Cloud unmanned aerial vehicle path optimization method and system

Abstract

The invention relates to the technical field of unmanned aerial vehicle scheduling, in particular to a cloud unmanned aerial vehicle path optimization method and system, and the method comprises the following steps: obtaining task priority distribution and resource state atlas and link quality data, extracting priority fluctuation and resource occupancy features, recognizing interference evaluation and comparison records to be coincident, and obtaining a cloud unmanned aerial vehicle path optimization result. And screening compliance check node mapping task content and standard judgment deviation, extracting an abnormal task group to calculate resource task ratio record period difference, and generating a task progress monitoring structure index. According to the invention, through the cloud platform and dynamic task scheduling optimization, the real-time performance and flexibility of unmanned aerial vehicle path planning are improved, the path planning is optimized, the correlation degree of path nodes is corrected, the path precision is ensured, communication interference is dynamically monitored, task failure is avoided, resource allocation and task cycle matching are optimized, and the task execution efficiency and safety are improved. Efficient completion is ensured, task priorities and path selection can be flexibly adjusted, and multi-task execution conflicts are reduced.

Description

Technical Field

[0001] This invention relates to the field of drone scheduling technology, and in particular to a cloud-based drone path optimization method and system. Background Technology

[0002] The field of drone scheduling technology involves the rational planning and scheduling of drone flight paths through information technology and communication methods, thereby improving the efficiency and safety of mission execution. Core aspects of this field include drone flight path optimization, task allocation, resource management, and real-time adjustment of flight plans. With the widespread development of drone applications, especially in logistics, inspection, and agriculture, how to efficiently and reliably schedule a large number of drones to complete complex tasks has become a critical issue that urgently needs to be addressed. Drone scheduling not only requires consideration of the geographical constraints of flight missions but also needs to address challenges such as flight safety, energy consumption optimization, and communication networks. Traditional drone path optimization methods, based on static or simple dynamic models, calculate fixed optimal paths using pre-set algorithms. However, these methods struggle to cope with sudden changes or dynamic adjustments in complex environments and changing mission requirements. To improve the real-time performance and flexibility of path planning, existing technologies employ cloud-based drone path optimization methods. These methods rely on cloud computing platforms, utilizing big data analytics and real-time communication technologies to execute drone mission scheduling and path planning in the cloud, enabling dynamic optimization based on real-time mission requirements, environmental changes, and flight status.

[0003] Existing technologies have limitations when facing complex environmental changes and ever-changing mission requirements. Traditional path optimization methods rely heavily on static models, which cannot flexibly adjust when mission requirements deviate or the environment changes. Because existing technologies fail to monitor mission execution status dynamically in real time, and cannot update mission priorities and resource allocation in real time, mission path planning and resource allocation are based on preset models, lacking necessary adaptability and real-time response capabilities. For example, when communication links are interfered with, existing methods cannot detect and adjust path planning in time, leading to flight mission interruptions or failures. Furthermore, if weather changes or geographical changes occur during flight, traditional path optimization methods cannot react quickly, resulting in reduced mission efficiency or safety hazards. Existing technologies have weak capabilities in handling multi-task parallel scheduling and resource allocation, failing to effectively reduce conflicts between tasks or improve mission coordination, thus compromising the efficiency and safety of simultaneous multi-task execution. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a cloud-based drone path optimization method and system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a cloud-based drone path optimization method, comprising the following steps: S1: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, and generate task priority deviation labels; S2: Based on the deviation amount and pattern classification in the task priority deviation label, select the path optimization mode that matches the pattern classification from a variety of preset path optimization modes, extract the path node correlation degree distribution range, and perform deviation correction and classification processing with the baseline path correlation degree to obtain the path optimization mode matching label. S3: Call the path optimization mode matching label, extract the path node correlation segment number, identify the communication link interference assessment data in the segment, compare the communication link interference with the path node segment, record the number of interference overlap time periods, and generate a list of communication link interference affected segments. S4: Based on the list of communication link interference affected sections, filter task compliance verification nodes, extract task compliance specifications and operation data, map the distribution of task content and compliance requirements, determine whether there is any deviation in task compliance, and obtain the UAV task compliance adjustment set.

[0006] As a further embodiment of the present invention, the task priority deviation label includes deviation number, priority status label, resource usage deviation amount, and mode classification; the path optimization mode matching label includes fluctuation type, benchmark comparison result, and task association number; the communication link interference affected segment list includes device type, affected time segment, number of overlapping time periods, and affected task number; and the UAV task compliance adjustment set includes task distribution unevenness number, operation duration record, task completion deviation amount, and operation matching degree.

[0007] As a further aspect of the present invention, the step of obtaining the task priority deviation label specifically includes: S111: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, match task priority requirements with real-time distribution, compare time range with priority requirements, and generate task record time periods for partition nodes. S112: Based on the task record time period of the partition node, extract the overlapping time period between the task priority status and the required time interval, calculate the ratio of overlapping time to the total required time, filter nodes with an overlap ratio lower than the benchmark value, and obtain the priority coverage deviation rate of the partition node according to the number of priority status annotations. S113: Based on the partition node priority coverage deviation rate, determine the deviation status of the node number, identify the node number whose deviation rate exceeds the node synchronization threshold, integrate the node number, priority coverage information and deviation rate value, and generate a task priority deviation label.

[0008] As a further aspect of the present invention, the step of obtaining the path optimization pattern matching label specifically includes: S211: Based on the task priority deviation label, identify the path optimization mode and path node correlation distribution characteristics that are adapted to the characteristics of the current task, extract the start and end fluctuation range of real-time path node correlation, calculate the start and end fluctuation difference of the path node correlation segment, compare it with the baseline path correlation fluctuation range, and obtain the path node correlation fluctuation deviation value. S212: Call the path node correlation fluctuation deviation value, combine the segment distribution, fluctuation trend and adjustment frequency, uniformly collect the path optimization mode deviation data, identify and calculate the adaptability deviation degree according to the segment number, determine the fluctuation direction according to the adjustment frequency, and obtain the path optimization mode matching label.

