An intelligent control system for UAV swarms

Abstract

The present invention relates to the field of UAV intelligent control technology, and particularly to a UAV swarm intelligent control system, which includes a data acquisition module, a first determination module, a second determination module, a determination module, a control module, and an adjustment module. Through a highly integrated perception and decision-making module, the present invention can quickly respond to complex climate changes in polar regions, accurately locate and adjust the UAV mission area. The system can efficiently identify water accumulation grids, temporary grids, and observation grids by collecting spectral reflectance, temperature, and crack width data in real time and combining intelligent analysis, so as to achieve precise task allocation and scheduling. In practical applications, the system can not only flexibly adjust the reflectance threshold according to the environmental changes of the grids, but also optimize the allocation of UAVs among multiple grids through comparison and fluctuation calculation, effectively solving the problem of low response speed of the system in the face of emergencies due to relying on a single control strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of unmanned aerial vehicles, and particularly to an intelligent control system for an unmanned aerial vehicle cluster. Background Art

[0002] With the rapid development of unmanned aerial vehicle technology and the continuous increase in application requirements, especially in complex environments such as polar regions and disaster areas, unmanned aerial vehicle cluster systems are widely used to perform tasks such as monitoring, rescue, and exploration. These tasks usually require the system to be able to efficiently and accurately complete large-scale real-time data collection and mission planning. However, existing systems often have problems such as untimely response and uneven resource utilization when facing large-scale cluster scheduling, and it is difficult to cope with the dynamic changes in complex environments. Therefore, improving the intelligent level and autonomous decision-making ability of unmanned aerial vehicle clusters has become the key to solving the above problems.

[0003] The patent document with the publication number CN119002518A discloses a control method for a cluster unmanned aerial vehicle system based on DPPO deep reinforcement learning. The method includes: 1) Construction of a three-level structure of the cluster unmanned aerial vehicle system: constructing a three-level structure of cluster - formation - single aircraft, analyzing the collaborative relationship and topological configuration characteristics between formations within the cluster and between unmanned aerial vehicles within the formation, and forming a three-level dynamic control system; 2) Multi-dimensional situation awareness and feature extraction: designing a deep neural network model, processing task area features, formation coverage features, unmanned aerial vehicle position features, and damage area features through a ResNet module, extracting the temporal features of high-dimensional information using an LSTM module, and using an Attention mechanism to focus on intra-formation and cross-level interactions to obtain multi-dimensional situation information; 3) Generation of a control strategy based on Distributed Proximal Policy Optimization (DPPO): applying the Advantage Actor-Critic policy gradient algorithm, performing asynchronous updates in combination with the experience replay technique, maximizing the expected policy reward, and training and updating the neural network using TD(λ), V-trace, and UPGO algorithms to generate a dynamic control strategy for the cluster unmanned aerial vehicle system;

[0004] 4) Design and training of a coalition learning mechanism: through designing coalition games and a virtual self-learning mechanism, including three types of agents: a master agent, a master explorer, and a coalition explorer, performing distributed learning and training to optimize the control strategy of the cluster unmanned aerial vehicle system; 5) Execution of the dynamic control strategy: during the process of the unmanned aerial vehicle cluster performing tasks, based on the generated control strategy, the formation topology structure and the unmanned aerial vehicle path planning are adjusted in real time to ensure the continuity of task execution and the optimization of system performance.

[0005] It can be seen that the control method of the cluster UAV system based on DPPO deep reinforcement learning has the following problems: This method relies on a large amount of high-dimensional data and historical data for training in multi-dimensional situation awareness and feature extraction, and updates the control strategy through reinforcement learning. This requires a large amount of training data and is affected by factors such as data quality, sampling frequency, and scene complexity, resulting in differences in the stability of model performance in specific environments; Reinforcement learning algorithms such as DPPO often encounter problems such as slow convergence speed and local optimal solutions during the training process. Although various algorithms are used for optimization (such as TD(λ), V-trace, UPGO), in complex dynamic environments, the training effect of reinforcement learning is still uncertain, resulting in too long training time or unstable policy performance; This method generates a dynamic control strategy through a deep learning model, and the response speed to the system in case of emergency is slow. Especially when the UAV cluster faces emergencies or environmental changes, it will delay adjustment and affect the execution efficiency and continuity of tasks. Summary of the Invention

[0006] To this end, the present invention provides an intelligent control system for a UAV cluster, which is used to overcome the problem of low response speed of the existing technology when the system faces emergencies due to relying on a single control strategy through a dynamic monitoring and adjustment mechanism based on a variety of real-time data.

[0007] The polar region is an ice layer area with multiple cracks. Due to long-term climate change, ice sheet sliding, and other natural factors on the surface of the ice sheet in this region, cracks of different sizes and widths have been formed. The changes in these cracks not only directly affect the stability of the ice sheet, but also have an impact on the local ecological environment and global sea level changes. In this environment, it is particularly crucial to use a UAV cluster for efficient and accurate real-time monitoring.

[0008] To achieve the above object, the present invention provides an intelligent control system for a UAV cluster, including:

[0009] A data acquisition module, which is used to acquire the real-time spectral reflectance, real-time temperature, and real-time ice sheet crack width collected by a preset number of UAVs in each to-be-determined grid within the polar region divided by a grid.

[0010] A first determination module, which is connected to the data acquisition module and is used to determine a number of water accumulation grids and non-water accumulation grids according to the real-time spectral reflectance and a preset reflectance threshold.

[0011] A second determination module, which is respectively connected to the first determination module and the data acquisition module, and is used to determine a number of temporary grids according to the real-time temperature and the real-time ice sheet crack width of each of the water accumulation grids.

