal. Attached Figure Description

[0056] Figure 1 This is a flowchart of an implementation of a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment according to this application;

[0057] Figure 2 This is a flowchart of step S30 in an embodiment of a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment according to this application;

[0058] Figure 3 This is a flowchart of step S50 in an embodiment of a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment according to this application;

[0059] Figure 4 This is a schematic diagram of a computer device according to this application. Detailed Implementation

[0060] The following is in conjunction with the appendix Figures 1-4 This application will be described in further detail.

[0061] In the following embodiments, such as Figure 1 As shown, this application discloses a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment, which specifically includes the following steps:

[0062] S10: When a detection request is received from the user, obtain the detection object information from the detection request;

[0063] In this embodiment, the user terminal refers to a PC, mobile PC, or APP terminal used by the testing personnel. The user terminal sends a testing request, and the backend sends the pre-entered and stored testing object information to the user terminal for the testing personnel to select. The testing object information includes pre-entered text descriptions and image information of the large-scale equipment.

[0064] S20: Retrieve a preset 3D model of the equipment based on the information of the object to be detected, and generate a detection task based on the 3D model of the equipment;

[0065] In this embodiment, the 3D model of the equipment is constructed based on the CAD drawings or oblique photogrammetry data of the lifting equipment, creating a high-precision 3D point cloud or mesh model. Within this model, a set of key components to be inspected (such as wire ropes, pulleys, hooks, tracks, connecting bolts, etc.) is defined, and each component is labeled with its inspection difficulty coefficient, inspection data type, required image resolution, sensor requirements, and standard inspection viewpoint, i.e., quality indicators.

[0066] The detection task is the sum of the tasks of the selected detection parts of the selected detection equipment in a single detection request.

[0067] Specifically, the 3D model of the equipment has multiple pre-set detection viewpoints, and step S20 includes the following steps:

[0068] S21: The user terminal selects the viewpoint to be inspected for this inspection task based on the inspection viewpoint of the device's 3D model;

[0069] S22: Obtain the task list associated with the viewpoint to be detected, and the user selects the viewpoint task to be executed for the current viewpoint based on the task list;

[0070] S23: The 3D model generates the detection task for this device detection based on all viewpoint tasks associated with the selected viewpoints to be detected.

[0071] Each inspection viewpoint has different components and inspection parameters. For example, the inspection at the same viewpoint may test two parameters: wear and temperature. Therefore, after each inspection viewpoint appears in the task list, the user must select the specific viewpoint task to clarify the content to be inspected, the required inspection data, and the required drone model. The viewpoint tasks selected in a single inspection request are associated and packaged to obtain the inspection task for this crane equipment drone inspection. Further, based on the specific viewpoint tasks in the inspection task, the number and model of drones are selected, and a collaborative flight route is planned.

[0072] S30: Perform requirements analysis on the detection task and output task requirement information, including detection point information, number of drones and drone performance information.

[0073] In this embodiment, the requirements analysis for the detection task involves analyzing the characteristics of each viewpoint task within the detection task, including the spatial location of each viewpoint task, the required UAV type for each viewpoint task, and the number of UAVs required for this detection task while minimizing detection time and considering detection time synchronization. The performance information of the UAVs is reflected in the different types of sensors and detection devices they are equipped with, such as cameras of different resolutions, and sensors such as laser, thermal imaging, infrared, and temperature sensors.

[0074] Specifically, refer to Figure 2 Step S30 includes the following steps:

[0075] S31: Obtain the number of viewpoint tasks in this detection task, the sensor requirements for each viewpoint task, and the observation quality indicators;

[0076] S32: Select the appropriate UAV model based on the sensor requirements and observation quality indicators for each viewpoint mission;

[0077] S33: Calculate the required drone models for each current viewpoint mission and match the corresponding number of drones for each model.

[0078] The purpose of calculating the required drone models for each viewpoint task is to first determine the workload differences between different models, and then, based on the individual detection time for each viewpoint task (e.g., the time for acquiring images and the time for acquiring videos are different), finally calculate the number of drones required for each model.

[0079] Furthermore, step S33 specifically includes:

[0080] S331: Mark viewpoint tasks with the same sensor requirements and observation quality indicators as the same type of viewpoint task;

[0081] S332: Count the number of tasks of the same type of viewpoint, and based on the consensus binding algorithm with time window constraints, solve the problem of minimizing the total detection time and balancing the working time of each type of viewpoint task, and match the number of drones required for different types of viewpoint tasks.