[0009] As a further aspect of the present invention, the step of obtaining the list of communication link interference-affected segments specifically includes: S311: Call the path optimization mode matching label, filter the task segment number of the path node correlation deviation, extract the path node correlation time period according to the node correlation table, process the segment path node time period according to the time dimension, identify the path node correlation time period index table, and obtain the path node correlation task time period set. S312: Based on the path node correlation task time period set, collect communication link interference assessment data for the same period segment, identify the communication link interference impact table, determine the daily communication link interference based on the interference threshold, match it with the path node correlation task time period, determine whether there is abnormal correlation of path node correlation, and generate a list of communication link interference affected segments.

[0010] As a further aspect of the present invention, the step of obtaining the UAV mission compliance adjustment set specifically includes: S411: Based on the list of communication link interference affected sections, filter the task compliance verification nodes, extract the task list, and obtain the task set of nodes not affected by communication link interference. S412: Call the task set of nodes not affected by communication link interference, match the task compliance specifications and operation data, extract the planned task quantity by task number, count the number of operators and real-time operation time period, and generate a task compliance execution status matching dataset. S413: Based on the task compliance execution status matching dataset, evaluate the degree of matching between task execution efficiency and operation distribution, identify efficiency fluctuation nodes, calculate fluctuation identification difference value, mark tasks with fluctuations exceeding the benchmark value as abnormal nodes, identify task compliance execution matching deviations, and obtain the UAV task compliance adjustment set.

[0011] As a further aspect of the present invention, the method further includes step S5: S5: Call the UAV mission compliance adjustment set, extract the abnormal fluctuation mission group, calculate the ratio of resource input to mission volume in the mission model, record the difference distribution between resource deployment cycle and mission execution cycle, and generate mission progress monitoring structure indicators. The task progress monitoring structure indicators include resource allocation ratio, task intensity level, execution cycle difference, and task efficiency indicators.

[0012] As a further aspect of the present invention, the steps for obtaining the task progress monitoring structure indicators are specifically as follows: S511: Call the UAV mission compliance adjustment set, filter nodes that exceed the threshold, record the time interval and the magnitude of changes in the task volume, and obtain the progress fluctuation abnormality identifier set; S512: Based on the task model resources corresponding to the nodes in the progress fluctuation anomaly identifier set, identify the node resource input ratio sequence, extract the abnormal distribution interval and compare the critical coefficient, record the ratio offset direction and node number, and form a task resource matching offset index group. S513: Based on the task resource matching offset index group, extract the task model resource deployment and task execution time period, identify the difference between the resource deployment cycle and the operation cycle, and sort and label them according to the progress benchmark to generate task progress monitoring structure indexes.

[0013] The cloud-based drone path optimization system is used to execute the aforementioned cloud-based drone path optimization method. The system includes: The task priority deviation extraction module obtains task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extracts the task priority fluctuation range and resource occupancy intensity distribution characteristics, compares the task priority demand with the real-time distribution, and generates task priority deviation labels. The path optimization mode classification module locates the path optimization mode that matches the characteristics of the current task based on the task priority deviation label, extracts the task node identifier, process node and baseline path correlation degree, calculates the difference between the on-site path correlation degree and the baseline path correlation degree, classifies and labels the difference type, and generates path optimization mode matching labels. The communication link interference identification module, based on the path optimization mode matching label, filters path node correlation lag task components, locates the corresponding segment, extracts the communication link interference assessment interference period, determines overlap with the operation period, filters frequently interfered segments, and generates a list of communication link interference affected segments. The task compliance diagnosis module, based on the list of communication link interference-affected segments, removes tasks in the interference segments, extracts task compliance specifications and operation data, matches task assignment and operation time periods, calculates the ratio of operation volume to task, identifies tasks with dense tasks and inefficient execution of components, and obtains the UAV task compliance adjustment set. Based on the UAV mission compliance adjustment set, the resource allocation analysis module locates the resource input records of the mission in the mission model, extracts the ratio of mission quantity to resource quantity allocation, compares the difference between the operation cycle and the input cycle, maps the mission resource usage and progress status, and forms a mission progress monitoring structure index.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention improves the real-time performance, adaptability, and flexibility of UAV path planning by combining a cloud platform and dynamic task scheduling optimization. By acquiring real-time data such as task priority fluctuations, resource usage, and communication link quality, the path planning model is optimized and the correlation of path nodes is corrected, making path planning more accurate and reliable in complex tasks. Furthermore, dynamic monitoring of communication link interference and the generation of interference-affected segments ensure timely path adjustments when interference is encountered during flight, preventing flight interruptions or mission failures. The application of a task compliance adjustment set enables real-time compliance verification of tasks, promptly identifying and correcting deviations in task execution, optimizing the matching of resource allocation and task execution cycles, and improving the safety and efficiency of flight missions. Through system resource monitoring and the generation of task progress structure indicators, the UAV mission execution process becomes more transparent and controllable, reducing risks and ensuring efficient mission completion in dynamic environments. Moreover, the adaptability of the solution allows for flexible adjustment of task priorities and path selection under various flight missions, improving the efficiency of simultaneous multi-task execution and reducing the probability of task conflicts. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the workflow of the present invention; Figure 2 This is a flowchart of the process for obtaining task priority deviation labels in this invention; Figure 3 This is a flowchart illustrating the process of obtaining path optimization pattern matching tags in this invention. Figure 4 This is a flowchart illustrating the process of obtaining the list of communication link interference-affected sections in this invention. Figure 5 This is a flowchart illustrating the process of obtaining the drone mission compliance adjustment set in this invention. Figure 6 This is a flowchart illustrating the process of obtaining the task progress monitoring structure indicators in this invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] Example 1 Please see Figure 1 This invention provides a technical solution: a cloud-based drone path optimization method, comprising the following steps: S1: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, and generate task priority deviation labels; S2: Based on the deviation amount and pattern classification in the task priority deviation label, select the path optimization mode that matches the pattern classification from a variety of preset path optimization modes, combine with the original task database, extract the path node correlation distribution range, and perform deviation correction and classification processing with the baseline path correlation to obtain the path optimization mode matching label. S3: Call the path optimization mode matching label, extract the path node correlation segment number, identify the communication link interference assessment data in the segment, compare the communication link interference with the path node segment, record the number of interference overlap time periods, and generate a list of communication link interference affected segments. S4: Based on the list of communication link interference affected sections, filter task compliance verification nodes, extract task compliance specifications and operation data, map task content and compliance requirement distribution, determine whether there is any deviation in task compliance, and obtain the UAV task compliance adjustment set; S5: Call the drone mission compliance adjustment set, extract the abnormal fluctuation mission group, calculate the ratio of resource input to mission volume in the mission model, record the difference distribution between resource deployment cycle and mission execution cycle, and generate mission progress monitoring structure indicators.

