All-weather autonomous inspection method and system based on cluster task dynamic load balancing

Abstract

The invention provides an all-weather autonomous inspection method and system based on cluster task dynamic load balancing, and relates to the technical field of inspection monitoring, and the method comprises the steps: collecting the remaining electric quantity, the positioning position, the task queue length and the sensor health state of an unmanned aerial vehicle cluster in real time through an airborne terminal; fusing the meteorological data and the multi-source environment sensing data, and constructing a dynamic obstacle map and a meteorological influence model; dividing an inspection area into a plurality of sub-areas based on an electronic fence, dynamically adjusting the inspection priority of each sub-area, setting task weights of a water taking head and a raw water pipeline facility, and distributing inspection tasks according to the priorities; based on the remaining power of the unmanned aerial vehicle, the task priority, the current load and the meteorological data, a bipartite graph matching model of the unmanned aerial vehicle and the task is constructed, the matching weight is dynamically calculated, the optimal matching distribution of the unmanned aerial vehicle and the inspection task is completed, and the efficient inspection operation of the unmanned aerial vehicle cluster in the complex environment is realized.

Description

Technical Field

[0001] The present invention relates to the field of patrol inspection and monitoring technology, and in particular to an all-weather autonomous patrol inspection method and system based on cluster task dynamic load balancing. Background Art

[0002] With the development of intelligent technology, drone swarms have been widely used for inspection and monitoring of large-scale infrastructure, especially in the fields of electricity, water supply, transportation, and security. However, existing drone inspection systems still have multiple technical bottlenecks that affect their efficiency and effectiveness. The main problems are reflected in the following aspects:

[0003] 1. Rigid task allocation: Traditional round-robin and static weighting algorithms cannot adapt to changes in the inspection environment in real time, resulting in uneven task loads across drones. Some drones may shut down due to overload, or the resources of other drones may not be fully utilized, resulting in low efficiency.

[0004] 2. Poor environmental adaptability: Existing drone inspection systems often fail to account for changes in real-time weather data. In adverse weather conditions such as rain, fog, and strong winds, existing systems are unable to automatically adjust flight parameters, resulting in inspection mission interruption rates exceeding 40%.

[0005] 3. Low collaborative efficiency: When multiple drones conduct collaborative inspections, there are problems such as path conflicts and improper resource allocation, which lead to redundant flights, increased inspection time, and low efficiency.

[0006] 4. Insufficient data integration: The current system's limited sensor types cannot meet the multi-dimensional monitoring needs of complex inspection scenarios. In particular, traditional sensors cannot provide sufficiently accurate data support for defect identification, affecting inspection quality.

[0007] Therefore, there is an urgent need for a new drone swarm control method that can dynamically adjust task allocation and realize all-weather inspection operations that adapt to various complex environments of water supply plants. Summary of the Invention

[0008] In view of this, in order to address the problem of uneven task distribution and inability to dynamically adapt to the all-weather environment in the all-weather autonomous inspection of large-scale water supply plants, resulting in low efficiency of multi-machine collaboration during inspection, the purpose of the present invention is to propose an all-weather autonomous inspection method and system based on dynamic load balancing of cluster tasks, which realizes efficient inspection operations of drone clusters in complex environments through real-time environmental perception, dynamic task scheduling, collaborative path planning and multi-source data fusion.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] Based on the above objectives, in a first aspect, the present invention provides an all-weather autonomous inspection method based on dynamic load balancing of cluster tasks, comprising the following steps:

[0011] The airborne terminal collects the remaining battery power, positioning position, task queue length, and sensor health status of the drone cluster in real time; integrates meteorological data with multi-source environmental perception data to build a dynamic obstacle map and meteorological impact model;

[0012] Divide the inspection area into multiple sub-areas based on electronic fences, dynamically adjust the inspection priority of each sub-area, set the task weights of water intake heads and raw water pipeline facilities, and assign inspection tasks according to priority;

[0013] Based on the remaining battery power of the drone, task priority, current load, and weather data, a bipartite graph matching model for drones and tasks is constructed, and matching weights are dynamically calculated to achieve the optimal matching allocation between drones and inspection tasks.

[0014] Access to the Meteorological Bureau's real-time API, dynamically adjust flight parameters based on real-time weather data, adjust flight altitude, speed coefficient and sensor mode according to weather conditions, and adaptively plan flight paths based on weather conditions;

[0015] The drone performs inspections according to the assigned tasks and continuously monitors the task execution status. It transmits key alarm data to the ground control center through the 5G slicing network and transmits the inspection data and task execution status back to the monitoring platform for analysis.