[0082] In this embodiment, similar viewpoint tasks refer to drones of the same type, including the detection equipment mounted on the drones, meaning the detection equipment and methods are identical. The similar viewpoint tasks are distinguished and their quantities are counted. A consensus binding algorithm is used to minimize the total detection time and balance the workload of each similar viewpoint task. This consensus binding algorithm is a distributed, market auction-based multi-agent task allocation algorithm. Its core idea is to mimic an "auction house," allowing each drone (agent) to "bid" for the viewpoint tasks it wants to perform, and to reach a consensus through mutual communication, ultimately forming a conflict-free and efficient task allocation scheme. Each drone is bound to non-conflicting viewpoint tasks. A time window constraint is added by collecting the detection completion time of each similar viewpoint task, ensuring that different drones face different workloads, minimizing and relatively balancing their detection time. Finally, the number of drones corresponding to the minimized and balanced detection time is matched.

[0083] S40: Generate a collaborative detection strategy based on task requirements, generate a collaborative flight route, retrieve the corresponding type and number of drones, and control multiple drones to collaboratively execute flight tasks according to the generated collaborative flight route;

[0084] In this embodiment, the collaborative detection strategy refers to a flight path design strategy where multiple drones of different models simultaneously detect lifting equipment, ensuring that there are no path conflicts between the drones in both spatial and temporal dimensions. The number of drones prepared for detection is greater than the number of drones retrieved for each mission.

[0085] Specifically, step S40 includes the following steps:

[0086] S41: Obtain spatial information of similar viewpoint tasks and generate associated node information;

[0087] S42: Based on multiple sets of different associated node information, generate multiple sets of non-conflicting flight routes according to the fast walking tree algorithm to obtain cooperative flight routes. Each set of flight routes consists of UAVs of the same type, and the flight routes of UAVs of the same type in each set of flight routes do not conflict.

[0088] Spatial information includes the spatial coordinates of viewpoint tasks in the device's 3D model. The coordinates of viewpoint tasks of the same type are associated to obtain associated node information. Different models generate multiple sets of associated node information.

[0089] The fast tree algorithm, a spatiotemporal RRT method based on priority and collision detection, assigns a priority to each drone and plans paths for each drone sequentially according to priority. When planning the path of the i-th drone, it is treated as a dynamic obstacle (whose path is known), and the planning avoids conflicts with previously planned paths in both time and space. The spatiotemporal RRT method includes a time dimension in the state space and considers time when expanding the tree. Thus, each node contains not only spatial coordinates but also time. During collision detection, we check whether the expanded new state conflicts with existing paths, achieving collision detection both temporally and spatially.

[0090] When there are two or more drones of the same model in a flight path, the route for each drone in the same flight path is obtained using the fast walking tree algorithm. Then, the number of associated node information to be responsible for each drone in the same flight path is allocated according to its battery level and performance. That is, after segmenting the drones in the same flight path based on conflict detection, the drones corresponding to each segment are selected. This allocation method makes the working time in the same flight path more balanced, allowing drones with more battery and better performance to be responsible for more viewpoint tasks.

[0091] S50: Based on a preset communication link, receive the detection data acquired by the UAV, map the detection data to the corresponding detection points in the three-dimensional model of the device, and send it to the user terminal.

[0092] In this embodiment, each operating drone is connected to each other for collision detection, and each drone has a unique mapping relationship for its viewpoint task on the device's 3D model.

[0093] Specifically, refer to Figure 3 Step S50 includes the following steps:

[0094] S51: Identify the spatiotemporal stamp identifier corresponding to the detection data acquired by the UAV. The spatiotemporal stamp identifier includes spatial information, time information, and the data type of the detection data.

[0095] S52: Task of identifying the viewpoint corresponding to the spatiotemporal stamp identifier;

[0096] S53: Based on the viewpoint to be detected to which the viewpoint task belongs, map the detection data to the matching detection points in the device's 3D model.

[0097] In this embodiment, the spatiotemporal stamp identifier can identify the detection data acquired at a specific time and in a specific space, as well as the data type, enabling a one-to-one match with the viewpoint task. After matching, the mapping path of the detection data is confirmed. The inspection personnel can click on each viewpoint to be inspected through the 3D model of the device displayed on the user terminal to access the detection data of the viewpoint task completed at that viewpoint.