[0019] Task priority deviation labels include deviation number, priority status label, resource usage deviation, and mode classification; path optimization mode matching labels include fluctuation type, benchmark comparison result, and task association number; communication link interference affected segment list includes device type, affected time segment, number of overlapping time periods, and affected task number; UAV task compliance adjustment set includes task distribution unevenness number, operation duration record, task completion deviation, and operation matching degree; task progress monitoring structure indicators include resource allocation ratio, task intensity level, execution cycle difference, and task efficiency indicator.

[0020] Please see Figure 2 The specific steps for obtaining the task priority deviation tag are as follows: S111: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, match task priority requirements with real-time distribution, compare time range with priority requirements, and generate task record time periods for partition nodes. The system acquires task priority distribution information, resource occupancy status graphs, and communication link quality assessment data from the cloud-based drone task scheduling platform. The API interface continuously polls and subscribes to the historical task log database of the cloud-based drone task scheduling platform (DJI Airport Scheduling System), retrieving drone task priority field records from the past 24 hours. These records include task ID, priority level (P1, P2, P3), task initiation time, and completion time. Simultaneously, it collects drone resource status sensor data from each zone node (City A's "Central Business District," "Residential Area B," and "Industrial Park C"), including the number of currently available drones, remaining battery power, and real-time ground station operator load rate (percentage). The network monitoring module also acquires real-time assessment data of the communication link between each drone and the cloud, such as signal strength (dBm), latency (ms), and packet loss rate (%). For example, at 8:00 AM on a certain day, the system acquires data showing that 5 drones are available in the "Central Business District" node, with an average remaining battery of 95%, an operator load rate of 85%, a communication link signal strength of -70dBm, a latency of 50ms, and a packet loss rate of 0.5%. This data is then compared to the data from the past 1... Hourly task priority data statistics for the central business district revealed that P1 tasks accounted for 70%, P2 tasks for 20%, and P3 tasks for 10%. Simultaneously, the average drone utilization rate was 80%, the average operator load was 75%, and the average packet loss rate of the communication link was 1.2%. When extracting the task priority fluctuation range and resource occupancy intensity distribution characteristics, based on the historical task priority distribution information, a sliding time window (1 hour) was used to statistically analyze the task priority for each partition node. The calculation showed that in the past hour, the proportion of P1 tasks in the "central business district" fluctuated from 65% to 75%, P2 tasks from 15% to 25%, and P3 tasks from 5% to 15%, determining the real-time priority fluctuation range for P1 tasks to be [65%, 75%]. Next, the resource occupancy status map data for each partition node was analyzed to extract the resource occupancy intensity distribution characteristics for each partition node. By calculating the mean and standard deviation of resource indicators such as the number of available drones and operator load rate in the "central business district" over the past 24 hours, the intensity distribution of the number of available drones was found to be a mean of 4.5 and a standard deviation of 1.Two machines were used, with an operator load intensity distribution of mean 70% and standard deviation 10%. When matching task priority requirements with real-time distribution, the task priority requirements for each partition node were first obtained from the preset task strategy library. For the morning peak (8:00-10:00) in the "Central Business District," priority tasks P1 accounted for 80% of the requirements, P2 tasks accounted for 15%, and P3 tasks accounted for 5%. The real-time detected priority distribution (P1: 70%, P2: 20%, P3: 10%) in the "Central Business District" within the current time window (8:00-9:00) was compared with this requirement to identify... If the real-time distribution of tasks P1 is below the demand value, while the real-time distribution of tasks P2 is above the demand value, when comparing the time range with priority requirements, check whether the current time period (8:00-9:00) falls within the preset high-priority demand time period (morning peak 8:00-10:00). If the current time period is confirmed to be within the morning peak high-priority demand range, based on the above matching and comparison results, encapsulate the task status, priority distribution, resource usage, communication link quality, and other information for the "Central Business District" 8:00-9:00 time period as the partition node task record time period, and generate the partition node task record time period.