[0016] As a further solution of the present invention, the remaining power, positioning position, task queue length and sensor health status of the drone cluster are collected in real time through the airborne terminal, wherein the sensor accuracy is: the remaining power error is ±5%, the positioning error is ±0.5m, and the millimeter wave radar accuracy used is ±2cm.

[0017] As a further solution of the present invention, a bipartite graph matching model of drones and tasks is constructed, and the matching weight is dynamically calculated. The calculation formula is:

[0018]

[0019] Where W is the matching weight between the task and the UAV, which is used to represent the priority of the task assigned to the UAV; α is the weight coefficient of the UAV's remaining power, which indicates the impact of the remaining power on task matching. The value of α is between 0 and 1. The larger the value, the greater the impact of the remaining power on task matching. Remaining power / total power is the ratio of the UAV's current remaining power to the total power, reflecting the UAV's power status; β is the weight coefficient of the current load rate, which indicates the impact of the UAV's current load on task matching.

[0020] The current load rate is the ratio of the current load of the drone to its maximum load capacity, reflecting whether the drone is overloaded. A larger value indicates a heavier load. γ is the weight coefficient of the task priority, indicating the impact of the urgency of the task on task matching. Task priority represents the importance or urgency of the task. δ is the weight coefficient of the meteorological influence factor, indicating the impact of weather conditions on task allocation. The value range of δ is 0.1-0.3. The meteorological influence factor affects the execution of tasks according to the current weather conditions. A global load assessment is performed every 30 seconds. When the load rate of a single machine exceeds the threshold, 30% of the tasks will be migrated to the drone with the lowest load. The task migration process is recorded through a blockchain smart contract, and the task block contains a SHA-256 hash check value to ensure that the data cannot be tampered with.

[0021] As a further solution of the present invention, the calculation formula of the weight coefficient δ of the meteorological impact factor is:

[0022] δ = 0.4 × rainfall intensity score + 0.3 × wind speed score + 0.3 × visibility score

[0023] Among them, rainfall intensity score = rainfall / 10mm / h; wind speed score = wind speed / 15m / s; visibility score = 1-visibility / 1000m.

[0024] As a further solution of the present invention, time-slot spatial channel planning is adopted during task migration, and the migration path is optimized through reinforcement learning algorithm to avoid multi-machine communication interference and ensure that the instruction accessibility rate of the migration process is ≥99.99%.

[0025] As a further solution of the present invention, a flight path is adaptively planned according to meteorological conditions. The flight adjustment strategy for adaptively planning the flight path according to meteorological conditions is:

[0026] Light rain (<10 mm / h): Flight altitude +5 m, speed factor × 0.9, radar obstacle avoidance enabled;

[0027] Heavy fog (visibility <50m): Flight altitude -10m, speed factor ×0.6, switch to infrared imaging;

[0028] Strong winds (>15m / s): Enable wind resistance mode, increase the flight speed factor by 0.5, and disable gimbal stabilization.

[0029] As a further embodiment of the present invention, when adjusting the flight altitude, speed coefficient, and sensor mode according to meteorological conditions, the sensor mode switching includes:

[0030] In rainy and foggy weather, the infrared thermal imaging module and laser methane detector are activated. The detection accuracy of the infrared thermal imaging module is ±0.5°C, and the detection limit of the laser methane detector is 0.1ppm.

[0031] In strong winds, the gimbal stabilization function is disabled, the aircraft switches to low-power mode, and the flight speed is reduced to 50% of the standard value.

[0032] As a further solution of the present invention, the all-weather autonomous inspection method based on cluster task dynamic load balancing also includes a heterogeneous task allocation mechanism; wherein the heterogeneous task allocation mechanism based on dynamic alliance game and blockchain consensus includes:

[0033] The UAV cluster is divided into multiple heterogeneous resource alliances, and each alliance is optimized for resource balance through K-medoids clustering. The clustering conditions are: the matching degree between the remaining power of the UAV and the energy consumption of the mission, the adaptability of the sensor type to the mission requirements, and the meteorological influencing factors.

[0034] As a further solution of the present invention, a blockchain consensus mechanism is used to distribute tasks across alliances, including:

[0035] Automatically trigger task orders through smart contracts and generate inspection task blocks containing encrypted time and space stamps;

[0036] Based on DPoS (Delegated Proof of Stake), task coordination nodes are dynamically elected, and nodes are required to pledge computing power resources as credit endorsement;

[0037] During task migration, cross-regional collaboration is achieved through the shard chain structure, and major failures trigger cross-chain verification, with a processing capacity of up to 1000TPS.