[0012] A determination module, which is respectively connected to the second determination module and the data acquisition module, and is used to determine a number of observation grids according to the real-time spectral reflectance and the real-time ice sheet crack width of any two adjacent temporary grids;

[0013] A control module, which is connected to the determination module, and is used to control a number of the drones to move from the non-ponding grids to the observation grids according to the number of grids, grid positions of the observation grids, and the preset number;

[0014] An adjustment module, which is respectively connected to the data acquisition module, the control module, and the first determination module, and is used to adjust the preset reflectance threshold based on the temporary grids determined after scheduling a number of the drones.

[0015] Further, the first determination module includes:

[0016] A reflectance comparison unit, which is used to compare the real-time spectral reflectance and the preset reflectance threshold to form a reflectance comparison result;

[0017] A first determination unit, which is connected to the reflectance comparison unit, and is used to determine that the grid to be determined is the ponding grid when the reflectance comparison result is that the real-time spectral reflectance is greater than the preset reflectance threshold, and to determine that the grid to be determined is the non-ponding grid when the reflectance comparison result is that the real-time spectral reflectance is less than or equal to the preset reflectance threshold.

[0018] Further, the second determination module includes:

[0019] A temperature fluctuation calculation unit, which is used to calculate the standard deviation of the real-time temperature within a preset temporary determination duration to form a temperature fluctuation value;

[0020] A first width fluctuation calculation unit, which is used to calculate the standard deviation of the real-time ice sheet crack width within the preset temporary determination duration to form a first width fluctuation value;

[0021] A second determination unit, which is respectively connected to the temperature fluctuation calculation unit and the first width fluctuation calculation unit, and is used to determine a number of the temporary grids according to the temperature fluctuation value and the width fluctuation value.

[0022] Further, the second determination unit includes:

[0023] A temperature fluctuation curve drawing sub-unit, which is used to draw a change curve of the temperature fluctuation value to form a temperature fluctuation curve;

[0024] A first width fluctuation curve drawing sub-unit, which is used to draw a change curve of the first width fluctuation value to form a first width fluctuation curve;

[0025] A consistency calculation subunit, which is respectively connected to the temperature fluctuation curve plotting subunit and the width fluctuation curve plotting subunit, and is used to calculate the cosine similarity between the temperature fluctuation curve and the width fluctuation curve to form a change consistency;

[0026] A temporary determination subunit, which is connected to the consistency calculation subunit, and is used to determine that the waterlogged grid is a temporary grid when the change consistency is less than a preset consistency threshold.

[0027] Further, the determination module includes:

[0028] A reflectance fluctuation calculation unit, which is used to calculate the standard deviation of the real-time spectral reflectance of a single temporary grid within a preset determination duration to form a reflectance fluctuation value;

[0029] A second width fluctuation calculation unit, which is used to calculate the standard deviation of the real-time ice sheet crack width of a single temporary grid within the preset determination duration to form a second width fluctuation value;

[0030] A synchronization degree calculation unit, which is respectively connected to the reflectance fluctuation calculation unit and the second width fluctuation calculation unit, and is used to calculate the synchronization degree according to the reflectance fluctuation value and the second width fluctuation value;

[0031] A determination unit, which is connected to the synchronization degree calculation unit, and is used to determine a number of observation grids according to the synchronization degree of any two adjacent temporary grids.

[0032] Further, the synchronization degree calculation unit includes:

[0033] A reflectance fluctuation curve plotting subunit, which is used to plot the change curve of the reflectance fluctuation value to form a reflectance fluctuation curve;

[0034] A second width fluctuation curve plotting subunit, which is used to plot the change curve of the second width fluctuation value to form a second width fluctuation curve;

[0035] A synchronization degree calculation subunit, which is respectively connected to the reflectance fluctuation curve plotting subunit and the second width fluctuation curve plotting subunit, and is used to calculate the cosine similarity between the reflectance fluctuation curve and the second width fluctuation curve to form a synchronization degree.

[0036] Further, the determination unit includes:

[0037] A synchronization deviation calculation subunit, which is used to calculate the relative deviation of the synchronization degree of any two adjacent temporary grids to form a synchronization deviation;

[0038] A determination sub-unit, which is connected to the synchronization deviation calculation sub-unit, and is used to determine that the temporary grid is the observation grid when the synchronization deviation is greater than a preset synchronization deviation threshold.

[0039] Further, the control module includes:

[0040] An observation ratio calculation unit, which is used to calculate the ratio of the number of grids in the observation grid to the number of all grids to be determined, and form an observation ratio;

[0041] A control unit, which is connected to the observation ratio calculation unit, and is used to control a number of the drones to move from the non-flooded grids to the observation grid according to the observation ratio and the preset quantity.

[0042] Further, the control unit includes:

[0043] A distribution degree calculation sub-unit, which is used to calculate the standard deviation of the distances from all the grid positions to a preset center point when the observation ratio is greater than a preset ratio threshold, and form a distribution degree;

[0044] A scheduling quantity calculation sub-unit, which is connected to the distribution degree calculation sub-unit, and is used to calculate the total scheduling quantity according to the relative deviation between the distribution degree and the preset distribution degree threshold, a preset scheduling coefficient, the number of grids in the observation grid, and the preset quantity when the distribution degree is greater than a preset distribution degree threshold;

[0045] A control sub-unit, which is connected to the scheduling quantity calculation sub-unit, and is used to control the drones with the total scheduling quantity to move to each of the observation grids on average from the non-flooded grids.