[0021] S112: Based on the task record time period of the partition node, extract the overlapping time period between the task priority status and the required time interval, calculate the ratio of overlapping time to the total required time, filter nodes with an overlap ratio lower than the benchmark value, and obtain the priority coverage deviation rate of the partition node according to the number of priority status annotations. Based on the task record time periods of each partition node, the task priority status of each node within a specific time period is extracted (in the "Central Business District," the actual execution rate of P1 tasks is 70% during the 8:00-9:00 time period) and the corresponding priority demand time interval (the demand rate of P1 tasks during this time period is 80%). In the 8:00-9:00 task record time period of the "Central Business District," the actual P1 task rate is 70%, while the demand rate of P1 tasks during this time period is set to 80%. The overlap ratio between the actual priority status duration and the total demand duration is calculated. The overlap time refers to the length of time the actual priority status meets the demand status, and the total demand duration refers to the total length of time the demand status is set. If a partition node is defined as having a high demand range for P1 priority from 8:00 to 10:00 AM, but the actual detected duration of high demand for P1 priority is only 1 hour and 30 minutes, then the overlap time is 1.5 hours, the total demand duration is 2 hours, and the calculated overlap ratio is 75% (1.5 hours / 2 hours = 0.75). Nodes with an overlap ratio lower than the benchmark value are filtered out. At that time, a priority coverage overlap benchmark value of 0.8 is set. This value is determined through historical data analysis and expert experience, indicating that 80% of the priority demand time is covered by the actual priority status. If the overlap ratio is 75%, which is lower than the benchmark value of 0.8, the "Central Business District" node is selected. In a certain scheduling cycle, the overlap ratio of P1 tasks in "Central Business District" is monitored to be 0.75, while the overlap ratio of P1 tasks in "Residential Area B" is 0.90. Only "Central Business District" is selected as a node with an overlap ratio lower than the benchmark value. Based on the number of priority status labels, the number of time segments in which a specific priority status (P1 task) does not meet the demand is counted among the selected nodes. For example, in the P1 task demand interval of 8:00-10:00, there are actually 3 P1 task segments with a length of 10 minutes in "Central Business District" with a P1 task ratio lower than the demand. Then the number of P1 priority status labels is 3. The priority coverage deviation rate of the partition node is calculated by the formula: Deviation rate = (Total duration of time segments with unmet demand / Total duration of demand) × 100%. For example, if the total duration of the "Central Business District" P1 task requirement is 2 hours (120 minutes), and the total duration of the unmet requirement time segment is 30 minutes (3 × 10 minutes), then its priority coverage deviation rate is (30 minutes / 120 minutes) × 100%.

[0022] S113: Based on the partition node priority coverage deviation rate, determine the deviation status of the node number, identify the node number whose deviation rate exceeds the node synchronization threshold, integrate the node number, priority coverage information and deviation rate value, and generate a task priority deviation label. Based on the priority coverage deviation rate of partition nodes, the priority coverage deviation rate of task P1 in "Central Business District" is 25%, and the priority coverage deviation rate of task P2 in "Industrial Park C" is 10%. When determining the deviation status of node numbers, a preset node synchronization threshold of 20% is used. This threshold is obtained through regression analysis of the degree to which task execution efficiency is affected by priority deviation in historical scheduling data. When the priority deviation rate exceeds 20%, the average task latency increases by more than 15%. When identifying node numbers whose deviation rate exceeds the node synchronization threshold, the priority coverage deviation rate of each partition node is compared with the threshold. Comparing the 5% deviation rate with the 20% threshold, 25% is greater than 20%. Therefore, "Central Business District" is identified as a node with a deviation rate exceeding the threshold. "Industrial Park C," with a 10% deviation rate, is not identified as a deviation node as it does not exceed the 20% threshold. When integrating node numbers, priority coverage information, and deviation rate values, all information related to identified deviation nodes is structured and summarized. The node number "Z_CSQ_001" for "Central Business District," its P1 task priority coverage information "Actual P1 percentage 70%, Demand 80%," and the deviation rate value of 25% are integrated to form a data record and generate a task priority deviation label. For "Central Business District," the label is {"Deviation Number": "Z_CSQ_001", "Priority Status Label": "P1 Deviation", "Resource Occupation Deviation": (specific value), "Mode Classification": (specific classification), "Additional Information": {...}}.

[0023] Please see Figure 3 The specific steps for obtaining the path optimization pattern matching tags are as follows: S211: Based on task priority deviation labels, identify the path optimization mode and path node correlation distribution characteristics that are suitable for the current task characteristics, extract the start and end fluctuation ranges of real-time path node correlation, calculate the start and end fluctuation difference of the path node correlation segment, and compare it with the baseline path correlation fluctuation range using the formula: ; Obtain the path node correlation fluctuation deviation value; in, This represents the deviation value of the correlation between path nodes. The segment representing the path node correlation degree The correlation value of each node. The segment representing the path node correlation degree The correlation value of each node. The number of nodes in a segment representing the relevance of a path node. Reference values ​​representing the fluctuation range of the baseline path correlation; Based on the task priority deviation label, and finding a 25% deviation in task priority for P1 at the "Central Business District" node, when identifying and adapting the current task characteristic path optimization mode and path node correlation distribution characteristics, the path optimization mode suitable for task P1 is determined to be "shortest time path mode" based on the P1 task priority information in the deviation label and the preset task characteristic-path optimization mode mapping rule library. Further, historical path node correlation distribution characteristics under the "shortest time path mode" are retrieved. Historical data shows that when executing task P1, the path node correlation (i.e., the degree of spatial matching between the actual flight path and the planned path, ranging from 0-100%) remains between 90% and 98%. To monitor the fluctuation range of real-time path node correlation, path node data during the current task execution process of the UAV is sampled, and the maximum and minimum values ​​of path node correlation within each time period are calculated. During the 5 minutes of the UAV executing task P1, the path node correlation value is sampled once per second. Statistics show that during this period, the maximum path node relevance was 95%, and the minimum was 88%. Therefore, the fluctuation range of real-time path node relevance is [88%, 95%], using the following formula: Calculate the fluctuation deviation value of the path node correlation degree; in, This indicates the deviation value of the path node correlation; the larger the deviation, the better. The more significant the fluctuation in the real-time path node correlation, the further it deviates from the baseline value. The segment representing the path node correlation degree The correlation value of each node. The segment representing the path node correlation degree The correlation value of each node. This refers to the number of nodes within the path node correlation segment (e.g., one node per second in a 5-minute sampling period). If one node is sampled every 10 seconds, then ), The baseline path relevance fluctuation range is set to 3.0. This value was derived from the analysis of a large amount of historical P1 task execution data, indicating that when the path node relevance fluctuation deviation is less than 3.0, the task completion time fluctuation remains within an acceptable range. In the actual calculation process, we selected 10 consecutive path node relevance sampling values ​​(sampled once every 30 seconds) of a UAV performing a P1 task in the "central business district" as the baseline. sequence: ,and ; First calculate Part: |90 92∣+∣93 90∣+∣89 93∣+∣91 89∣+∣94 91∣+∣88 94∣+∣92 88∣+∣90 92∣+∣91 90∣=27; Then, calculate part: ; Choose to divide by in the formula This is to eliminate the linear influence of the number of sampling points on the results when calculating the fluctuation deviation of path node correlation, so that it more accurately reflects the fluctuation intensity itself. This is because as the number of sampling points increases, the total fluctuation roughly increases with... Growth, rather than strictly linear growth, if directly divided by The results will be excessively diluted, making the values ​​at different sampling frequencies incomparable; using Normalization, similar to the standard deviation approach, matches the indicator to the statistical scale of fluctuation, thereby ensuring that the results calculated under different sampling conditions are accurate. It can objectively measure the actual fluctuation range of the correlation between path nodes; Finally, calculate : ; This formula can effectively measure the frequency and magnitude of fluctuations in the relevance of path nodes, and can be combined with a benchmark value. The comparison quantifies the degree of path deviation; in this example, Significantly higher This indicates that the correlation between current path nodes fluctuates significantly and abnormally, with UAVs frequently deviating from the planned path or making unnecessary corrections, leading to mission delays. This method can provide quantitative support for path optimization decisions, identify potential path planning problems, and provide a basis for subsequent optimization measures.