[0038] In a second aspect, the present invention also provides an all-weather autonomous inspection system based on dynamic load balancing of cluster tasks, comprising the following components:

[0039] Airborne terminal: used to collect the remaining power, positioning position, task queue length and sensor health status of the drone cluster in real time; the accuracy of the sensors includes: remaining power error of ±5%, positioning error of ±0.5m, and millimeter wave radar accuracy of ±2cm.

[0040] Environmental Perception Module: This module integrates meteorological data with multi-source environmental perception data to build dynamic obstacle maps and meteorological impact models, providing support for meteorological data, environmental changes, and obstacle information. This module can connect to the Meteorological Bureau's API interface in real time to obtain real-time meteorological data.

[0041] Inspection area division and priority scheduling module: The inspection area is divided into multiple sub-areas based on electronic fences, and the inspection priority of each sub-area is dynamically adjusted; the task weights of water intake heads and raw water pipeline facilities are set, and inspection tasks are allocated according to priority.

[0042] Task allocation and matching module: Based on the remaining power of the drone, task priority, current load and weather data, a bipartite graph matching model between drones and tasks is constructed, matching weights are dynamically calculated, and the optimal matching allocation between tasks and drones is completed.

[0043] Flight parameter adaptive adjustment module: Dynamically adjusts flight parameters according to real-time meteorological data, adjusts flight altitude, speed coefficient and sensor mode, and adaptively plans flight paths according to meteorological conditions.

[0044] Mission execution monitoring module: used for drones to perform inspection missions and monitor mission execution status in real time. It transmits key alarm data to the ground control center through the 5G slicing network, and transmits inspection data and mission execution status back to the monitoring platform for analysis.

[0045] Task migration and load assessment module: A global load assessment is performed every 30 seconds. When the load rate of a single drone exceeds the threshold, 30% of the tasks will be migrated to the drone with the lowest load. The task migration process is recorded through blockchain smart contracts to ensure that the data cannot be tampered with.

[0046] Meteorological impact factor calculation module: used to calculate the weight coefficient δ of the meteorological impact factor. Its calculation formula is based on the rainfall intensity score, wind speed score and visibility score to evaluate the impact of meteorological conditions on task execution.

[0047] Flight Path Optimization: This module adaptively plans flight paths based on weather conditions and adjusts altitude, speed, and sensor modes to ensure the flight path matches mission requirements. This module also adjusts flight strategies based on weather conditions, including adjusting flight parameters for light rain, heavy fog, and strong winds.

[0048] Heterogeneous Task Allocation Mechanism: This module divides the drone cluster into multiple heterogeneous resource alliances based on dynamic alliance game theory and blockchain consensus, and optimizes resource balance through K-medoids clustering. This module uses blockchain consensus to allocate tasks across alliances, ensuring collaborative execution across different alliances.

[0049] Smart Contract and Task Coordination Module: This module automatically triggers task orders, generates inspection task blocks with encrypted time and space stamps, and dynamically elects task coordination nodes based on Delegated Proof of Stake (DPoS). Nodes must pledge computing power as a credit endorsement. During task migration, cross-regional collaboration is achieved through a sharded chain structure, and major failures trigger cross-chain verification.

[0050] Control Center: Used to receive the flight status, mission execution data and alarm information of all drones, monitor and adjust task allocation in real time to ensure efficient completion of tasks.

[0051] Blockchain network module: used to realize encrypted storage of task blocks during task migration and data transmission, ensure the security of task migration and the non-tamperability of data, and ensure the transparency of cross-regional collaboration and task coordination.

[0052] Compared with the existing technology, the present invention proposes an all-weather autonomous inspection method and system based on cluster task dynamic load balancing, which has the following beneficial effects:

[0053] This invention achieves optimal matching between drones and tasks through a bipartite graph matching model based on drone remaining battery life, task priority, load status, and meteorological data. This method dynamically adjusts task allocation based on real-time flight status, environmental conditions, and task requirements, avoiding excessive task concentration or unloaded drone operation. This significantly improves operational efficiency, ensures the timely completion of inspection tasks, and enhances both the efficiency of drone swarm operations and the accuracy of task allocation.