[0098] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0099] In one embodiment, a cluster scheduling system for collaborative detection of multiple UAVs for lifting equipment is provided. This cluster scheduling system corresponds to the cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment described in the above embodiments. As shown in Figure x, the cluster scheduling system for collaborative detection of multiple UAVs for lifting equipment includes:

[0100] The detection object acquisition module is used to obtain detection object information from the detection request when a detection request is received from the user.

[0101] The inspection task generation module is used to retrieve a preset 3D model of the equipment based on the inspection object information, and generate an inspection task based on the 3D model of the equipment.

[0102] The requirements analysis module is used to perform requirements analysis on the detection task and output task requirements information, including detection point information, number of drones, and drone performance information.

[0103] The strategy generation module is used to generate collaborative detection strategies and collaborative flight routes based on task requirements, retrieve the corresponding drone models and quantities, and control multiple drones to collaboratively execute flight tasks according to the generated collaborative flight routes.

[0104] The data mapping module is used to receive detection data acquired by the UAV based on a preset communication link, map the detection data to the corresponding detection points in the three-dimensional model of the device, and send it to the user terminal.

[0105] Optionally, the device's 3D model has multiple pre-set detection viewpoints, and the detection task generation module includes:

[0106] The viewpoint selection submodule is used by the user to select the viewpoint to be detected for this detection task based on the device's 3D model.

[0107] The task selection submodule is used to obtain the task list associated with the viewpoint to be detected. The user selects the viewpoint task to be executed for the current viewpoint based on the task list.

[0108] The detection task generation submodule is used to generate the detection task for this equipment inspection based on all viewpoint tasks associated with the selected viewpoints to be inspected in the 3D model.

[0109] Optionally, the requirements analysis module includes:

[0110] The requirement identification submodule is used to obtain the number of viewpoint tasks in this detection task, the sensor requirements of each viewpoint task, and the observation quality indicators.

[0111] The model selection submodule is used to select the corresponding UAV model based on the sensor requirements and observation quality indicators of each viewpoint mission.

[0112] The quantity matching submodule is used to count the required drone models for each viewpoint task and match the corresponding number of drones for each model.

[0113] Optional, the quantity matching submodule includes:

[0114] Same-type marking unit, used to mark viewpoint tasks with the same sensor requirements and observation quality indicators as same-type viewpoint tasks;

[0115] The quantity calculation unit is used to count the number of each type of viewpoint task. Based on the consensus binding algorithm with time window constraints, it solves the problem of minimizing the total detection time and balancing the working time of each type of viewpoint task, and matches the number of drones required for different types of viewpoint tasks.

[0116] Optionally, the policy generation module includes:

[0117] The node association submodule is used to obtain spatial information of similar viewpoint tasks and generate associated node information;

[0118] The route generation submodule is used to generate multiple sets of non-conflicting flight routes based on multiple sets of different associated node information and according to the fast walking tree algorithm to obtain cooperative flight routes. Each set of flight routes consists of UAVs of the same type, and the flight routes of UAVs of the same type in each set of flight routes do not conflict.

[0119] Optionally, the route generation submodule also includes: when the number of drones of the same model in the same flight route is greater than or equal to two; after obtaining the route of each drone in the same flight route according to the fast travel tree algorithm; and assigning the number of associated node information to each drone in the same flight route according to the battery level and performance of each drone.

[0120] Specific limitations regarding the cluster scheduling system for collaborative detection of multiple UAVs for lifting equipment can be found in the limitations of the cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment mentioned above, and will not be repeated here. Each module in the aforementioned cluster scheduling system for collaborative detection of multiple UAVs for lifting equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0121] In one embodiment, a computer device, which may be a server, is provided, and its internal structure is shown in Figure 4. The computer device includes a processor, memory, a network interface, and a database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment.

[0122] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment.

[0123] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements a cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment.

[0124] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0125] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0126] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs, characterized in that: When a detection request is received from the user, the detection object information is obtained from the detection request; The system retrieves a pre-defined 3D model of the equipment based on the information of the object to be inspected, and generates an inspection task based on the 3D model of the equipment. Perform a requirements analysis on the detection task and output task requirement information, including detection point information, number of drones, and drone performance information. Based on task requirements, a collaborative detection strategy is generated to create a collaborative flight route. The corresponding drone model and number are then retrieved, and multiple drones are controlled to collaboratively execute the flight mission according to the generated collaborative flight route. Based on a preset communication link, the system receives detection data acquired by the drone, maps the detection data to the corresponding detection points in the device's 3D model, and sends it to the user terminal.