[0024] S212: Call the path node correlation fluctuation deviation value, combine the segment distribution, fluctuation trend and adjustment frequency, uniformly collect the path optimization mode deviation data, identify and calculate the adaptability deviation degree according to the segment number, determine the fluctuation direction according to the adjustment frequency, and obtain the path optimization mode matching label. The path node correlation fluctuation deviation value of 5.54 was used to analyze its distribution, fluctuation trend (upward trend), and path adjustment frequency (12 times / 5 minutes) across different time periods. Information such as the deviation value, occurrence time, fluctuation trend, and adjustment frequency was stored as path optimization mode deviation data. Using the segment number "P1 Task_Central Business District_8:00-9:00" and the evaluation model, the adaptability deviation was calculated. The adaptability deviation equals the path node correlation fluctuation deviation value of 5.54 multiplied by the adjustment frequency of 12, then divided by... The ratio of the total task segment duration of 60 minutes to the unit time of 5 minutes is 5.54. The adjustment frequency threshold is set to 5 times / 5 minutes. The actual adjustment frequency is 12 times / 5 minutes, which exceeds the threshold. It is judged as "the direction of increasing fluctuation". The path optimization mode matching label is obtained, for example, {"Task segment number": "P1 task_central business district_8:00-9:00", "Path optimization mode": "shortest time path mode", "Adaptability deviation": "5.54", "Fluctuation direction": "the direction of increasing fluctuation"}.

[0025] Please see Figure 4 The specific steps for obtaining the list of communication link interference-affected sections are as follows: S311: Call the path optimization mode matching label, filter the task segment number of the path node correlation deviation, extract the path node correlation time period according to the node correlation table, process the segment path node time period by time dimension, identify the path node correlation time period index table, and obtain the path node correlation task time period set. The path optimization pattern matching label is retrieved from step S212. The matching label for "P1 Task_Central Business District_8:00-9:00" includes an adaptation deviation of 5.54 and the direction of increasing fluctuation. The task segment numbers with path node correlation deviation are filtered. Based on the adaptation deviation indicated in the matching label, the filtering benchmark is set to an adaptation deviation greater than 5.0. The task segment number "P1 Task_Central Business District_8:00-9:00" with an adaptation deviation of 5.54 is selected. The path node correlation time period is extracted from the node correlation table. A pre-established "node-task-time" correlation table is queried. This table records the drone node ID corresponding to each task segment number and its actual execution time period. The drone node ID associated with the task segment number "P1 Task_Central Business District_8:00-9:00" is found to be "Drone_001," and its actual execution time period is from 8:00:00 to 8:59:59. The segment path node time period is processed according to the time dimension. The extracted path node correlation time period (8:00:00-8:59:59) is sorted chronologically and sliced ​​in 1-minute increments to generate a series of smaller path node time periods, such as [8:00:00-8:00:59], [8:01:00-8:01:59], etc. A path node correlation time period index table is identified, and a unique index number is assigned to each sliced ​​path node time period, recording its start and end times. Finally, the set of path node correlation task time periods is obtained. [{"index": "001", "time period": "8:00:00-8:00:59", "node ID": "Drone_001"}, {"index": "002", "time period": "8:01:00-8:01:59", "node ID": "Drone_001"}, ..., {"index": "060", "time period": "8:59:00-8:59:59", "node ID": "Drone_001"}].

[0026] S312: Based on the path node correlation task time period set, collect communication link interference assessment data of the same period segment, identify the communication link interference impact table, judge the daily communication link interference based on the interference threshold, match it with the path node correlation task time period, determine whether there is abnormal correlation of path node correlation, and generate a list of communication link interference affected segments. Obtain the task time period set for path node correlation, including 60 one-minute slices of the "Drone_001" drone in the "Central Business District" from 8:00:00 to 8:59:59. Request communication link quality log data for "Drone_001" within the corresponding time period from network monitoring, collecting average signal strength, latency, packet loss rate, etc., and construct a communication link interference impact table, as shown in Table 1. Set the communication link interference threshold: signal below -85dBm, or latency above 200ms, or packet loss above 5%. If the signal is -90dBm, latency is 150ms, and packet loss is 2% during the 8:05:00-8:05:59 period, then communication link interference is determined to exist. Match the interference period with the path node correlation task time period. If the path node correlation drops from 92% to 80% (below the historical average of 90%) during the interference period, an abnormal correlation is determined, and a list of communication link interference-affected segments is generated. Table 1: Examples of UAV Communication Link Interference Assessment Data See Table 1, which shows the communication link quality indicators of the UAV at different times to determine interference.