[0054] The system of the present invention can also adaptively adjust flight parameters, including flight altitude, speed coefficient, and sensor mode, based on real-time meteorological data. This adjustment mechanism can effectively cope with adverse weather conditions such as rain, fog, strong winds, and heavy fog. While ensuring flight safety, it ensures that mission execution is not excessively interfered with by the weather, enabling all-weather autonomous inspection operations. Based on the remaining power and load assessment of the drone, the system can monitor and adjust the working status of the drone in real time to avoid situations where the battery is too low or the load is too heavy. By dynamically adjusting task allocation and migrating some tasks to drones with lighter loads through a task migration mechanism when the drone load is too high, the load balance of the entire drone cluster is ensured, thereby improving the sustainability and stability of the inspection mission.

[0055] When a drone's load exceeds a threshold, the system automatically assesses and migrates tasks to less-loaded drones, ensuring continuity and efficient execution. Blockchain smart contracts record the task migration process, ensuring the immutability of task data. Reinforcement learning algorithms optimize the task migration path, avoiding multi-machine communication interference and ensuring high command availability, thus improving the reliability of task migration.

[0056] The present invention also uses an intelligent task priority scheduling mechanism to dynamically adjust the execution order and strategy of tasks according to the urgency of the task, environmental changes, and the working status of the drone. The task allocation and migration mechanism based on blockchain ensures the transparency and security of all task allocation and migration processes. Each task block contains an encrypted time and space stamp, and the task work order is triggered by a smart contract to ensure that the task process data cannot be tampered with, thereby improving the credibility and security of the entire system. The present invention has strong adaptability through the deep integration of environmental perception, meteorological data and task priority scheduling. The system can automatically make decisions and adjust inspection strategies based on changing environmental conditions, task requirements and the operating status of the drone, thereby ensuring that the drone cluster completes tasks efficiently and safely in different complex situations.

[0057] To sum up, the all-weather autonomous inspection method and system based on dynamic load balancing of cluster tasks in the present invention not only significantly improves the inspection efficiency and reliability of drone clusters, but also can flexibly adapt to changing environments and task requirements, providing solid technical guarantees for the intelligent, automated and efficient execution of drone inspection tasks.

[0058] These and other aspects of the present application will be more clearly understood in the following description of the embodiments. It should be understood that the above general description and the following detailed description are merely exemplary and explanatory and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the following briefly introduces the drawings required for the exemplary embodiments or related technical descriptions. The drawings are used to provide a further understanding of the present invention and constitute part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the drawings:

[0060] Figure 1 This is a flow chart of an all-weather autonomous inspection method based on dynamic load balancing of cluster tasks according to an embodiment of the present invention.

[0061] Figure 2 Schematic diagram of sensor mode switching in an all-weather autonomous inspection method based on dynamic load balancing of cluster tasks according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] Below, the present application is further described in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0063] To make the purpose, technical solutions and advantages of the present invention more clearly understood, the following is a further detailed description of the embodiments of the present invention in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0064] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are intended to distinguish two non-identical entities or non-identical parameters with the same name. Therefore, "first" and "second" are used for convenience of expression only and should not be understood as limitations on the embodiments of the present invention. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, other steps or units inherent to a process, method, system, product, or device that includes a series of steps or units.

[0065] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0066] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0067] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0068] In response to the problem of uneven task distribution and inability to dynamically adapt to the all-weather environment in the all-weather autonomous inspection of large-scale water supply plants, which leads to low efficiency of multi-machine collaboration during inspection, the present invention proposes an all-weather autonomous inspection method and system based on dynamic load balancing of cluster tasks. Through real-time environmental perception, dynamic task scheduling, collaborative path planning and multi-source data fusion, efficient inspection operations of drone clusters in complex environments can be achieved.

[0069] See also Figure 1 As shown, an embodiment of the present invention provides an all-weather autonomous inspection method based on cluster task dynamic load balancing, comprising the following steps:

[0070] Step S10: The remaining power, positioning position, task queue length, and sensor health status of the drone cluster are collected in real time through the airborne terminal; meteorological data and multi-source environmental perception data are integrated to build a dynamic obstacle map and meteorological impact model.

[0071] In this step, the remaining power, positioning position, task queue length and sensor health status of the drone cluster are collected in real time through the airborne terminal. The sensor accuracy is: the remaining power error is ±5%, the positioning error is ±0.5m, and the millimeter wave radar accuracy used is ±2cm.

[0072] Step S20: Divide the inspection area into multiple sub-areas based on the electronic fence, dynamically adjust the inspection priority of each sub-area, set the task weights of the water intake head and raw water pipeline facilities, and allocate inspection tasks according to priority.

