Task-driven autonomous inspection decision-making method for UAVs
By employing an autonomous inspection decision-making method for unmanned aerial vehicles (UAVs), and utilizing 3D modeling and data acquisition requirements analysis, the data acquisition path of UAVs in complex environments is optimized, solving the problem of insufficient data acquisition by UAVs in complex environments and achieving efficient and accurate data acquisition.
Patent Information
- Application Number
- CN202411166015.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-08-23
AI Technical Summary
In complex environments, drones have difficulty adjusting their data collection paths to meet mission requirements, resulting in insufficient data collection.
By employing methods such as 3D modeling, data collection needs analysis, path formation, weight adjustment, data comparison and decomposition, and target adjustment step sequencing, an autonomous inspection decision-making scheme is formed to optimize the data collection path.
It improves the efficiency of data acquisition by drones in complex environments, reduces environmental identification errors, and enhances the accuracy and speed of adjustments.
Smart Images

Figure CN118672285B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle automatic control, and in particular to a task-driven unmanned aerial vehicle autonomous inspection decision-making method. Background Art
[0002] UAV inspections can greatly improve the efficiency of target inspections in complex backgrounds. Compared with manual inspections, drone aerial inspections have many advantages, such as fast speed and strong maneuverability. They can autonomously perform inspection tasks, fly around obstacles, reconstruct images in three dimensions, and transmit high-definition images back, making up for the blind spots that may exist in manual inspections. The efficiency and quality of inspections can be improved by analyzing images and videos in real time.
[0003] However, due to the complex inspection environment, the data collected by drones during inspection may not meet the mission requirements. When the mission requirements cannot be met, due to the complexity of the environment, it is difficult for drones to adjust the collection path by themselves to complete the data collection requirements required for the mission. Summary of the invention
[0004] In order to solve the above technical problems, a task-driven UAV autonomous inspection decision-making method is provided. This technical solution solves the problem raised in the above background technology that due to the complex inspection environment, the data collected by the UAV during the inspection may not meet the task requirements. When the task requirements cannot be met, due to the complexity of the environment, it is difficult for the UAV to adjust the collection path by itself to complete the data collection requirements required for the task.
[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:
[0006] The task-driven UAV autonomous inspection decision-making method includes:
[0007] The drone receives the inspection mission and performs 3D modeling of the environment;
[0008] Based on the inspection task, at least one data collection requirement is obtained through analysis;
[0009] Form inspection routes based on data collection requirements and 3D modeling;
[0010] Use the inspection path to collect actual inspection data;
[0011] Acquire at least one basic collection requirement based on historical data, and form a collection adjustment plan for the basic collection requirement, wherein the collection adjustment plan consists of at least one adjustment step;
[0012] forming an adjustment weight for at least one adjustment step, wherein the adjustment weight is used to quantify the adjustment effect of the adjustment step on data collection;
[0013] Compare the actual inspection data with the data collection requirements to determine whether the actual inspection data meets the data collection requirements. When the actual inspection data meets the data collection requirements, no processing is performed. When the actual inspection data does not meet the data collection requirements, the data collection requirements are decomposed based on the basic collection requirements to obtain at least one target basic collection requirement.
[0014] Summarize at least one target basic acquisition requirement to obtain a target basic acquisition requirement set;
[0015] Merging the collection adjustment plans corresponding to at least one target basic collection requirement to obtain a target adjustment plan;
[0016] The adjustment steps in the target adjustment scheme are used as target adjustment steps, and the target adjustment steps are sorted from large to small according to the adjustment weights to obtain a target adjustment step sequence;
[0017] Use the target adjustment steps in the target adjustment step sequence in sequence to adjust data collection;
[0018] After each target adjustment step is used to adjust data collection, the target basic collection requirements are deleted from the target basic collection requirements set based on the collected and summarized data;
[0019] When the target basic collection requirement set becomes an empty set, the drone completes the data collection requirement.