[0046] Further, the adjustment module includes:

[0047] A grid fluctuation calculation unit, which is used to calculate the standard deviation of the number of grids in the temporary grid within a preset adjustment duration, and form a grid number fluctuation value;

[0048] An adjustment unit, which is connected to the grid fluctuation calculation unit, and is used to increase the preset reflectivity threshold according to the relative deviation between the grid number fluctuation value and the preset number fluctuation threshold and a preset adjustment coefficient when the grid number fluctuation value is greater than a preset number fluctuation threshold.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows. Through a highly integrated perception and decision-making module, it can quickly respond to complex climate changes in polar regions, accurately locate and adjust the UAV mission area. By collecting spectral reflectance, temperature, and crack width data in real time and combining intelligent analysis, the system can efficiently identify water accumulation grids, temporary grids, and observation grids, achieving precise task allocation and scheduling. In practical applications, the system can not only flexibly adjust the reflectance threshold according to the environmental changes of the grids, but also optimize the allocation of UAVs among multiple grids through comparison and fluctuation calculation, avoiding resource waste and improving the collection efficiency. In addition, the adaptive adjustment mechanism of the system can quickly adjust the number and distribution of UAVs according to real-time data fluctuations, ensuring an efficient working state in various extreme environments. Overall, the system significantly improves the task execution ability, response speed, and data collection accuracy of UAV swarms in polar complex environments, effectively reducing the operation cost and enhancing the comprehensiveness and accuracy of polar environmental monitoring, and effectively solving the problem of low response speed of the system when facing emergencies due to relying on a single control strategy.

[0050] Furthermore, by comparing the real-time spectral reflectance with the preset reflectance threshold, it is possible to accurately and quickly determine whether there is water accumulation in each grid, reducing the complexity of manual judgment and improving the efficiency and accuracy of data processing.

[0051] Furthermore, by calculating the volatility of temperature and crack width, areas with unstable characteristics can be identified, providing a scientific basis for the determination of temporary grids, thereby enhancing the accurate identification ability of water accumulation areas. By real-time monitoring the fluctuations of temperature and crack width, the accuracy of UAV scheduling can be ensured, and subsequent environmental monitoring and data collection can be further optimized, improving the response speed and flexibility of the system, and contributing to enhancing the reliability and accuracy of water accumulation identification in polar regions.

[0052] Furthermore, by analyzing the fluctuation consistency of temperature and crack width, temporary grids can be effectively identified, which helps to further optimize the UAV scheduling and observation strategies. By analyzing the consistency of the changes in temperature and crack width, it is possible to accurately judge areas with unstable or abnormal changes, avoiding misjudgment and improving the adaptability and processing efficiency of the system in complex environments.

[0053] Furthermore, by calculating the synchronization degree of the spectral reflectance and crack width fluctuation values of the temporary grids, grids showing similar change trends can be effectively identified, ensuring more accurate selection of observation grids, reducing the interference of errors, and improving the accuracy and reliability of overall grid division and monitoring.

[0054] Furthermore, through the calculation of the synchronization degree, it is possible to effectively identify whether the spectral reflectance and crack width change trends of adjacent grids are consistent within the same time period, thereby providing a basis for further screening of the temporary grids and improving the accuracy of grid determination. Using cosine similarity to calculate the synchronization degree can eliminate misjudgments caused by instantaneous fluctuations and ensure the stability and reliability of the determination.

[0055] Furthermore, by comparing the synchronization deviation with the preset synchronization deviation threshold, it is possible to accurately judge the synchronization degree difference between temporary grids, ensuring that only when the changes are significantly inconsistent will the temporary grids be determined as observation grids. This can effectively filter out grids that do not meet the observation conditions, improve the accuracy of data analysis, and optimize the subsequent UAV scheduling and regional monitoring efficiency.

[0056] Furthermore, through the calculation of the observation ratio and the reasonable scheduling of UAVs, it is possible to ensure that the observation grids are covered by enough UAVs, avoid resource waste, optimize the allocation of UAVs, and improve the efficiency and accuracy of polar region monitoring.

[0057] Furthermore, by dynamically adjusting the scheduling quantity and UAV allocation, it is possible to efficiently cover the target area, avoid over-concentration or sparse distribution, and improve the working efficiency of UAVs in polar region monitoring and the comprehensiveness of data collection.

[0058] Furthermore, by dynamically adjusting the reflectance threshold, it is possible to avoid over-reliance on unstable grid data, improve the robustness and adaptability of the system, ensure that the system can respond flexibly under large environmental changes, and optimize the UAV task allocation and resource scheduling. Description of the Drawings

[0059] Figure 1 It is a schematic diagram of the UAV swarm intelligent control system of this embodiment;

[0060] Figure 2 It is a determination logic diagram of the first determination unit of this embodiment for determining waterlogged grids and non-waterlogged grids;

[0061] Figure 3 It is a determination logic diagram of the temporary determination subunit of this embodiment for determining temporary grids;

[0062] Figure 4 It is a determination logic diagram of the determination subunit of this embodiment for determining observation grids. Detailed Embodiment

[0063] In order to make the purpose and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0064] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principle of the present invention and do not limit the protection scope of the present invention.