[0073] In this step, a bipartite graph matching model of drones and tasks is constructed, and the matching weight is dynamically calculated. The calculation formula is:

[0074]

[0075] Where W is the matching weight between the task and the UAV, which is used to represent the priority of the task assigned to the UAV; α is the weight coefficient of the UAV's remaining power, which indicates the impact of the remaining power on task matching. The value of α is between 0 and 1. The larger the value, the greater the impact of the remaining power on task matching. Remaining power / total power is the ratio of the UAV's current remaining power to the total power, reflecting the UAV's power status; β is the weight coefficient of the current load rate, which indicates the impact of the UAV's current load on task matching.

[0076] The current load rate is the ratio of the current load of the drone to its maximum load capacity, reflecting whether the drone is overloaded. A larger value indicates a heavier load. γ is the weight coefficient of the task priority, indicating the impact of the urgency of the task on task matching. Task priority represents the importance or urgency of the task. δ is the weight coefficient of the meteorological influence factor, indicating the impact of weather conditions on task allocation. The value range of δ is 0.1-0.3. The meteorological influence factor affects the execution of tasks according to the current weather conditions. A global load assessment is performed every 30 seconds. When the load rate of a single machine exceeds the threshold, 30% of the tasks will be migrated to the drone with the lowest load. The task migration process is recorded through a blockchain smart contract, and the task block contains a SHA-256 hash check value to ensure that the data cannot be tampered with.

[0077] In this embodiment, the calculation formula of the weight coefficient δ of the meteorological impact factor is:

[0078] δ = 0.4 × rainfall intensity score + 0.3 × wind speed score + 0.3 × visibility score

[0079] Among them, rainfall intensity score = rainfall / 10mm / h; wind speed score = wind speed / 15m / s; visibility score = 1-visibility / 1000m.

[0080] Among them, time-slot airspace channel planning is adopted during task migration, and the migration path is optimized through reinforcement learning algorithm to avoid multi-machine communication interference and ensure that the instruction accessibility rate of the migration process is ≥99.99%.

[0081] Step S30: Based on the remaining power of the drone, task priority, current load and weather data, a bipartite graph matching model between the drone and the task is constructed, and the matching weight is dynamically calculated to complete the optimal matching allocation between the drone and the inspection task.

[0082] Step S40: Access the real-time API of the Meteorological Bureau, dynamically adjust flight parameters according to real-time meteorological data, adjust flight altitude, speed coefficient and sensor mode according to meteorological conditions, and adaptively plan the flight path according to meteorological conditions.

[0083] In this step, the flight path is adaptively planned according to the weather conditions. The flight adjustment strategy for adaptively planning the flight path according to the weather conditions is:

[0084] Light rain (<10 mm / h): Flight altitude +5 m, speed factor × 0.9, radar obstacle avoidance enabled;

[0085] Heavy fog (visibility <50m): Flight altitude -10m, speed factor ×0.6, switch to infrared imaging;

[0086] Strong winds (>15m / s): Enable wind resistance mode, increase the flight speed factor by 0.5, and disable gimbal stabilization.

[0087] When adjusting the flight altitude, speed factor, and sensor mode based on weather conditions, the sensor mode switching includes:

[0088] Step S401: In rainy and foggy weather, the infrared thermal imaging module and the laser methane detector are activated, wherein the detection accuracy of the infrared thermal imaging module is ±0.5°C, and the detection limit of the laser methane detector is 0.1ppm;

[0089] Step S402: In a strong wind environment, the gimbal stabilization function is turned off, the system switches to low power consumption mode, and the flight speed is reduced to 50% of the standard value.

[0090] Step S50: The drone performs inspections according to the assigned tasks and continuously monitors the task execution status. It transmits key alarm data to the ground control center through the 5G slicing network, and transmits the inspection data and task execution status back to the monitoring platform for analysis.

[0091] In some embodiments, the all-weather autonomous inspection method based on cluster task dynamic load balancing also includes a heterogeneous task allocation mechanism; wherein the heterogeneous task allocation mechanism based on dynamic alliance game and blockchain consensus includes:

[0092] The UAV cluster is divided into multiple heterogeneous resource alliances, and each alliance is optimized for resource balance through K-medoids clustering. The clustering conditions are: the matching degree between the remaining power of the UAV and the energy consumption of the mission, the adaptability of the sensor type to the mission requirements, and the meteorological influencing factors.