[0020] Preferably, the analyzing and obtaining at least one data collection requirement based on the inspection task includes the following steps:
[0021] The inspection tasks are divided according to the inspection locations to obtain at least one basic inspection task, where the basic inspection task is a part of the inspection tasks corresponding to the same inspection location;
[0022] The basic inspection tasks are classified according to the types of data requirements to obtain at least one data collection requirement.
[0023] Preferably, forming an inspection path based on data collection requirements and 3D modeling includes the following steps:
[0024] Obtain at least one inspection location involved in data collection requirements in 3D modeling;
[0025] At least one single link is formed for at least one inspection location, where the single link passes through the inspection location only once;
[0026] The shortest single link is obtained as the inspection path.
[0027] Preferably, acquiring at least one basic collection requirement based on historical data and forming a collection adjustment plan for the basic collection requirement comprises the following steps:
[0028] At least one collection environment facing basic collection requirements is obtained based on historical data, and at least one target collection environment is selected from the at least one collection environment, where the data collected by the drone in the target collection environment cannot meet the basic collection requirements;
[0029] In the target collection environment, the missing data of the data collected by the drone is obtained relative to the basic collection requirements;
[0030] At least one sampling point is evenly set in the target collection environment, and the drone collects data at the sampling point. When the data collected at the sampling point overlaps with the missing data, the sampling point is used as the target sampling point, and the overlapping part of the data collected at the target sampling point and the missing data is used as the target missing data;
[0031] Obtaining a sampling point corresponding to at least one target sampling point where the target missing data summary data covers the missing data as an adjustment point;
[0032] Aggregate at least one adjustment point into an adjustment point set, and take a subset of the adjustment point set as a test set;
[0033] A trajectory formed by traversing the adjustment points in the test set is used as a test trajectory, and the test trajectory is used for data collection to obtain test data;
[0034] Obtain the set with the least number of elements in the test set where the test data covers the missing data as the adjustment set;
[0035] The test track corresponding to the adjustment set is used as the adjustment step of the target acquisition environment;
[0036] The adjustment steps of at least one target acquisition environment are summarized as an acquisition adjustment plan, and the acquisition adjustment plan is paired with a basic acquisition requirement corresponding to the target acquisition environment.
[0037] Preferably, forming an adjustment weight for at least one adjustment step comprises the following steps:
[0038] Extracting features of the target acquisition environment to obtain at least one indicator feature, and taking the total number of the at least one indicator feature as the number of features;
[0039] Identify indicator features in at least one acquisition environment, and when the number of indicator features appearing in an area of a preset size in the acquisition environment accounts for a proportion of the number of features that exceeds a preset proportion, the area of the preset size is regarded as a similar area;
[0040] The total area of the similar regions in at least one acquisition environment is divided by the total area of the at least one acquisition environment to obtain a similarity value of the target acquisition environment, and the similarity value of the target acquisition environment is used as an adjustment weight of the adjustment step corresponding to the target acquisition environment.
[0041] Preferably, decomposing the data collection requirements to obtain at least one target basic collection requirement comprises the following steps:
[0042] Compare the data collection requirement with at least one basic collection requirement, and when the basic collection requirement appears in the data collection requirement, use the data collection requirement as the target basic collection requirement;
[0043] Repeat the previous step to continuously obtain the target basic collection requirements. When at least one target basic collection requirement obtained by comparison covers the data collection requirements for the first time, the decomposition of the data collection requirements is completed.
[0044] Preferably, the step of merging the acquisition adjustment schemes corresponding to at least one target basic acquisition requirement to obtain the target adjustment scheme comprises the following steps:
[0045] The adjustment steps in the collection adjustment plan corresponding to the target basic collection requirements are used as feature adjustment steps;
[0046] Combining at least one feature adjustment step to obtain a pre-adjustment solution;
[0047] Taking one of at least one characteristic adjustment step as the adjustment step to be tested;
[0048] Comparing at least one characteristic adjustment step, when the adjustment step to be tested can be included in the remaining at least one characteristic adjustment step, deleting the adjustment step to be tested in the pre-adjustment scheme;
[0049] When the adjustment step to be tested traverses at least one characteristic adjustment step, the merging is completed and the pre-adjustment scheme is used as the target adjustment scheme.
