Aviation robot dynamic task allocation and optimization method

Abstract

The invention discloses a dynamic task allocation and optimization method for aviation robots, relates to the technical field of task allocation, is used for solving the problems of equipment overload and frequent airspace conflicts, obtains a current task list of each aviation robot through a task database, calls a task level of a task target, collects residual electric quantity and calculates endurance time, and provides a dynamic task allocation and optimization method for the aviation robots. Screening and marking task targets in combination with task levels; importing position coordinates of the marked task targets and available time into a clustering analysis algorithm to generate candidate areas; marking overlapped parts of the candidate areas of different robots; obtaining a robot list of the marked candidate areas; according to the method, optimal scores are evaluated and executed in combination with track density characteristics, sorting is carried out in marked candidate areas according to the scores, a sorting result is imported into an analytic hierarchy process algorithm to generate a candidate robot list, and the safety and execution efficiency of a multi-aviation-robot collaborative task are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of task allocation, more particularly, the present application relates to an aviation robot dynamic task allocation and optimization method. BACKGROUND

[0002] With the development of unmanned aviation systems and multi-unmanned aerial vehicle cooperation technology, aviation robots are increasingly widely used in fields such as logistics distribution, environmental monitoring, disaster rescue, and military reconnaissance. In the prior art, when multiple aviation robots perform complex tasks, they usually use pre-planned or static task allocation methods for scheduling.

[0003] The prior art has the following disadvantages: Currently, the prior art relies on a static scheduling model driven by pure algorithms, lacks a dynamic control and feedback mechanism for the real-time physical state of aviation robots (such as power, motor temperature, and airspace occupation), and cannot achieve collaborative optimization of task scheduling algorithms and robot energy management, hardware protection, and airspace control, resulting in unbalanced energy distribution, equipment overload, and frequent airspace conflicts during task execution, which reduces the overall system operation safety and task completion efficiency. Therefore, an aviation robot dynamic task allocation and optimization method is proposed.

[0004] The above information disclosed in the background section is only intended to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present application provide an aviation robot dynamic task allocation and optimization method, which uses dual constraints of task level and endurance characteristics to perform intelligent filtering, clustering analysis, and spatial overlap identification of task targets, realizes adaptive division and conflict detection of candidate regions, combines multi-dimensional execution optimization scoring of motor temperature rise, power, and airspace occupation degree, and generates a candidate aviation robot list using an analytic hierarchy process algorithm to solve the problems raised in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solution, an aviation robot dynamic task allocation and optimization method, comprising the following steps: Step S1: Obtain the current task list corresponding to each aviation robot through the task database, retrieve the task level of the task target in the current task list, collect the current remaining power of the aviation robot to analyze the endurance time, and filter and mark the task target in the current task list according to the task level; Step S2: Import the location coordinates of the marked task target and the available time of the task into a clustering analysis algorithm to generate each candidate region, and mark the overlapping regions in the candidate regions corresponding to different aviation robots; Step S3: Obtain the aerial robot list of the marked candidate area, detect the motor temperature rise gradient and airspace occupation degree of each aerial robot in the aerial robot list, analyze the track density characteristics by using the airspace occupation degree, and evaluate the execution preference score of the aerial robot by combining the motor temperature rise gradient; Step S4: Sort the aerial robots in each marked candidate area by using the execution preference score, and import the sorting results of each marked candidate area into the analytic hierarchy process algorithm to generate a candidate aerial robot list.

[0007] In a preferred embodiment, in step S1, the identifier ID of each aerial robot is matched with the task database to obtain the corresponding current task target of each aerial robot, and the current task list of the aerial robot is integrated; The task ID of each task target in the current task list of the aerial robot is matched with the task database, and the hierarchical depth of the task target is counted as the level of the task target; If the task level of the task target is less than the preset task level threshold, the task target is determined to be high priority; If the task level of the task target is greater than or equal to the preset task level threshold, the task target is determined to be low priority.

[0008] In a preferred embodiment, in step S1, the remaining power of each aerial robot is read by calling the BMS interface of each aerial robot; The historical power consumption of each aerial robot is called, and the average historical power consumption when the same type of robot performs similar tasks is selected as the average power consumption of each aerial robot; The remaining power of each aerial robot is multiplied by the nominal capacity of the battery to obtain the remaining energy of each aerial robot, and the remaining energy of each aerial robot is divided by the average power consumption to obtain the endurance time of each aerial robot; The predicted execution time of each task target is read by traversing the current task list corresponding to each aerial robot; If the priority of the task target is high priority, and the predicted execution time of the task target is less than the endurance time of the corresponding aerial robot, the task target is marked; Otherwise, the task target is not marked.