[0065] Please refer to Figure 1 as shown, which is a schematic diagram of the intelligent control system of the UAV cluster in this embodiment;

[0066] This embodiment provides an intelligent control system for a UAV cluster, including:

[0067] a data acquisition module for acquiring the real-time spectral reflectance, real-time temperature, and real-time ice sheet crack width collected by a preset number of UAVs in each to-be-determined grid in the polar region divided by grids;

[0068] a first determination module connected to the data acquisition module for determining a number of water accumulation grids and non-water accumulation grids according to the real-time spectral reflectance and a preset reflectance threshold;

[0069] a second determination module respectively connected to the first determination module and the data acquisition module for determining a number of temporary grids according to the real-time temperature and the real-time ice sheet crack width of each water accumulation grid;

[0070] a determination module respectively connected to the second determination module and the data acquisition module for determining a number of observation grids according to the real-time spectral reflectance and the real-time ice sheet crack width of any two adjacent temporary grids;

[0071] a control module connected to the determination module for controlling a number of the UAVs to move from the non-water accumulation grids to the observation grids according to the number of grids, grid positions, and the preset number of the observation grids;

[0072] an adjustment module respectively connected to the data acquisition module, the control module, and the first determination module for adjusting the preset reflectance threshold based on the temporary grids determined after scheduling a number of the UAVs.

[0073] The data acquisition module simultaneously collects data through multiple drones within each grid area and ensures the representativeness and accuracy of the acquired data through averaging. Specifically, each drone is equipped with a spectral reflectance sensor, a temperature sensor, and a crack width measurement device, which continuously collect data at different positions within the same area. Through real-time data transmission, the module aggregates the acquisition results of multiple drones within each grid, calculates the average values of the spectral reflectance, temperature, and ice sheet crack width within each grid area. These average values reflect the overall situation of the environmental characteristics within the grid and serve as the basis for subsequent analysis and decision-making, thus ensuring the comprehensiveness and high precision of the data.

[0074] The preset number refers to the number of drones participating in data acquisition within each grid area, which depends on the size of the area, the complexity of the environment, and the accuracy requirements of data acquisition. According to the actual situation, the preset number is usually between 5 and 20. In this embodiment, it is set to 10 to ensure a relatively high data acquisition density and representativeness, and the number of drones is reasonable, avoiding resource waste.

[0075] The preset reflectance threshold refers to the spectral reflectance value used to distinguish between waterlogged and non-waterlogged grids, which depends on the reflection characteristics of different types of ground or surfaces in a specific spectral band. It is usually set between 0.3 and 0.5. In this embodiment, it is set to 0.4, which can better identify waterlogged areas and non-waterlogged areas, avoiding misjudgment and improving the accuracy of the system.

[0076] Through a multi-level data acquisition and analysis process, intelligent division and dynamic scheduling of grids are carried out in the polar region. First, the data acquisition module collects spectral reflectance, temperature, and crack width data in real time for subsequent analysis. The first determination module determines waterlogged and non-waterlogged grids based on the spectral reflectance and the preset threshold. The second determination module further analyzes the temperature and crack width of the waterlogged grids to determine the temporary grids. Then, the determination module determines the final observation grids based on the synchrony of adjacent grids. The control module schedules the drones to move from non-waterlogged grids to the observation grids based on the number, position of the observation grids, and the preset number. Finally, the adjustment module dynamically adjusts the preset reflectance threshold according to the drone scheduling situation, thereby improving the adaptability and accuracy of the system.

[0077] Through a highly integrated perception and decision-making module, it can quickly respond to complex climate changes in polar regions, accurately locate and adjust the UAV mission area. By collecting spectral reflectance, temperature, and crack width data in real time and combining intelligent analysis, the system can efficiently identify water accumulation grids, temporary grids, and observation grids, achieve precise task allocation and scheduling. In practical applications, the system can not only flexibly adjust the reflectance threshold according to the environmental changes of the grids, but also optimize the allocation of UAVs among multiple grids through comparison and fluctuation calculation, avoiding resource waste and improving the acquisition efficiency. In addition, the adaptive adjustment mechanism of the system can quickly adjust the number and distribution of UAVs according to real-time data fluctuations to ensure efficient working status in various extreme environments. Overall, the system significantly improves the task execution ability, response speed, and data acquisition accuracy of UAV swarms in polar complex environments, effectively reduces the operation cost, and enhances the comprehensiveness and accuracy of polar environmental monitoring, effectively solving the problem of low response speed of the system when facing emergencies due to relying on a single control strategy.

[0078] Please continue to refer to Figure 2 As shown, it is the determination logic diagram of the water accumulation grid and the non-water accumulation grid determined by the first determination unit of this embodiment;

[0079] The first determination module includes:

[0080] A reflectance comparison unit for comparing the real-time spectral reflectance with the preset reflectance threshold to form a reflectance comparison result;

[0081] A first determination unit connected to the reflectance comparison unit for determining that the grid to be determined is the water accumulation grid when the reflectance comparison result is that the real-time spectral reflectance is greater than the preset reflectance threshold, and for determining that the grid to be determined is the non-water accumulation grid when the reflectance comparison result is that the real-time spectral reflectance is less than or equal to the preset reflectance threshold.

[0082] The real-time spectral reflectance is compared with the preset reflectance threshold by the reflectance comparison unit to form a reflectance comparison result. If the real-time spectral reflectance is greater than the preset reflectance threshold, the first determination unit determines that the grid is a water accumulation grid; conversely, if the real-time spectral reflectance is less than or equal to the preset reflectance threshold, it is determined as a non-water accumulation grid. This process realizes the automatic discrimination of water accumulation conditions in polar regions.

[0083] By comparing the real-time spectral reflectance with the preset reflectance threshold, it is possible to accurately and quickly judge whether there is water accumulation in each grid, reducing the complexity of manual judgment and improving the efficiency and accuracy of data processing.