[0050] Preferably, the deleting of the target basic collection requirements in the target basic collection requirement set based on the collected and summarized data comprises the following steps:
[0051] Compare the collected and summarized data with the target basic collection requirements in the target basic collection requirements set. When the collected and summarized data covers all or part of the target basic collection requirements in the target basic collection requirements set, delete the part covered by the collected and summarized data from the target basic collection requirements in the target basic collection requirements set.
[0052] When the target basic collection requirement is reduced to 0, the target basic collection requirement is deleted from the target basic collection requirement set.
[0053] Compared with the prior art, the present invention has the following beneficial effects:
[0054] By forming a collection adjustment plan, forming adjustment weights for adjustment steps, decomposing data collection requirements and deleting target basic collection requirements in the target basic collection requirement set, various adjustment steps can be generated for complex environments, and based on the decomposition of data collection requirements, adjustments can be made to the target adjustment plan. The target adjustment plan includes adjustments for different environments, so that it can avoid environmental identification and reduce errors caused by handling complex environments. When making adjustments, the target adjustment steps with more obvious adjustment effects are given priority, thereby improving the efficiency of the adjustment and completing data collection requirements faster. As a result, the drone can make decisions on its own and adopt better decision-making plans. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 It is a flowchart of the task-driven UAV autonomous inspection decision-making method of the present invention;
[0056] Figure 2 A schematic diagram of a process of analyzing and obtaining at least one data collection requirement based on an inspection task of the present invention;
[0057] Figure 3 A schematic diagram of a process for forming an inspection path based on data collection requirements and 3D modeling in the present invention;
[0058] Figure 4 A schematic diagram of a process of acquiring at least one basic collection requirement based on historical data and forming a collection adjustment plan for the basic collection requirement according to the present invention;
[0059] Figure 5 A schematic diagram of a process of forming an adjustment weight for at least one adjustment step of the present invention;
[0060] Figure 6 A schematic diagram of a process for decomposing data collection requirements to obtain at least one target basic collection requirement;
[0061] Figure 7 A flow chart of merging acquisition adjustment schemes corresponding to at least one target basic acquisition requirement of the present invention to obtain a target adjustment scheme;
[0062] Figure 8 The present invention is a flow chart of deleting target basic collection requirements in a target basic collection requirement set based on collected and summarized data. DETAILED DESCRIPTION
[0063] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.
[0064] Reference Figure 1 As shown in the figure, the task-driven UAV autonomous inspection decision-making method includes:
[0065] The drone receives the inspection mission and performs 3D modeling of the environment;
[0066] Based on the inspection task, at least one data collection requirement is obtained through analysis;
[0067] Form inspection routes based on data collection requirements and 3D modeling;
[0068] Use the inspection path to collect actual inspection data;
[0069] Acquire at least one basic collection requirement based on historical data, and form a collection adjustment plan for the basic collection requirement, wherein the collection adjustment plan consists of at least one adjustment step;
[0070] forming an adjustment weight for at least one adjustment step, wherein the adjustment weight is used to quantify the adjustment effect of the adjustment step on data collection;
[0071] Compare the actual inspection data with the data collection requirements to determine whether the actual inspection data meets the data collection requirements. When the actual inspection data meets the data collection requirements, no processing is performed. When the actual inspection data does not meet the data collection requirements, the data collection requirements are decomposed based on the basic collection requirements to obtain at least one target basic collection requirement.