[0009] In a preferred embodiment, in step S2, the location coordinates and task available time of each marked task target are obtained by matching each marked task target with the task database; The location coordinates and task available time of each marked task target are standardized and integrated into a spatiotemporal feature vector set; The standardized spatiotemporal feature vector set is input into a clustering algorithm model, and a judgment is made based on a pre-set logic; Assign a region number to each marked task target that conforms to the predefined logic; Marked task targets with the same region number are grouped together to form a candidate region.

[0010] In a preferred embodiment, in step S2, if a candidate region has some location coordinates that are the same as another candidate region, and the available time of the candidate region has repeated time periods, then it is determined that the space where the candidate region has the same location coordinates as another candidate region is marked to obtain a marked candidate region. Conversely, if the condition is not met, then no marking is performed.

[0011] In a preferred embodiment, in step S3, the aerial robots corresponding to each marked candidate region are integrated into an aerial robot list for each marked candidate region; Match the marked task target of each aerial robot in the aerial robot list of each marked candidate area with the historical task database to obtain the maximum motor temperature when each marked task target is executed; The current temperature of each aviation robot's motor is obtained by using the temperature sensor built into each aviation robot's motor. The ratio of the maximum motor temperature when the marked task target is executed to the current motor temperature of the corresponding aerial robot, and then to the corresponding expected execution time, is used as the motor temperature rise gradient for each aerial robot in the aerial robot list.

[0012] In a preferred embodiment, in step S3, the maximum value of the x-axis coordinate, the maximum value of the y-axis coordinate, and the maximum value of the z-axis coordinate in the position coordinates of each candidate marker region are subtracted from the minimum value of the x-axis coordinate, the minimum value of the y-axis coordinate, and the minimum value of the z-axis coordinate, respectively, to obtain the length, width, and height of each candidate marker region; The spatial size of each candidate region is obtained by calculating its length, width, and height using the ellipsoid approximation algorithm. The ratio of the spatial size of each candidate marking region to the spatial size required by the marking task target of each aerial robot in the aerial robot list corresponding to the candidate marking region is used as the airspace occupancy of each aerial robot in the aerial robot list.

[0013] In a preferred embodiment, in step S3, the average airspace occupancy of all aerial robots within a marked candidate area is calculated; The ratio of the airspace occupancy of each aerial robot corresponding to the candidate region to the average airspace occupancy of the candidate region is used as the track density feature of each aerial robot in the aerial robot list of the candidate region. The temperature rise gradient and track density characteristics of the motors of each aerial robot in each list of aerial robots are standardized to obtain the temperature rise gradient factor and track density factor. The optimal performance score for each aerial robot in each candidate marked region is calculated by combining the temperature gradient factor and the track density factor. ,in, The temperature gradient factor. For track density factor, The optimal performance score is given to each aerial robot in each marked candidate area.

[0014] In a preferred embodiment, in step S4, the aerial robots in the same candidate region are sorted in descending order according to their performance preference scores. The sorting results formed by each marked candidate region are input into the hierarchical analysis algorithm model. The aerial robot is used as the scheme layer, the optimization scoring is used as the criterion layer, and the candidate aerial robot list generation is used as the target layer. The sorting results of each marked candidate region are converted into relative weights by linear mapping and normalized to obtain the comprehensive weight of each aerial robot in the overall candidate list. After the overall weights are verified for consistency, the overall priority index of each aerial robot is output, which is used to generate the final list of candidate aerial robots.