[0093] Among them, the blockchain consensus mechanism is used for cross-alliance task allocation, including:

[0094] Automatically trigger task orders through smart contracts and generate inspection task blocks containing encrypted time and space stamps;

[0095] Based on DPoS (Delegated Proof of Stake), task coordination nodes are dynamically elected, and nodes are required to pledge computing power resources as credit endorsement;

[0096] During task migration, cross-regional collaboration is achieved through the shard chain structure, and major failures trigger cross-chain verification, with a processing capacity of up to 1000TPS.

[0097] This invention achieves optimal matching between drones and tasks through a bipartite graph matching model based on drone remaining battery life, task priority, load status, and meteorological data. This method dynamically adjusts task allocation based on real-time flight status, environmental conditions, and task requirements, avoiding excessive task concentration or unloaded drone operation. This significantly improves operational efficiency, ensures the timely completion of inspection tasks, and enhances both the efficiency of drone swarm operations and the accuracy of task allocation.

[0098] The embodiment of the present invention further provides an all-weather autonomous inspection system based on cluster task dynamic load balancing, comprising the following components:

[0099] On the airborne terminal: used to collect the remaining power, positioning position, task queue length and sensor health status of the drone cluster in real time; the accuracy of the sensors includes: remaining power error of ±5%, positioning error of ±0.5m, and millimeter wave radar accuracy of ±2cm.

[0100] Environmental Perception Module: This module integrates meteorological data with multi-source environmental perception data to build dynamic obstacle maps and meteorological impact models, providing support for meteorological data, environmental changes, and obstacle information. This module can connect to the Meteorological Bureau's API interface in real time to obtain real-time meteorological data.

[0101] Inspection area division and priority scheduling module: The inspection area is divided into multiple sub-areas based on electronic fences, and the inspection priority of each sub-area is dynamically adjusted; the task weights of water intake heads and raw water pipeline facilities are set, and inspection tasks are allocated according to priority.

[0102] Task allocation and matching module: Based on the remaining power of the drone, task priority, current load and weather data, a bipartite graph matching model between drones and tasks is constructed, matching weights are dynamically calculated, and the optimal matching allocation between tasks and drones is completed.

[0103] Flight parameter adaptive adjustment module: Dynamically adjusts flight parameters according to real-time meteorological data, adjusts flight altitude, speed coefficient and sensor mode, and adaptively plans flight paths according to meteorological conditions.

[0104] Mission execution monitoring module: used for drones to perform inspection missions and monitor mission execution status in real time. It transmits key alarm data to the ground control center through the 5G slicing network, and transmits inspection data and mission execution status back to the monitoring platform for analysis.

[0105] Task migration and load assessment module: A global load assessment is performed every 30 seconds. When the load rate of a single drone exceeds the threshold, 30% of the tasks will be migrated to the drone with the lowest load. The task migration process is recorded through blockchain smart contracts to ensure that the data cannot be tampered with.

[0106] Meteorological impact factor calculation module: used to calculate the weight coefficient δ of the meteorological impact factor. Its calculation formula is based on the rainfall intensity score, wind speed score and visibility score to evaluate the impact of meteorological conditions on task execution.

[0107] Flight Path Optimization: This module adaptively plans flight paths based on weather conditions and adjusts altitude, speed, and sensor modes to ensure the flight path matches mission requirements. This module also adjusts flight strategies based on weather conditions, including adjusting flight parameters for light rain, heavy fog, and strong winds.

[0108] Heterogeneous Task Allocation Mechanism: This module divides the drone cluster into multiple heterogeneous resource alliances based on dynamic alliance game theory and blockchain consensus, and optimizes resource balance through K-medoids clustering. This module uses blockchain consensus to allocate tasks across alliances, ensuring collaborative execution across different alliances.

[0109] Smart Contract and Task Coordination Module: This module automatically triggers task orders, generates inspection task blocks with encrypted time and space stamps, and dynamically elects task coordination nodes based on Delegated Proof of Stake (DPoS). Nodes must pledge computing power as a credit endorsement. During task migration, cross-regional collaboration is achieved through a sharded chain structure, and major failures trigger cross-chain verification.

[0110] Control Center: Used to receive the flight status, mission execution data and alarm information of all drones, monitor and adjust task allocation in real time to ensure efficient completion of tasks.

[0111] Blockchain network module: used to realize encrypted storage of task blocks during task migration and data transmission, ensure the security of task migration and the non-tamperability of data, and ensure the transparency of cross-regional collaboration and task coordination.