[0072] Summarize at least one target basic acquisition requirement to obtain a target basic acquisition requirement set;
[0073] Merging the collection adjustment plans corresponding to at least one target basic collection requirement to obtain a target adjustment plan;
[0074] The adjustment steps in the target adjustment scheme are used as target adjustment steps, and the target adjustment steps are sorted from large to small according to the adjustment weights to obtain a target adjustment step sequence;
[0075] Use the target adjustment steps in the target adjustment step sequence in sequence to adjust data collection;
[0076] After each target adjustment step is used to adjust data collection, the target basic collection requirements are deleted from the target basic collection requirements set based on the collected and summarized data;
[0077] When the target basic collection requirement set becomes an empty set, the drone completes the data collection requirement.
[0078] In this scheme, at least one basic collection requirement is formed. At least one basic collection requirement can be combined to form any task requirement. For the basic collection requirement, adjustment plans are formed in different environments. Thus, the adjustment plans are summarized. Since the plans include different environmental factors, when making adjustments according to the plans, there is no need to identify the environment. Because the actual environment is extremely complex, it is necessary to design extremely complex algorithms to correctly identify it. Otherwise, the recognition accuracy is insufficient, and the adjustment based on the recognition result is difficult to meet the needs. At the same time, it takes a lot of computing power and time during identification. It is faster than pre-setting an adjustment plan containing various environmental factors in this scheme. When adjusting, it is sorted and adjusted according to the frequency of occurrence of the environment corresponding to each adjustment step, and the adjustment steps with a higher probability of occurrence in the corresponding environment are given priority. Thus, the efficiency of the adjustment can be further improved.
[0079] Reference Figure 2 As shown, based on the inspection task, analyzing and obtaining at least one data collection requirement includes the following steps:
[0080] The inspection tasks are divided according to the inspection locations to obtain at least one basic inspection task, where the basic inspection task is a part of the inspection tasks corresponding to the same inspection location;
[0081] The basic inspection tasks are classified according to the types of data requirements to obtain at least one data collection requirement.
[0082] There may be multiple inspection locations included in the inspection task. Therefore, in order to facilitate the subsequent status of the plan, the tasks need to be classified according to the inspection locations. Otherwise, tasks at different locations will be mixed together for adjustment.
[0083] Reference Figure 3 As shown, based on data collection requirements and 3D modeling, forming an inspection path includes the following steps:
[0084] Obtain at least one inspection location involved in data collection requirements in 3D modeling;
[0085] At least one single link is formed for at least one inspection location, where the single link passes through the inspection location only once;
[0086] The shortest single link is obtained as the inspection path.
[0087] The inspection path uses the shortest path, which can effectively reduce the inspection time. Otherwise, a longer path will increase the inspection time.
[0088] Reference Figure 4As shown, obtaining at least one basic collection requirement based on historical data and forming a collection adjustment plan for the basic collection requirement includes the following steps:
[0089] At least one collection environment facing basic collection requirements is obtained based on historical data, and at least one target collection environment is selected from the at least one collection environment, where the data collected by the drone in the target collection environment cannot meet the basic collection requirements;
[0090] In the target collection environment, the missing data of the data collected by the drone is obtained relative to the basic collection requirements;
[0091] At least one sampling point is evenly set in the target collection environment, and the drone collects data at the sampling point. When the data collected at the sampling point overlaps with the missing data, the sampling point is used as the target sampling point, and the overlapping part of the data collected at the target sampling point and the missing data is used as the target missing data;
[0092] Obtaining a sampling point corresponding to at least one target sampling point where the target missing data summary data covers the missing data as an adjustment point;
[0093] Aggregate at least one adjustment point into an adjustment point set, and take a subset of the adjustment point set as a test set;
[0094] A trajectory formed by traversing the adjustment points in the test set is used as a test trajectory, and the test trajectory is used for data collection to obtain test data;
[0095] Obtain the set with the least number of elements in the test set where the test data covers the missing data as the adjustment set;
[0096] The test track corresponding to the adjustment set is used as the adjustment step of the target acquisition environment;
[0097] The adjustment steps of at least one target acquisition environment are summarized as an acquisition adjustment plan, and the acquisition adjustment plan is paired with a basic acquisition requirement corresponding to the target acquisition environment.