[0084] Specifically, the second determination module includes:

[0085] A temperature fluctuation calculation unit is used to calculate the standard deviation of the real-time temperature within a preset temporarily determined duration to form a temperature fluctuation value, where, , R is the temperature fluctuation value, Ti is the real-time temperature at each moment within the preset temporarily determined duration, and T0 is the average value of all real-time temperatures;

[0086] A first width fluctuation calculation unit is used to calculate the standard deviation of the real-time ice sheet crack width within the preset temporarily determined duration to form a first width fluctuation value, where, Y is the first width fluctuation value, Wi is the real-time ice sheet crack width at each moment within the preset temporarily determined duration, W0 is the average value of all real-time ice sheet crack widths, and N is the number of real-time temperatures or real-time ice sheet crack widths;

[0087] A second determination unit, which is respectively connected to the temperature fluctuation calculation unit and the first width fluctuation calculation unit, is used to determine a plurality of the temporary grids according to the temperature fluctuation value and the width fluctuation value.

[0088] The preset temporarily determined duration is the time period for calculating the real-time temperature and the crack width fluctuation, which depends on the speed of environmental change and the requirements of the system's response to changes. Usually, factors such as weather fluctuations and the ice sheet crack change cycle are considered. It is usually set between several hours and one day. In this embodiment, it is set to 12 hours, which can balance the real-time performance and the stability of data fluctuations, avoid misjudgment of fluctuations within too short a time, and at the same time ensure that the change trend within a long enough time can be captured, which helps to more accurately judge the temporary state of the area and improve the response accuracy and stability of the system.

[0089] The standard deviation of the real-time temperature within the preset temporarily determined duration is calculated by the temperature fluctuation calculation unit to obtain the temperature fluctuation value; at the same time, the standard deviation of the real-time ice sheet crack width is calculated by the first width fluctuation calculation unit to obtain the first width fluctuation value. Then, the second determination unit determines a plurality of temporary grids according to the change situations of the temperature fluctuation value and the crack width fluctuation value. This process can effectively identify the areas with fluctuations within a specific time period and provide a basis for further observation and analysis.

[0090] By calculating the volatility of the temperature and the crack width, the areas with unstable characteristics can be identified, providing a scientific basis for the determination of the temporary grids, thereby enhancing the accurate identification ability of the water accumulation area. By real-time monitoring the fluctuations of the temperature and the crack width, the accuracy of the UAV scheduling can be ensured, and the subsequent environmental monitoring and data collection can be further optimized, improving the response speed and flexibility of the system, which helps to improve the reliability and accuracy of the water accumulation identification in the polar region.

[0091] Please continue to refer to Figure 3As shown, it is the determination logic diagram of the temporary grid determined by the temporary determination subunit in this embodiment;

[0092] The second determination unit includes:

[0093] A temperature fluctuation curve drawing subunit, configured to draw a change curve of the temperature fluctuation value to form a temperature fluctuation curve;

[0094] A first width fluctuation curve drawing subunit, configured to draw a change curve of the first width fluctuation value to form a first width fluctuation curve;

[0095] A consistency calculation subunit, which is respectively connected to the temperature fluctuation curve drawing subunit and the width fluctuation curve drawing subunit, and is configured to calculate the cosine similarity of the temperature fluctuation curve and the width fluctuation curve to form a change consistency;

[0096] Among them, the formula for calculating the change consistency is: , where Q is the change consistency.

[0097] A temporary determination subunit, which is connected to the consistency calculation subunit, and is configured to determine that the water accumulation grid is a temporary grid when the change consistency is less than a preset consistency threshold.

[0098] The preset consistency threshold is a standard value used to determine whether the similarity between the temperature fluctuation curve and the crack width fluctuation curve is high enough to judge whether it is a temporary grid. It depends on the required sensitivity and accuracy. Usually, according to the actual application scenario, the change range of environmental factors, and the system's tolerance for abnormal changes, it is usually set between 0.7 and 0.9. In this embodiment, it is set to 0.8, which helps to balance the risks of misjudgment and missed judgment, and ensures that only when the changes in temperature and crack width fluctuations have strong consistency, it is determined as a temporary grid, thereby improving the accuracy and stability of the system.

[0099] Change curves are drawn through the temperature fluctuation value and the crack width fluctuation value, and a temperature fluctuation curve and a width fluctuation curve are respectively generated. Then, the cosine similarity of these two fluctuation curves, that is, the change consistency, is calculated. If the change consistency is less than the preset consistency threshold, it is determined that the water accumulation grid is a temporary grid, indicating that the state of this area changes greatly and has temporary instability, and further observation and processing are required.

[0100] By analyzing the fluctuation consistency of temperature and crack width, temporary grids can be effectively identified, which helps to further optimize the scheduling and observation strategies of the unmanned aerial vehicle. Through the consistency analysis of the changes in temperature and crack width, the areas with unstable or abnormal changes can be accurately judged, avoiding misjudgment and improving the adaptability and processing efficiency of the system in complex environments.

[0101] Specifically, the determination module includes:

[0102] A reflectivity fluctuation calculation unit for calculating the standard deviation of the real-time spectral reflectivity of a single temporary grid within a preset determination duration to form a reflectivity fluctuation value, where , U is the reflectivity fluctuation value, Pi is the real-time spectral reflectivity at each moment within the preset determination duration, and P0 is the mean value of all real-time spectral reflectivities;

[0103] A second width fluctuation calculation unit for calculating the standard deviation of the real-time ice sheet crack width of a single temporary grid within the preset determination duration to form a second width fluctuation value, where F is the second width fluctuation value, Wi’ is the real-time ice sheet crack width at each moment within the preset determination duration, W0’ is the mean value of all real-time ice sheet crack widths, and n is the number of real-time spectral reflectivities or real-time ice sheet crack widths;

[0104] A synchronization degree calculation unit, which is respectively connected to the reflectivity fluctuation calculation unit and the second width fluctuation calculation unit, for calculating the synchronization degree according to the reflectivity fluctuation value and the second width fluctuation value;

[0105] A determination unit, which is connected to the synchronization degree calculation unit, for determining a number of observation grids according to the synchronization degree of any two adjacent temporary grids.