[0112] The system of the present invention can adaptively adjust flight parameters, including flight altitude, speed coefficient, and sensor mode, based on real-time meteorological data. This adjustment mechanism can effectively cope with adverse weather conditions such as rain, fog, strong winds, and heavy fog. While ensuring flight safety, it ensures that mission execution is not excessively interfered with by the weather, thus achieving all-weather autonomous inspection operations. Based on the remaining power and load assessment of the drone, the system can monitor and adjust the working status of the drone in real time to avoid situations where the battery is too low or the load is too heavy. By dynamically adjusting task allocation and migrating some tasks to drones with lighter loads through the task migration mechanism when the drone load is too high, the load balance of the entire drone cluster is ensured, thereby improving the sustainability and stability of the inspection tasks.

[0113] When a drone's load exceeds a threshold, the system automatically assesses and migrates tasks to less-loaded drones, ensuring continuity and efficient execution. Blockchain smart contracts record the task migration process, ensuring the immutability of task data. Reinforcement learning algorithms optimize the task migration path, avoiding multi-machine communication interference and ensuring high command availability, thus improving the reliability of task migration.

[0114] The present invention also uses an intelligent task priority scheduling mechanism to dynamically adjust the execution order and strategy of tasks according to the urgency of the task, environmental changes, and the working status of the drone. The task allocation and migration mechanism based on blockchain ensures the transparency and security of all task allocation and migration processes. Each task block contains an encrypted time and space stamp, and the task work order is triggered by a smart contract to ensure that the task process data cannot be tampered with, thereby improving the credibility and security of the entire system. The present invention has strong adaptability through the deep integration of environmental perception, meteorological data and task priority scheduling. The system can automatically make decisions and adjust inspection strategies based on changing environmental conditions, task requirements and the operating status of the drone, thereby ensuring that the drone cluster completes tasks efficiently and safely in different complex situations.

[0115] To sum up, the all-weather autonomous inspection method and system based on dynamic load balancing of cluster tasks in the present invention not only significantly improves the inspection efficiency and reliability of drone clusters, but also can flexibly adapt to changing environments and task requirements, providing solid technical guarantees for the intelligent, automated and efficient execution of drone inspection tasks.

[0116] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope of the embodiments disclosed in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any particular order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless expressly limited to the singular.

[0117] It should be understood that, as used herein, the singular form "a" or "an" is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations of one or more of the items listed in association. The serial numbers of the embodiments disclosed in the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0118] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples. Within the spirit of the embodiments of the present invention, the technical features of the above embodiments or different embodiments may be combined, and there are many other variations of different aspects of the above embodiments of the present invention, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of the embodiments of the present invention.

Claims

1. An all-weather autonomous inspection method based on cluster task dynamic load balancing, characterized in that: The following steps are involved: The airborne terminal collects the remaining battery power, positioning position, task queue length, and sensor health status of the drone cluster in real time; integrates meteorological data with multi-source environmental perception data to build a dynamic obstacle map and meteorological impact model; Divide the inspection area into multiple sub-areas based on electronic fences, dynamically adjust the inspection priority of each sub-area, set the task weights of water intake heads and raw water pipeline facilities, and assign inspection tasks according to priority; Based on the remaining battery power of the drone, task priority, current load, and weather data, a bipartite graph matching model for drones and tasks is constructed, and matching weights are dynamically calculated to achieve the optimal matching allocation between drones and inspection tasks. Access to the Meteorological Bureau's real-time API, dynamically adjust flight parameters based on real-time weather data, adjust flight altitude, speed coefficient and sensor mode according to weather conditions, and adaptively plan flight paths based on weather conditions; The drone performs inspections according to the assigned tasks and continuously monitors the task execution status. It transmits key alarm data to the ground control center through the 5G slicing network and transmits the inspection data and task execution status back to the monitoring platform for analysis.

2. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 1, characterized in that: The airborne terminal collects the remaining power, positioning position, task queue length and sensor health status of the drone cluster in real time. The sensor accuracy is: the remaining power error is ±5%, the positioning error is ±0.5m, and the millimeter wave radar accuracy is ±2cm.

3. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 1, characterized in that: Construct a bipartite graph matching model for drones and tasks, and dynamically calculate the matching weight. The calculation formula is: Where W is the matching weight between tasks and drones, α is the weight coefficient of the drone's remaining battery power, β is the weight coefficient of the current load rate, γ is the weight coefficient of the task priority, and δ is the weight coefficient of the meteorological influence factor, with the value range of δ being 0.1-0.