[0098] At least one basic collection requirement can be combined to form any task requirement. Therefore, once the task requirement is decomposed, as long as a collection adjustment plan can be formed for each basic collection requirement, the collection adjustment plan can be summarized to complete the adjustment of the task requirement. In this solution, not only a collection adjustment plan is formed for the basic collection requirement, but also the number of steps in the collection adjustment plan is ensured to be as small as possible, thereby improving its adjustment efficiency. Therefore, when forming a collection adjustment plan, the set with the least number of elements in the test set in which the test data covers the missing data is obtained as the adjustment set.
[0099] Reference Figure 5As shown, forming an adjustment weight for at least one adjustment step includes the following steps:
[0100] Extracting features of the target acquisition environment to obtain at least one indicator feature, and taking the total number of the at least one indicator feature as the number of features;
[0101] Identify indicator features in at least one acquisition environment, and when the number of indicator features appearing in an area of a preset size in the acquisition environment accounts for a proportion of the number of features that exceeds a preset proportion, the area of the preset size is regarded as a similar area;
[0102] The total area of the similar regions in at least one acquisition environment is divided by the total area of the at least one acquisition environment to obtain a similarity value of the target acquisition environment, and the similarity value of the target acquisition environment is used as an adjustment weight of the adjustment step corresponding to the target acquisition environment.
[0103] The collection adjustment plan corresponds to the basic collection requirements, but the basic collection requirements may face multiple data collection environments. Therefore, adjustment steps are formed for each target collection environment. However, since the probability of the target collection environment occurring in actual situations is different, in order to improve the efficiency of the adjustment, it is necessary to use the adjustment steps corresponding to the environment with a high probability of occurrence. Therefore, it is necessary to form an adjustment weight to describe the probability of occurrence of the target collection environment corresponding to the adjustment step.
[0104] Reference Figure 6 As shown, decomposing the data collection requirements to obtain at least one target basic collection requirement includes the following steps:
[0105] Compare the data collection requirement with at least one basic collection requirement, and when the basic collection requirement appears in the data collection requirement, use the data collection requirement as the target basic collection requirement;
[0106] Repeat the previous step to continuously obtain the target basic collection requirements. When at least one target basic collection requirement obtained by comparison covers the data collection requirements for the first time, the decomposition of the data collection requirements is completed.
[0107] Reference Figure 7 As shown, merging the acquisition adjustment schemes corresponding to at least one target basic acquisition requirement to obtain the target adjustment scheme includes the following steps:
[0108] The adjustment steps in the collection adjustment plan corresponding to the target basic collection requirements are used as feature adjustment steps;
[0109] Combining at least one feature adjustment step to obtain a pre-adjustment solution;
[0110] Taking one of at least one characteristic adjustment step as the adjustment step to be tested;
[0111] Comparing at least one characteristic adjustment step, when the adjustment step to be tested can be included in the remaining at least one characteristic adjustment step, deleting the adjustment step to be tested in the pre-adjustment scheme;
[0112] When the adjustment step to be tested traverses at least one characteristic adjustment step, the merging is completed and the pre-adjustment scheme is used as the target adjustment scheme.
[0113] There may be duplication in the adjustment steps of the direct merging scheme. Therefore, in order to avoid repeated adjustments, the adjustment steps need to be deduplicated. The time spent on deduplication is much less than the time spent by the drone to adjust according to the adjustment steps, thereby reducing the overall adjustment time.