[0106] The preset determination duration refers to the time period used when judging the synchronization degree of temporary grids, which depends on the geographical environment, data acquisition frequency, and actual monitoring requirements. It is usually set between several minutes and several hours. In this embodiment, it is set to 30 minutes, which can ensure that the collected data has a sufficient range of changes, avoid the interference of instantaneous fluctuations, and at the same time, it will not cause too drastic environmental changes due to too long a duration, ensuring the stability and accuracy of the determination.

[0107] The reflectivity fluctuation calculation unit and the second width fluctuation calculation unit respectively calculate the standard deviations of the spectral reflectivity and crack width within the temporary grid within the preset determination duration to obtain the reflectivity fluctuation value and the second width fluctuation value. The synchronization degree calculation unit calculates the synchronization degree according to these two fluctuation values. Finally, the determination unit determines whether they meet the conditions according to the synchronization degree of any two adjacent temporary grids, so as to determine a number of observation grids.

[0108] By calculating the synchronization degree of the spectral reflectivity and crack width fluctuation values of the temporary grid, grids showing similar change trends can be effectively identified, ensuring that the selection of observation grids is more accurate, reducing the interference of errors, and improving the accuracy and reliability of the overall grid division and monitoring.

[0109] Specifically, the synchronization degree calculation unit includes:

[0110] A reflectivity fluctuation curve plotting subunit, configured to plot a change curve of the reflectivity fluctuation value to form a reflectivity fluctuation curve;

[0111] A second width fluctuation curve plotting subunit, configured to plot a change curve of the second width fluctuation value to form a second width fluctuation curve;

[0112] A synchronization degree calculation subunit, which is respectively connected to the reflectivity fluctuation curve plotting subunit and the second width fluctuation curve plotting subunit, and is configured to calculate a cosine similarity between the reflectivity fluctuation curve and the second width fluctuation curve to form a synchronization degree;

[0113] Wherein, the formula for calculating the synchronization degree is: , Q is the change consistency degree, and G is the synchronization degree.

[0114] By plotting the change curves of the reflectivity fluctuation value and the crack width fluctuation value, a reflectivity fluctuation curve and a second width fluctuation curve are respectively formed. Then, the cosine similarity between these two fluctuation curves is calculated to obtain a synchronization degree value. The higher the synchronization degree, the more consistent the fluctuation trends between the two, reflecting the synchronous change of the water accumulation grid and the crack width.

[0115] Through the calculation of the synchronization degree, it can effectively identify whether the spectral reflectivity and the crack width change trends of adjacent grids are consistent within the same time period, thereby providing a basis for further screening of the temporary grids and improving the accuracy of grid determination. Using the cosine similarity to calculate the synchronization degree can eliminate the misjudgment caused by instantaneous fluctuations and ensure the stability and reliability of the determination.

[0116] Please continue to refer to Figure 4 as shown, which is a determination logic diagram of the determination subunit for determining the observation grid in this embodiment;

[0117] The determination unit includes:

[0118] A synchronization deviation calculation subunit, configured to calculate a relative deviation of the synchronization degree between any two adjacent temporary grids to form a synchronization deviation;

[0119] A determination subunit, which is connected to the synchronization deviation calculation subunit, and is configured to determine the temporary grid as the observation grid when the synchronization deviation is greater than a preset synchronization deviation threshold.

[0120] The preset synchronization deviation threshold is a standard value used to determine whether a temporary grid is an observation grid, depending on the expected variation ranges of factors such as temperature fluctuations and crack width changes within the grid, as well as how to distinguish normal fluctuations from abnormal fluctuations. It is usually set between 0.1 and 1.0. In this embodiment, it is set to 0.5 to ensure that abnormal grids that need attention can be effectively identified and, to a large extent, over-response to normal minor changes can be avoided, thereby improving the stability and practicality of the system.

[0121] First, the relative deviation between the synchronization degrees of two adjacent temporary grids is calculated by the synchronization deviation calculation subunit to obtain the synchronization deviation value. Then, the determination subunit determines whether the temporary grid belongs to the observation grid according to whether the synchronization deviation is greater than the preset synchronization deviation threshold. If the synchronization deviation is greater than the threshold, it is considered that the changes in the adjacent grids are inconsistent, and thus it is determined as an observation grid.

[0122] By comparing the synchronization deviation with the preset synchronization deviation threshold, the difference in synchronization degree between temporary grids can be accurately judged, ensuring that only when the changes are significantly inconsistent will the temporary grid be determined as an observation grid. Grids that do not meet the observation conditions can be effectively filtered out, improving the accuracy of data analysis and optimizing the subsequent UAV scheduling and regional monitoring efficiency.

[0123] Specifically, the control module includes:

[0124] An observation ratio calculation unit for calculating the ratio of the number of grids of the observation grid to the number of all grids to be determined to form an observation ratio;

[0125] A control unit connected to the observation ratio calculation unit for controlling a number of the UAVs to move from the non-ponding grids to the observation grid according to the observation ratio and the preset quantity.

[0126] The ratio of the number of observation grids to the number of all grids to be determined, that is, the observation ratio, is calculated by the observation ratio calculation unit. Then, the control unit selects and controls a number of UAVs from the non-ponding grids to move to the observation grid for further observation or data collection according to the relationship between this ratio and the preset quantity.