3. A global load assessment is performed every 30 seconds. When the load rate of a single drone exceeds the threshold, 30% of the tasks will be migrated to the drone with the lowest load. The task migration process is recorded through a blockchain smart contract, and the task block contains a SHA-256 hash check value to ensure that the data cannot be tampered with.

4. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 3, characterized in that: The calculation formula of the weight coefficient δ of the meteorological impact factor is: δ = 0.4 × rainfall intensity score + 0.3 × wind speed score + 0.3 × visibility score Among them, rainfall intensity score = rainfall / 10mm / h; wind speed score = wind speed / 15m / s; visibility score = 1-visibility / 1000m.

5. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 4, characterized in that: Time-slot spatial channel planning is adopted during task migration, and the migration path is optimized through reinforcement learning algorithm. The instruction accessibility rate of the migration process is ≥99.99%.

6. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 1, characterized in that: The flight path is adaptively planned according to weather conditions. The flight adjustment strategy for adaptively planning the flight path according to weather conditions is as follows: Light rain, <10 mm / h: Flight altitude +5 m, speed factor × 0.9, radar obstacle avoidance enabled; Heavy fog, visibility <50m: Flight altitude -10m, speed factor ×0.6, switch to infrared imaging; Strong winds (>15m / s): Enable wind resistance mode, multiply the flight speed coefficient by 0.5, and disable gimbal stabilization.

7. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 6, characterized in that: When adjusting the flight altitude, speed factor, and sensor mode according to weather conditions, the sensor mode switching includes: In rainy and foggy weather, the infrared thermal imaging module and laser methane detector are activated. The detection accuracy of the infrared thermal imaging module is ±0.5°C, and the detection limit of the laser methane detector is 0.1ppm. In strong winds, the gimbal stabilization function is disabled, the aircraft switches to low-power mode, and the flight speed is reduced to 50% of the standard value.

8. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to any one of claims 1 to 7, characterized in that: The all-weather autonomous inspection method based on cluster task dynamic load balancing also includes a heterogeneous task allocation mechanism; wherein the heterogeneous task allocation mechanism based on dynamic alliance game and blockchain consensus includes: The UAV cluster is divided into multiple heterogeneous resource alliances, and each alliance is optimized for resource balance through K-medoids clustering. The clustering conditions are: the matching degree between the remaining power of the UAV and the energy consumption of the mission, the adaptability of the sensor type to the mission requirements, and the meteorological influencing factors.

9. The all-weather autonomous inspection method based on cluster task dynamic load balancing according to claim 8, characterized in that: Adopt blockchain consensus mechanism for cross-alliance task allocation, including: Automatically trigger task orders through smart contracts and generate inspection task blocks containing encrypted time and space stamps; Based on DPoS dynamic election task coordination nodes, nodes need to pledge computing power resources as credit endorsement; During task migration, cross-regional collaboration is achieved through the shard chain structure, and major failures trigger cross-chain verification, with a processing capacity of up to 1000TPS.

10. An all-weather autonomous inspection system based on cluster task dynamic load balancing, characterized in that: The system is used to perform the all-weather autonomous inspection method based on cluster task dynamic load balancing according to any one of claims 1 to 9, and comprises the following components: Airborne terminal: used to collect real-time data on the remaining battery power, positioning position, task queue length, and sensor health status of the drone cluster; Environmental perception module: used to integrate meteorological data with multi-source environmental perception data, build dynamic obstacle maps and meteorological impact models, and provide support for meteorological data, environmental changes, and obstacle information. This environmental perception module is connected to the Meteorological Bureau's API interface in real time to obtain real-time meteorological data. Inspection area division and priority scheduling module: The inspection area is divided into multiple sub-areas based on electronic fences, and the inspection priority of each sub-area is dynamically adjusted; the task weights of water intake heads and raw water pipeline facilities are set, and inspection tasks are allocated according to priority; Task allocation and matching module: Based on the remaining battery power of the drone, task priority, current load and weather data, a bipartite graph matching model is constructed between the drone and the task, matching weights are dynamically calculated, and the optimal matching allocation between the task and the drone is completed; Flight parameter adaptive adjustment module: dynamically adjusts flight parameters based on real-time weather data, adjusts flight altitude, speed coefficient and sensor mode, and adaptively plans flight paths based on weather conditions; Mission execution monitoring module: used for drones to perform inspection missions and monitor mission execution status in real time. It transmits key alarm data to the ground control center through the 5G slicing network, and transmits inspection data and mission execution status back to the monitoring platform for analysis.

Images