[0114] Reference Figure 8 As shown, based on the collected and summarized data, deleting the target basic collection requirements in the target basic collection requirements set includes the following steps:
[0115] Compare the collected and summarized data with the target basic collection requirements in the target basic collection requirements set. When the collected and summarized data covers all or part of the target basic collection requirements in the target basic collection requirements set, delete the part covered by the collected and summarized data from the target basic collection requirements in the target basic collection requirements set.
[0116] When the target basic collection requirement is reduced to 0, the target basic collection requirement is deleted from the target basic collection requirement set.
[0117] The data obtained after each adjustment can partially meet the collection requirements. Therefore, real-time updates are performed to delete or partially delete the satisfied requirements. When the target basic collection requirements in the target basic collection requirements set are completely deleted, it means that the data collected by the adjusted path and the original path can meet the task requirements.
[0118] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned task-driven UAV autonomous inspection decision method is executed.
[0119] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).
[0120] In summary, the advantages of the present invention are: by forming a collection adjustment plan, forming adjustment weights for adjustment steps, decomposing data collection requirements and deleting target basic collection requirements in the target basic collection requirement set, various adjustment steps can be generated for complex environments, and based on the decomposition of data collection requirements, adjustments are made to the target adjustment plan. The target adjustment plan includes adjustments under different environments, so that it can avoid environmental identification and reduce errors caused by handling complex environments. When making adjustments, the target adjustment steps with more obvious adjustment effects are given priority, thereby improving the efficiency of the adjustment and completing the data collection requirements faster. As a result, the drone can make decisions on its own and adopt better decision-making plans at the same time.
[0121] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. A task-driven UAV autonomous inspection decision-making method, characterized by: include: The drone receives the inspection mission and performs 3D modeling of the environment; Based on the inspection task, at least one data collection requirement is obtained through analysis; Form inspection routes based on data collection requirements and 3D modeling; Use the inspection path to collect actual inspection data; Acquire at least one basic collection requirement based on historical data, and form a collection adjustment plan for the basic collection requirement, wherein the collection adjustment plan consists of at least one adjustment step; forming an adjustment weight for at least one adjustment step, wherein the adjustment weight is used to quantify the adjustment effect of the adjustment step on data collection; Compare the actual inspection data with the data collection requirements to determine whether the actual inspection data meets the data collection requirements. When the actual inspection data meets the data collection requirements, no processing is performed. When the actual inspection data does not meet the data collection requirements, the data collection requirements are decomposed based on the basic collection requirements to obtain at least one target basic collection requirement. Summarize at least one target basic acquisition requirement to obtain a target basic acquisition requirement set; Merging the collection adjustment plans corresponding to at least one target basic collection requirement to obtain a target adjustment plan; The adjustment steps in the target adjustment scheme are used as target adjustment steps, and the target adjustment steps are sorted from large to small according to the adjustment weights to obtain a target adjustment step sequence; Use the target adjustment steps in the target adjustment step sequence in sequence to adjust data collection; After each target adjustment step is used to adjust data collection, the target basic collection requirements are deleted from the target basic collection requirements set based on the collected and summarized data; When the target basic collection requirement set becomes an empty set, the drone completes the data collection requirement; The obtaining of at least one basic collection requirement based on historical data and forming a collection adjustment plan for the basic collection requirement comprises the following steps: At least one collection environment facing basic collection requirements is obtained based on historical data, and at least one target collection environment is selected from the at least one collection environment, where the data collected by the drone in the target collection environment cannot meet the basic collection requirements; In the target collection environment, the missing data of the data collected by the drone is obtained relative to the basic collection requirements; At least one sampling point is evenly set in the target collection environment, and the drone collects data at the sampling point. When the data collected at the sampling point overlaps with the missing data, the sampling point is used as the target sampling point, and the overlapping part of the data collected at the target sampling point and the missing data is used as the target missing data; Obtaining a sampling point corresponding to at least one target sampling point where the target missing data summary data covers the missing data as an adjustment point; Aggregate at least one adjustment point into an adjustment point set, and take a subset of the adjustment point set as a test set; A trajectory formed by traversing the adjustment points in the test set is used as a test trajectory, and the test trajectory is used for data collection to obtain test data; Obtain the set with the least number of elements in the test set where the test data covers the missing data as the adjustment set; The test track corresponding to the adjustment set is used as the adjustment step of the target acquisition environment; Summarize the adjustment steps of at least one target collection environment as a collection adjustment plan, and pair the collection adjustment plan with the basic collection requirements corresponding to the target collection environment; The forming of adjustment weights for at least one adjustment step comprises the following steps: Extracting features of the target acquisition environment to obtain at least one indicator feature, and taking the total number of the at least one indicator feature as the number of features; Identify indicator features in at least one acquisition environment, and when the number of indicator features appearing in an area of a preset size in the acquisition environment accounts for a proportion of the number of features that exceeds a preset proportion, the area of the preset size is regarded as a similar area; The total area of the similar regions in at least one acquisition environment is divided by the total area of the at least one acquisition environment to obtain a similarity value of the target acquisition environment, and the similarity value of the target acquisition environment is used as an adjustment weight of the adjustment step corresponding to the target acquisition environment.