[0015] The technical effects and advantages of this invention are as follows: This invention obtains the current task list corresponding to each aerial robot from a task database, retrieves the task level of the task target in the task list, collects the current remaining power of the aerial robot to calculate the endurance time, and filters and marks the task targets in combination with the task level. The position coordinates of the marked task targets and the available time of the task are imported into a clustering analysis algorithm to generate candidate regions. Overlapping regions in the candidate regions corresponding to different aerial robots are marked to obtain a list of aerial robots in the marked candidate regions. The motor temperature rise gradient and airspace occupancy of each aerial robot in the list are detected. The airspace occupancy is used to analyze the track density characteristics, and the execution optimization score of the aerial robot is evaluated in combination with the motor temperature rise gradient. The aerial robots are sorted in each marked candidate region according to the execution optimization score. The sorting results are imported into the hierarchical analysis algorithm to generate a candidate aerial robot list. This realizes dynamic control and coordination of energy endurance, hardware temperature rise and airspace occupancy during the task allocation process, reduces the execution risks caused by insufficient power, equipment overheating and airspace overlap, and improves the safety and execution efficiency of multi-aerial robot collaborative tasks. Attached Figure Description

[0016] Fig. 1 This is a flowchart illustrating the implementation of the dynamic task allocation and optimization method for aerial robots according to the present invention.

[0017] Fig. 2 This is a schematic diagram illustrating the steps of the dynamic task allocation and optimization method for aerial robots according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention obtains the current task list corresponding to each aerial robot from a task database, retrieves the task level of the task target in the task list, collects the current remaining power of the aerial robot to calculate the endurance time, and filters and marks the task targets in combination with the task level. The position coordinates of the marked task targets and the available time of the task are imported into a clustering analysis algorithm to generate candidate regions. Overlapping regions in the candidate regions corresponding to different aerial robots are marked to obtain a list of aerial robots in the marked candidate regions. The motor temperature rise gradient and airspace occupancy of each aerial robot in the list are detected. The airspace occupancy is used to analyze the track density characteristics, and the execution optimization score of the aerial robot is evaluated in combination with the motor temperature rise gradient. The aerial robots are sorted in each marked candidate region according to the execution optimization score. The sorting results are imported into the hierarchical analysis algorithm to generate a candidate aerial robot list. This realizes dynamic control and coordination of energy endurance, hardware temperature rise and airspace occupancy during task allocation, reducing the execution risks caused by insufficient power, equipment overheating and airspace overlap.

[0020] Example 1: A method for dynamic task allocation and optimization of aerial robots, such as... Figs. 1-2 As shown, it includes the following steps: Step S1: Obtain the current task list corresponding to each aerial robot through the task database, retrieve the task level of the task target in the current task list, collect the current remaining battery power of the aerial robot to analyze the endurance time, and filter and mark the task targets in the current task list in combination with the task level. Step S2: Import the location coordinates of the marked task target and the available time of the task into the clustering analysis algorithm to generate candidate regions. Mark the overlapping regions in the candidate regions corresponding to different aerial robots. Step S3: Obtain a list of aerial robots with marked candidate regions, detect the motor temperature rise gradient and airspace occupancy of each aerial robot in the list, analyze the trajectory density characteristics using the airspace occupancy, and evaluate the execution optimization score of the aerial robot in combination with the motor temperature rise gradient. Step S4: Sort the aerial robots in each marked candidate region using the performance optimization score, and import the sorting results of each marked candidate region into the hierarchical analysis algorithm to generate a candidate aerial robot list.

[0021] The specific implementation is as follows: In step S1, the identifier ID of each aerial robot is matched with the task database to obtain the current task target corresponding to each aerial robot, and then integrated into the current task list corresponding to each aerial robot. Iterate through the current task list corresponding to each aerial robot, and read the task ID field of each task target to obtain the task ID of each task target in the current task list corresponding to each aerial robot; Match the task ID of each task target in the current task list corresponding to each aerial robot with the task database, and read the parent task identifier field of each task target; If the parent task identifier field of the task target is not empty, then there is a parent task. Continue to query recursively until the top-level task is found, and the level depth of the task target is counted as the level of the task target. If the parent task identifier field of the task target is empty, then there is no parent task, and the task target is determined to be at level one. It should be noted that the identifier ID of an aerial robot is a unique identifier for each aerial robot. This identifier is unique, non-repeatable, and traceable within the system, ensuring accurate differentiation of individuals in multi-aerial robot collaborative tasks. The task database refers to the data storage unit used to record and manage information related to aerial robot tasks. The task ID of the task target is an identifier used to uniquely identify a specific task being performed and is the core index field of the task database. The parent task identifier field is a field contained in each sub-task record in the task database, which stores the unique identifier of its parent task, i.e., the parent task's task ID. The top-level task refers to a task directly generated, initiated, or issued by the system's main control terminal. It does not belong to any parent task, therefore its parent task identifier field is empty.