[0027] Please see Figure 5 The specific steps for obtaining the drone mission compliance adjustment set are as follows: S411: Based on the list of communication link interference affected sections, filter the nodes for task compliance verification, extract the task list, and obtain the task set of nodes not affected by communication link interference; Obtain a list of communication link interference-affected segments. For example, during the abnormally associated time period "8:05:00-8:05:59", the drone "Drone_001" shows communication interference and abnormal path correlation. When filtering task compliance verification nodes, exclude the associated nodes and tasks in the list that are affected by communication link interference and abnormal path correlation. Remove the task "Drone_001" during this time period from the compliance verification scope. Extract a detailed list of all currently executing or planned drone tasks from the cloud task scheduling platform, including task number, drone ID, scheduled take-off and landing time, planned route, operator ID, etc. Obtain the task set of nodes not affected by communication link interference. For example, after excluding the "Drone_001" task from 8:05:00-8:05:59, obtain a set including tasks such as "Drone_002" executing "Material Transportation Task 002" and "Drone_003" executing "Inspection Task 003".

[0028] S412: Call the task set of nodes that are not affected by communication link interference, match the task compliance specifications and operation data, extract the planned task volume by task number, count the number of operators and real-time operation time period, and generate a task compliance execution status matching dataset. Based on a task set of nodes unaffected by communication link interference, including tasks "Material Transportation Task 002" and "Inspection Task 003", the corresponding task type compliance standards are retrieved from a pre-set task compliance specification library. The "Material Transportation Task" requires a single flight time of no more than 30 minutes and a payload of no more than 5 kg. The "Inspection Task" requires a flight path deviation of no more than 5 meters and a flight altitude maintained between 50 and 80 meters. Real-time data collection of UAV flight control reported operation data, including actual flight time, payload, GPS track, and flight altitude, is collected. The operation data is compared item by item with the compliance specifications. Based on the task number, the corresponding task plan execution quantity or workload is extracted from the original task plan. For example, "Material Transportation Task 002" plans to transport 5 packages, and "Inspection Task 003" plans to inspect an area of ​​2 square kilometers. The operator ID and actual start and end times of each task are recorded. For example, "Material Transportation Task 002" is executed by operator "Op_A" from 8:10 to 8:35. A task compliance execution matching dataset is generated.

[0029] S413: Match the dataset based on the task compliance execution status, evaluate the degree of matching between task execution efficiency and operation distribution, identify efficiency fluctuation nodes, and use the formula: ; Calculate the fluctuation identification difference value, mark tasks with fluctuations exceeding the benchmark value as abnormal nodes, identify the task compliance execution matching deviation, and obtain the UAV task compliance adjustment set; Where Q represents the fluctuation identification difference value, Representing the The execution efficiency value of each task. This represents the average execution efficiency of all tasks. Representing the Standard deviation of the operation distribution for each task This represents the set benchmark fluctuation value. Represents the total number of tasks; Based on a dataset matching task compliance execution, this method evaluates the degree of matching between the execution efficiency and operation distribution of the UAV "Materials Transportation Mission 002". First, it calculates the execution efficiency of each task. Efficiency is defined as the ratio of the amount of work completed to the planned amount of work completed, or the ratio of the time taken to the planned time taken. For example, if the plan is to complete 10 packages but 9 are actually completed, the efficiency is 9 / 10. Secondly, calculate the operation distribution for each task. This reflects the balance of the workload through the standard deviation of operator working hours, measuring the dispersion of operator behavior (such as the interval between control command transmissions) during the execution of the task. For example, the standard deviation of operator operation intervals for a certain task is... The core objective is to assess whether the two are matched in order to identify situations where execution efficiency is high but the workload is uneven, i.e., a few operators are overloaded. To identify points of efficiency fluctuation, a baseline value for efficiency fluctuation was set. By analyzing the efficiency data of all tasks over the past month, it was found that the efficiency fluctuation was less than [a certain value]. This is normal fluctuation, exceeding To mitigate significant fluctuations, the baseline value is set to [value missing]. Introducing fluctuation identification difference value The formula for quantifying task compliance matching deviation is as follows: ,in, A larger value indicates a worse match between task efficiency and operation distribution; and Representing the first The execution efficiency of each task and the average execution efficiency of all tasks (current dataset is...) ), It is the first Standard deviation of the operation distribution for each task It is to set the benchmark fluctuation value Seconds indicate the time elapsed during the operation. Fluctuations within the standard deviation of seconds are acceptable. It is the total number of tasks (the current dataset contains) (One task) To demonstrate the calculation process of this formula, three tasks were selected for example analysis. Assume... Execution efficiency They are respectively Average execution efficiency Operational distribution standard deviation They are respectively Seconds, benchmark fluctuation value Seconds, substitute into the formula to calculate: For task 1, ; For task 2, ; For task 3 ; The final calculation yielded ; Through analysis of a large amount of historical data, a pre-set... The baseline threshold is ,when When the value exceeds this threshold, the task requires intervention and adjustment; the current calculation result... Significantly greater than This indicates that the selected task samples as a whole have a significant mismatch between task execution efficiency and operation distribution, and need to be marked as abnormal nodes; The advantage of this formula lies in its comprehensive consideration of both the absolute degree of deviation of task execution efficiency from the average and the dispersion of operation distribution, thus fully reflecting the combined performance of individual task execution efficiency and operator operational stability. This is particularly relevant when operator behavior fluctuates significantly. (relatively large), for The value has a greater impact, highlighting the importance of operational standardization. In this example, Task 2... The value is the largest, and its pair The value also contributed the most, indicating that the operator's behavior in Task 2 fluctuated significantly. If the value exceeds the baseline threshold, all tasks are marked, and a task compliance adjustment set is generated.

[0030] Please see Figure 6 The specific steps for obtaining the task progress monitoring structure indicators are as follows: S511: Call the drone mission compliance adjustment set, filter nodes that exceed the threshold, record the time interval and the magnitude of changes in the task volume, and obtain the set of abnormal progress fluctuation indicators; The drone mission compliance adjustment set is invoked, which includes mission "Task 2" marked as an abnormal node with a Q value of 0.059063. When filtering nodes that exceed the threshold, the Q value filtering threshold is set to 0.05 to identify abnormal nodes that need intervention and adjustment. Mission "Task 2" with a Q value greater than 0.05 is filtered as a node that exceeds the threshold. When recording the time interval and the change in task volume, the actual execution time interval (8:40-9:10) and the change in actual task volume (planned to transport 10 packages, actual transported 9, change in volume is -10%) are extracted from the original scheduling record of mission "Task 2". The progress fluctuation abnormal identifier set is obtained as [{"Task Number": "Task 2", "Execution Time Interval": "8:40-9:10", "Change in Task Volume": "-10%"}].