2. The task-driven UAV autonomous inspection decision-making method according to claim 1 is characterized in that: The step of analyzing and obtaining at least one data collection requirement based on the inspection task includes the following steps: The inspection tasks are divided according to the inspection locations to obtain at least one basic inspection task, where the basic inspection task is a part of the inspection tasks corresponding to the same inspection location; The basic inspection tasks are classified according to the types of data requirements to obtain at least one data collection requirement.
3. The task-driven autonomous inspection decision-making method for unmanned aerial vehicles according to claim 2 is characterized in that: The formation of the inspection path based on data collection requirements and 3D modeling includes the following steps: Obtain at least one inspection location involved in data collection requirements in 3D modeling; At least one single link is formed for at least one inspection location, where the single link passes through the inspection location only once; The shortest single link is obtained as the inspection path.
4. The task-driven UAV autonomous inspection decision-making method according to claim 3 is characterized in that: Decomposing the data collection requirements to obtain at least one target basic collection requirement comprises the following steps: Compare the data collection requirement with at least one basic collection requirement, and when the basic collection requirement appears in the data collection requirement, use the data collection requirement as the target basic collection requirement; Repeat the previous step to continuously obtain the target basic collection requirements. When at least one target basic collection requirement obtained by comparison covers the data collection requirements for the first time, the decomposition of the data collection requirements is completed.
5. The task-driven UAV autonomous inspection decision-making method according to claim 4 is characterized in that: The step of merging the acquisition adjustment schemes corresponding to at least one target basic acquisition requirement to obtain the target adjustment scheme comprises the following steps: The adjustment steps in the collection adjustment plan corresponding to the target basic collection requirements are used as feature adjustment steps; Combining at least one feature adjustment step to obtain a pre-adjustment solution; Taking one of at least one characteristic adjustment step as the adjustment step to be tested; Comparing at least one characteristic adjustment step, when the adjustment step to be tested is included in the remaining at least one characteristic adjustment step, deleting the adjustment step to be tested in the pre-adjustment scheme; When the adjustment step to be tested traverses at least one characteristic adjustment step, the merging is completed and the pre-adjustment scheme is used as the target adjustment scheme.
6. The task-driven autonomous inspection decision-making method for unmanned aerial vehicles according to claim 5 is characterized in that: The method of deleting the target basic collection requirements from the target basic collection requirements set based on the collected and summarized data includes the following steps: Compare the collected and summarized data with the target basic collection requirements in the target basic collection requirements set. When the collected and summarized data covers all or part of the target basic collection requirements in the target basic collection requirements set, delete the part covered by the collected and summarized data from the target basic collection requirements in the target basic collection requirements set. When the target basic collection requirement is reduced to 0, the target basic collection requirement is deleted from the target basic collection requirement set.