[0127] By calculating the observation ratio and reasonably scheduling the UAVs, it can be ensured that the observation grid is covered by enough UAVs, while avoiding resource waste, optimizing the allocation of UAVs, and improving the efficiency and accuracy of polar region monitoring.

[0128] Specifically, the control unit includes:

[0129] A distribution degree calculation subunit for calculating the standard deviation of the distances from all the grid positions to the preset center point to form a distribution degree when the observation ratio is greater than the preset ratio threshold.

[0130] A scheduling quantity calculation subunit, which is connected to the distribution degree calculation subunit, and is used to calculate the total scheduling quantity according to the relative deviation between the distribution degree and a preset distribution degree threshold, a preset scheduling coefficient, the number of grids of the observation grid, and the preset quantity when the distribution degree is greater than the preset distribution degree threshold;

[0131] A control subunit, which is connected to the scheduling quantity calculation subunit, and is used to control the UAVs with the total scheduling quantity to move to each of the observation grids on average from the non-flooded grids.

[0132] Among them, to calculate the total scheduling quantity, an adjustment factor is calculated first. The adjustment factor is obtained by multiplying the preset quantity by a multiplier jointly determined by an adjustment coefficient and a relative deviation. Then, the calculated adjustment factor is multiplied by the number of observation grids, and finally the total number of UAVs to be scheduled is obtained.

[0133] The preset proportion threshold is an important parameter for the control unit to judge whether to adjust the UAV scheduling. It depends on the coverage requirements of the actual task and the area size, and determines when more UAVs need to be scheduled to the observation grids. It is usually set between 10% - 30%. In this embodiment, it is set to 20%, which can balance the requirements of area coverage and resource scheduling, and helps to ensure that there are enough UAVs in the observation grids for effective monitoring.

[0134] The preset distribution degree threshold is a standard value used to evaluate the distribution of the observation grid. It is set according to the size of the grid, the complexity of the task, and the characteristics of the area, and is usually set between 10% - 40%. In this embodiment, it is set to 25%, which helps to adjust the UAV distribution in time to achieve uniform area coverage.

[0135] The preset scheduling coefficient is an adjustment parameter used when calculating the UAV scheduling quantity. It depends on the actual task requirements, the capabilities of the UAVs, and the area conditions, and is usually set between 1.0 and 2.0. In this embodiment, it is set to 1.5, which can moderately increase the number of scheduled UAVs when the distribution degree is relatively high to ensure the task completion efficiency.

[0136] When the observation proportion is greater than the preset proportion threshold, the distribution degree calculation subunit calculates the standard deviation of the distance from the grid position to the preset center point to form the distribution degree. If the distribution degree is greater than the preset threshold, the scheduling quantity calculation subunit calculates the total quantity to be scheduled according to the distribution degree, deviation, scheduling coefficient, and preset quantity. Finally, the control subunit reasonably schedules the UAVs from the non-flooded grids according to the calculation result and evenly distributes them to each observation grid to execute the task.

[0137] By dynamically adjusting the scheduling quantity and UAV allocation, the target area can be efficiently covered, avoiding over-concentration or sparse distribution, and improving the working efficiency of UAVs in polar region monitoring and the comprehensiveness of data collection.

[0138] Specifically, the adjustment module includes:

[0139] A grid fluctuation calculation unit for calculating the standard deviation of the number of grids in the temporary grid within a preset adjustment duration to form a grid number fluctuation value;

[0140] An adjustment unit connected to the grid fluctuation calculation unit for increasing the preset reflectivity threshold according to the relative deviation between the grid number fluctuation value and the preset number fluctuation threshold and a preset adjustment coefficient when the grid number fluctuation value is greater than the preset number fluctuation threshold.

[0141] First, the grid fluctuation calculation unit calculates the standard deviation of the number of temporary grids within a preset adjustment duration to obtain a grid number fluctuation value. Then, the adjustment unit makes a judgment based on the relative deviation between the grid number fluctuation value and the preset number fluctuation threshold and the preset adjustment coefficient. If the grid number fluctuation value is greater than the preset threshold, the adjustment unit will increase the preset reflectivity threshold to adapt to the new environmental changes.

[0142] By dynamically adjusting the reflectivity threshold, over-reliance on unstable grid data can be avoided, the robustness and adaptability of the system can be improved, and it can be ensured that the system can respond flexibly and optimize the task allocation and resource scheduling of UAVs in the case of large environmental changes.

[0143] So far, the technical solution of the present invention has been described in combination with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or replacements to the relevant technical features, and the technical solutions after these changes or replacements will fall within the protection scope of the present invention.

Claims

1. An intelligent control system for drone swarms, characterized in that: include: A data acquisition module is used to obtain real-time spectral reflectance, real-time temperature and real-time ice sheet crack width collected by a preset number of drones in each grid to be determined in the polar region divided by grids; A first determination module, connected to the data acquisition module, for determining a number of waterlogging grids and non-waterlogging grids according to the real-time spectral reflectance and a preset reflectance threshold; A second determination module, which is connected to the first determination module and the data acquisition module respectively, and is used to determine a number of temporary grids according to the real-time temperature of each of the water accumulation grids and the real-time ice sheet crack width; A determination module, which is connected to the second determination module and the data acquisition module respectively, and is used to determine a number of observation grids according to the real-time spectral reflectance and the real-time ice sheet crack width of any two adjacent temporary grids; A control module connected to the determination module, for controlling a plurality of the drones to move from the non-waterlogging grid to the observation grid according to the number of grids, grid positions and the preset number of the observation grid; An adjustment module is connected to the data acquisition module, the control module and the first determination module respectively, and is used to adjust the preset reflectivity threshold based on the temporary grid determined after scheduling a number of the drones.