[0031] S512: Based on the task model resources corresponding to the nodes in the progress fluctuation anomaly identifier set, identify the node resource input ratio sequence, extract the abnormal distribution interval and compare the critical coefficient, record the ratio offset direction and node number, and form a task resource matching offset index group. Based on the task model resources corresponding to the nodes identified by the progress fluctuation anomaly identifier, including task "Task 2", whose task workload changes by -10%, when identifying the node resource input ratio sequence, the task model data corresponding to task "Task 2" is called. This model defines the standard resource input required to complete the task (standard UAV flight time, operator working hours, battery consumption), and is compared with the actual resource input to calculate the resource input ratio sequence. If the standard flight time of task "Task 2" is 25 minutes and the actual flight time is 30 minutes, then the flight time ratio is 1.2 (30 / 25). When extracting the abnormal distribution interval and comparing the critical coefficient, statistical analysis is performed on the resource input ratio sequence. An abnormal distribution interval deviating from the normal range was identified. During the execution of "Task 2", the flight time ratio was consistently higher than 1.2 in the interval of 9:00-9:10. The preset critical coefficient was 1.1. This coefficient was obtained by analyzing the resource input ratio that caused task delays or resource waste in historical data. When the ratio exceeded 1.1, it was considered that there was a significant abnormality in resource input. The flight time ratio of 1.2 was compared with the critical coefficient of 1.1. Since 1.2 is greater than 1.1, this interval was identified as an abnormal distribution interval. When recording the ratio offset direction and node number, the resource input ratio offset direction (higher than the standard value) and the corresponding task node number were also recorded to form the task resource matching offset index group.

[0032] S513: Based on the task resource matching offset index group, extract the task model resource deployment and task execution time period, identify the difference between the resource deployment cycle and the operation cycle, and sort and label them according to the progress benchmark to generate task progress monitoring structure indexes. Based on the task resource matching offset index group, it is found that the flight time resource investment of task "Task 2" is too high, with an abnormal range of 9:00-9:10. When extracting the task model resource deployment and task execution time period, the planned resource deployment time period of task "Task 2" (planned deployment of UAV from 8:40 to 9:05) is extracted from the task model, and its actual task execution time period (actual 8:40-9:10) is extracted from the actual execution log. When identifying the difference between the resource deployment cycle and the operation cycle, the planned resource deployment time period and the actual task execution time period are compared and the difference between the two is calculated. The planned resource deployment is 25 minutes, and the actual execution is 30 minutes, with a difference of 5 minutes. When sorting and labeling according to the progress benchmark, a preset progress benchmark value is set. A delay is defined as the task execution time exceeding the planned time by 10%. The current execution time of task "Task 2" exceeds the planned time by ((30-25) / 25)×100%=20%, which is far beyond the 10% benchmark value. Therefore, it is labeled as "severe delay", and the task progress monitoring structure index is generated.

[0033] The cloud-based drone path optimization system is used to execute the above-mentioned cloud-based drone path optimization method. The system includes: The task priority deviation extraction module obtains task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extracts the task priority fluctuation range and resource occupancy intensity distribution characteristics, compares the task priority demand with the real-time distribution, and generates task priority deviation labels. The path optimization mode classification module locates the path optimization mode that is suitable for the characteristics of the current task based on the task priority deviation label, extracts the task node identifier, process node and baseline path correlation degree, calculates the difference between the on-site path correlation degree and the baseline path correlation degree, classifies and labels the difference type, and generates path optimization mode matching labels. The communication link interference identification module uses path optimization pattern matching tags to filter path node correlation lagging task components, locate corresponding segments, extract communication link interference assessment interference time periods, determine overlap with operation time periods, filter frequently interfered segments, and generate a list of communication link interference affected segments. The task compliance diagnosis module is based on the list of communication link interference-affected segments, removes tasks in the interfering segments, extracts task compliance specifications and operation data, matches task assignment and operation time periods, calculates the ratio of operation volume to task, identifies tasks with dense tasks and inefficient execution of components, and obtains the UAV task compliance adjustment set. The resource allocation analysis module, based on the UAV mission compliance adjustment set, locates the resource input records of the mission in the mission model, extracts the ratio of mission quantity to resource quantity allocation, compares the difference between the operation cycle and the input cycle, maps the mission resource usage and progress status, and forms a mission progress monitoring structure index.

[0034] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A cloud-based drone path optimization method, characterized in that, Includes the following steps: S1: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, and generate task priority deviation labels; S2: Based on the deviation amount and pattern classification in the task priority deviation label, select the path optimization mode that matches the pattern classification from a variety of preset path optimization modes, extract the path node correlation degree distribution range, and perform deviation correction and classification processing with the baseline path correlation degree to obtain the path optimization mode matching label. S3: Call the path optimization mode matching label, extract the path node correlation segment number, identify the communication link interference assessment data in the segment, compare the communication link interference with the path node segment, record the number of interference overlap time periods, and generate a list of communication link interference affected segments. S4: Based on the list of communication link interference affected sections, filter task compliance verification nodes, extract task compliance specifications and operation data, map the distribution of task content and compliance requirements, determine whether there is any deviation in task compliance, and obtain the UAV task compliance adjustment set.

2. The cloud-based drone path optimization method according to claim 1, characterized in that, The task priority deviation label includes deviation number, priority status label, resource usage deviation amount, and mode classification. The path optimization mode matching label includes fluctuation type, benchmark comparison result, and task association number. The communication link interference affected segment list includes device type, affected time segment, number of overlapping time periods, and affected task number. The UAV task compliance adjustment set includes task distribution unevenness number, operation duration record, task completion deviation amount, and operation matching degree.