2. The cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs according to claim 1, characterized in that, The device 3D model has multiple pre-set detection viewpoints. The step of retrieving the pre-set device 3D model based on the detection object information and generating a detection task based on the device 3D model includes: The user selects the viewpoint to be inspected for this inspection task based on the inspection viewpoint of the device's 3D model; Obtain the task list associated with the viewpoint to be detected, and the user selects the viewpoint task to be executed for the current viewpoint based on the task list; The 3D model generates the detection task for this device inspection based on all viewpoint tasks associated with the selected viewpoints to be inspected.

3. The cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs according to claim 2, characterized in that, The steps of performing requirements analysis on the detection task and outputting task requirements information, including detection point information, number of drones, and drone performance information, include: Obtain the number of viewpoint tasks in this detection mission, the sensor requirements for each viewpoint task, and the observation quality indicators; Based on the sensor requirements and observation quality indicators of each viewpoint mission, select the corresponding UAV model; Analyze the required drone models for each viewpoint mission and match the corresponding number of drones for each model.

4. The cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment according to claim 3, characterized in that, The step of calculating the required drone models for each current viewpoint task and matching the corresponding number of drones for each model includes: Viewpoint tasks with the same sensor requirements and observation quality indicators are labeled as the same type of viewpoint task; The number of tasks of each type of viewpoint is counted. Based on the consensus binding algorithm with time window constraints, the total detection time is minimized and the working time of each type of viewpoint task is balanced to match the number of drones required for different types of viewpoint tasks.

5. A cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs according to claim 2, characterized in that, The steps of generating a collaborative flight route based on task requirements using a collaborative detection strategy, retrieving the corresponding type and number of drones, and controlling multiple drones to collaboratively execute the flight mission according to the generated collaborative flight route include: Acquire spatial information of similar viewpoint tasks and generate associated node information; Based on multiple sets of different associated node information, multiple sets of non-conflicting flight routes are generated according to the fast marching tree algorithm to obtain cooperative flight routes. Each set of flight routes consists of UAVs of the same type, and the flight routes of UAVs of the same type in each set of flight routes do not conflict.

6. A cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs according to claim 5, characterized in that, The step of generating multiple sets of non-conflicting flight paths based on the fast progress tree algorithm to obtain a cooperative flight path also includes: When the number of drones of the same model in the same flight route is greater than or equal to two; After obtaining the route of each drone in the same flight path using the fast walking tree algorithm, the number of associated node information to be responsible for each drone is allocated according to the battery level and performance of each drone in the same flight path.

7. A cluster scheduling method for collaborative detection of lifting equipment by multiple UAVs according to claim 2, characterized in that: Identify the spatiotemporal stamp identifier corresponding to the detection data acquired by the drone. The spatiotemporal stamp identifier includes spatial information, time information, and the data type of the detection data. The task of identifying viewpoints corresponding to spatiotemporal stamp identifiers; Based on the viewpoint to be detected to which the viewpoint task belongs, the detection data is mapped to the matching detection points in the device's 3D model.

8. A cluster scheduling system for collaborative detection of lifting equipment by multiple unmanned aerial vehicles (UAVs), characterized in that, The detection object acquisition module is used to obtain detection object information from the detection request when a detection request is received from the user. The inspection task generation module is used to retrieve a preset 3D model of the equipment based on the inspection object information, and generate an inspection task based on the 3D model of the equipment. The requirements analysis module is used to perform requirements analysis on the detection task and output task requirements information, including detection point information, number of drones, and drone performance information. The strategy generation module is used to generate collaborative detection strategies and collaborative flight routes based on task requirements, retrieve the corresponding drone models and quantities, and control multiple drones to collaboratively execute flight tasks according to the generated collaborative flight routes. The data mapping module is used to receive detection data acquired by the UAV based on a preset communication link, map the detection data to the corresponding detection points in the three-dimensional model of the device, and send it to the user terminal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the cluster scheduling method for collaborative detection of multiple UAVs for lifting equipment as described in any one of claims 1 to 7.

Images