2. The UAV swarm intelligent control system according to claim 1 is characterized in that: The first determining module comprises: A reflectivity comparison unit, used to compare the real-time spectral reflectivity with the preset reflectivity threshold to form a reflectivity comparison result; A first determination unit is connected to the reflectivity comparison unit, and is used to determine that the grid to be determined is the waterlogged grid when the reflectivity comparison result is that the real-time spectral reflectivity is greater than the preset reflectivity threshold, and to determine that the grid to be determined is the non-waterlogged grid when the reflectivity comparison result is that the real-time spectral reflectivity is less than or equal to the preset reflectivity threshold.

3. The UAV swarm intelligent control system according to claim 2 is characterized in that: The second determining module comprises: A temperature fluctuation calculation unit, used to calculate the standard deviation of the real-time temperature within a preset temporary determined time period to form a temperature fluctuation value; A first width fluctuation calculation unit, used for calculating the standard deviation of the real-time ice sheet crack width within the preset temporary determination time length to form a first width fluctuation value; A second determining unit is connected to the temperature fluctuation calculating unit and the first width fluctuation calculating unit respectively, and is used to determine a plurality of temporary grids according to the temperature fluctuation value and the width fluctuation value.

4. The UAV swarm intelligent control system according to claim 3 is characterized in that: The second determining unit includes: A temperature fluctuation curve drawing subunit is used to draw a variation curve of the temperature fluctuation value to form a temperature fluctuation curve; A first width fluctuation curve drawing subunit, used for drawing a variation curve of the first width fluctuation value to form a first width fluctuation curve; a consistency calculation subunit, which is connected to the temperature fluctuation curve drawing subunit and the width fluctuation curve drawing subunit respectively, and is used to calculate the cosine similarity of the temperature fluctuation curve and the width fluctuation curve to form a change consistency; A temporary determination subunit is connected to the consistency calculation subunit, and is used to determine that the waterlogging grid is a temporary grid when the change consistency is less than a preset consistency threshold.

5. The UAV swarm intelligent control system according to claim 4 is characterized in that: The determination module comprises: A reflectivity fluctuation calculation unit, used to calculate the standard deviation of the real-time spectral reflectivity of a single temporary grid within a preset determination time period to form a reflectivity fluctuation value; A second width fluctuation calculation unit, used for calculating the standard deviation of the real-time ice sheet crack width of a single temporary grid within the preset determination time period to form a second width fluctuation value; a synchronization degree calculation unit, connected to the reflectivity fluctuation calculation unit and the second width fluctuation calculation unit respectively, for calculating the synchronization degree according to the reflectivity fluctuation value and the second width fluctuation value; A determination unit is connected to the synchronization calculation unit and is used to determine a number of observation grids according to the synchronization between any two adjacent temporary grids.

6. The UAV swarm intelligent control system according to claim 5, characterized in that: The synchronization degree calculation unit comprises: A reflectivity fluctuation curve drawing subunit is used to draw a change curve of the reflectivity fluctuation value to form a reflectivity fluctuation curve; A second width fluctuation curve drawing subunit, used for drawing a variation curve of the second width fluctuation value to form a second width fluctuation curve; The synchronization degree calculation subunit is connected to the reflectivity fluctuation curve drawing subunit and the second width fluctuation curve drawing subunit respectively, and is used to calculate the cosine similarity of the reflectivity fluctuation curve and the second width fluctuation curve to form the synchronization degree.

7. The UAV swarm intelligent control system according to claim 6, characterized in that: The determination unit comprises: A synchronization deviation calculation subunit, used for calculating the relative deviation of the synchronization degree of any two adjacent temporary grids to form a synchronization deviation; A determination subunit is connected to the synchronization deviation calculation subunit, and is used to determine that the temporary grid is the observation grid when the synchronization deviation is greater than a preset synchronization deviation threshold.

8. The UAV swarm intelligent control system according to claim 7, characterized in that: The control module comprises: An observation ratio calculation unit is used to calculate the ratio of the number of grids of the observation grid to the number of grids of all grids to be determined, so as to form an observation ratio; A control unit is connected to the observation ratio calculation unit, and is used to control a plurality of the drones to move from the non-waterlogging grid to the observation grid according to the observation ratio and the preset number.

9. The UAV swarm intelligent control system according to claim 8, characterized in that: The control unit comprises: A distribution degree calculation subunit, used for calculating the standard deviation of the distances from all the grid positions to the preset center point to form a distribution degree when the observation proportion is greater than a preset proportion threshold; a dispatch quantity calculation subunit, connected to the distribution degree calculation subunit, for calculating the total dispatch quantity according to the relative deviation between the distribution degree and the preset distribution degree threshold, the preset dispatch coefficient, the number of grids of the observation grid and the preset number when the distribution degree is greater than the preset distribution degree threshold; A control subunit is connected to the dispatch quantity calculation subunit, and is used to control the total dispatch quantity of the drones to be moved evenly from the non-waterlogged grids to each of the observation grids.

10. The UAV swarm intelligent control system according to claim 9, characterized in that: The adjustment module comprises: A grid fluctuation calculation unit, used to calculate the standard deviation of the number of grids of the temporary grid within a preset adjustment time to form a grid number fluctuation value; An adjustment unit is connected to the grid fluctuation calculation unit and is used to increase the preset reflectivity threshold according to a relative deviation between the grid quantity fluctuation value and the preset quantity fluctuation threshold and a preset adjustment coefficient when the grid quantity fluctuation value is greater than the preset quantity fluctuation threshold.

Images