3. The cloud-based drone path optimization method according to claim 1, characterized in that, The specific steps for obtaining the task priority deviation label are as follows: S111: Obtain task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extract task priority fluctuation range and resource occupancy intensity distribution characteristics, match task priority requirements with real-time distribution, compare time range with priority requirements, and generate task record time periods for partition nodes. S112: Based on the task record time period of the partition node, extract the overlapping time period between the task priority status and the required time interval, calculate the ratio of overlapping time to the total required time, filter nodes with an overlap ratio lower than the benchmark value, and obtain the priority coverage deviation rate of the partition node according to the number of priority status annotations. S113: Based on the partition node priority coverage deviation rate, determine the deviation status of the node number, identify the node number whose deviation rate exceeds the node synchronization threshold, integrate the node number, priority coverage information and deviation rate value, and generate a task priority deviation label.

4. The cloud-based drone path optimization method according to claim 3, characterized in that, The specific steps for obtaining the path optimization mode matching tags are as follows: S211: Based on the task priority deviation label, identify the path optimization mode and path node correlation distribution characteristics that are adapted to the characteristics of the current task, extract the start and end fluctuation range of real-time path node correlation, calculate the start and end fluctuation difference of the path node correlation segment, compare it with the baseline path correlation fluctuation range, and obtain the path node correlation fluctuation deviation value. S212: Call the path node correlation fluctuation deviation value, combine the segment distribution, fluctuation trend and adjustment frequency, uniformly collect the path optimization mode deviation data, identify and calculate the adaptability deviation degree according to the segment number, determine the fluctuation direction according to the adjustment frequency, and obtain the path optimization mode matching label.

5. The cloud-based drone path optimization method according to claim 4, characterized in that, The specific steps for obtaining the list of communication link interference-affected segments are as follows: S311: Call the path optimization mode matching label, filter the task segment number of the path node correlation deviation, extract the path node correlation time period according to the node correlation table, process the segment path node time period according to the time dimension, identify the path node correlation time period index table, and obtain the path node correlation task time period set. S312: Based on the path node correlation task time period set, collect communication link interference assessment data for the same period segment, identify the communication link interference impact table, determine the daily communication link interference based on the interference threshold, match it with the path node correlation task time period, determine whether there is abnormal correlation of path node correlation, and generate a list of communication link interference affected segments.

6. The cloud-based drone path optimization method according to claim 5, characterized in that, The specific steps for obtaining the drone mission compliance adjustment set are as follows: S411: Based on the list of communication link interference affected sections, filter the task compliance verification nodes, extract the task list, and obtain the task set of nodes not affected by communication link interference. S412: Call the task set of nodes not affected by communication link interference, match the task compliance specifications and operation data, extract the planned task quantity by task number, count the number of operators and real-time operation time period, and generate a task compliance execution status matching dataset. S413: Based on the task compliance execution status matching dataset, evaluate the degree of matching between task execution efficiency and operation distribution, identify efficiency fluctuation nodes, calculate fluctuation identification difference value, mark tasks with fluctuations exceeding the benchmark value as abnormal nodes, identify task compliance execution matching deviations, and obtain the UAV task compliance adjustment set.

7. The cloud-based drone path optimization method according to claim 1, characterized in that, The method also includes step S5: S5: Call the UAV mission compliance adjustment set, extract the abnormal fluctuation mission group, calculate the ratio of resource input to mission volume in the mission model, record the difference distribution between resource deployment cycle and mission execution cycle, and generate mission progress monitoring structure indicators. The task progress monitoring structure indicators include resource allocation ratio, task intensity level, execution cycle difference, and task efficiency indicators.

8. The cloud-based drone path optimization method according to claim 7, characterized in that, The specific steps for obtaining the task progress monitoring structure indicators are as follows: S511: Call the UAV mission compliance adjustment set, filter nodes that exceed the threshold, record the time interval and the magnitude of changes in the task volume, and obtain the progress fluctuation abnormality identifier set; S512: Based on the task model resources corresponding to the nodes in the progress fluctuation anomaly identifier set, identify the node resource input ratio sequence, extract the abnormal distribution interval and compare the critical coefficient, record the ratio offset direction and node number, and form a task resource matching offset index group. S513: Based on the task resource matching offset index group, extract the task model resource deployment and task execution time period, identify the difference between the resource deployment cycle and the operation cycle, and sort and label them according to the progress benchmark to generate task progress monitoring structure indexes.

9. A cloud-based drone path optimization system, characterized in that, The system is used to implement the cloud-based drone path optimization method according to any one of claims 1-8, the system comprising: The task priority deviation extraction module obtains task priority distribution information, resource occupancy status map and communication link quality assessment data from the cloud-based UAV task scheduling platform, extracts the task priority fluctuation range and resource occupancy intensity distribution characteristics, compares the task priority demand with the real-time distribution, and generates task priority deviation labels. The path optimization mode classification module locates the path optimization mode that matches the characteristics of the current task based on the task priority deviation label, extracts the task node identifier, process node and baseline path correlation degree, calculates the difference between the on-site path correlation degree and the baseline path correlation degree, classifies and labels the difference type, and generates path optimization mode matching labels. The communication link interference identification module, based on the path optimization mode matching label, filters path node correlation lag task components, locates the corresponding segment, extracts the communication link interference assessment interference period, determines overlap with the operation period, filters frequently interfered segments, and generates a list of communication link interference affected segments. The task compliance diagnosis module, based on the list of communication link interference-affected segments, removes tasks in the interference segments, extracts task compliance specifications and operation data, matches task assignment and operation time periods, calculates the ratio of operation volume to task, identifies tasks with dense tasks and inefficient execution of components, and obtains the UAV task compliance adjustment set. Based on the UAV mission compliance adjustment set, the resource allocation analysis module locates the resource input records of the mission in the mission model, extracts the ratio of mission quantity to resource quantity allocation, compares the difference between the operation cycle and the input cycle, maps the mission resource usage and progress status, and forms a mission progress monitoring structure index.

Images