[0022] The priority of the task objectives in the current task list corresponding to each aerial robot is determined by comparing the task level with a preset task level threshold. If the task level of the task target in the current task list corresponding to each aerial robot is less than the preset task level threshold, then the task target is determined to be of high priority. If the task level of the task target in the current task list corresponding to each aerial robot is greater than or equal to the preset task level threshold, then the task target is determined to be of low priority. The remaining battery power of each aerial robot is read by calling its BMS interface; The historical power consumption of each aerial robot was retrieved, and the average historical power consumption of the same model of robot performing similar tasks was taken as the average power consumption of each aerial robot. The remaining power of each aerial robot is obtained by multiplying its remaining power by the nominal battery capacity. The remaining power of each aerial robot is divided by its average power consumption to obtain the endurance of each aerial robot. Iterate through the current task list corresponding to each aerial robot and read the estimated execution time of each task objective; The decision to mark a task is based on the priority and estimated execution time of each task objective in the current task list for each aerial robot: If the task objective has a high priority and its estimated execution time is less than the corresponding aerial robot's endurance, then the task objective is marked. Conversely, the task objective will not be marked.

[0023] It should be explained that the preset task level threshold is an important parameter for determining the priority of task objectives. The threshold is automatically adjusted according to the current hierarchical distribution of each objective task to keep the number of high-priority tasks in a reasonable proportion. For example, the level value corresponding to 20% of the total number of tasks can be taken as the threshold. The BMS interface refers to the data interaction and control interface provided by the battery management unit inside the aerospace robot. It is used to monitor the battery status, transmit power data, execute battery protection logic, and provide access channels for power-related information to the outside world. The nominal battery capacity refers to the rated charge or energy capacity that the battery can output under standard discharge conditions. The battery manufacturer provides the nominal capacity of the battery in the product specification. The estimated execution time refers to the estimated time required for the aerospace robot to perform a specific task objective from the start of the task to its completion. Furthermore, the core purpose of reading the remaining power and calculating the remaining battery life through the BMS interface in this step is to screen out tasks that can be safely executed without the risk of running out of power, based on the real-time energy status of the aerial robot. Essentially, it achieves pre-emptive control of the robot's energy safety during execution by monitoring energy data, avoiding the risk of the robot stopping midway due to the algorithm scheduling ignoring hardware energy limitations. In addition, the task hierarchy screening is not simply sorting, but prioritizing the execution of high-priority tasks within the robot's controllable energy range, reflecting the logic that task allocation follows physical control.

[0024] In step S2, each marked task target is matched with the task database to obtain the location coordinates and available time of each marked task target; It should be noted that the position coordinates of each marked task target refer to the specific location of each marked task target in three-dimensional space, which is used to determine the distribution of tasks in the airspace; the available time of each marked task target refers to the time window in which each task target can be executed in the airspace, that is, the allowed time range for the task to be executed by the aerial robot.

[0025] The location coordinates of each marked task target and the available time of the task are imported into the clustering analysis algorithm to generate each candidate region; The specific generation process is as follows: The location coordinates of each marked task target and the available time of the task are standardized and integrated into a spatiotemporal feature vector set; Density clustering algorithm was selected as the core analysis model. The core parameters for setting clustering algorithms mainly include: neighborhood radius and minimum number of samples; The standardized spatiotemporal feature vector set is input into the clustering algorithm model, and judgment is made based on pre-defined logic. The pre-defined logic is as follows: starting from a randomly labeled task target, check if there are more labeled task targets within its neighborhood radius than the minimum number of samples. If this logic holds, a new region is formed and continues to expand until no more marked task targets can be incorporated. If the logic is not followed, the point is marked as a noise point. Noise points do not participate in the generation of candidate regions in the current round, but their task objectives are still retained in the task list and will participate in cluster analysis again in the subsequent generation of candidate regions; Assign a region number to each marked task target that conforms to the predefined logic; Group the marked task targets with the same region number together to form a candidate region; For all marked task objectives within the same region number, the earliest start time is taken as the earliest start time of the region; the latest end time is taken as the latest end time of the region, and these are combined into the available time for the candidate region. Determine the candidate regions corresponding to different aerial robots: If the candidate region shares some location coordinates with one or more other candidate regions, and the available time of the candidate regions overlaps within the same time period, then the space where the candidate region shares the same location coordinates with one or more other candidate regions is marked to obtain a marked candidate region. Conversely, if the condition is not met, then no marking is performed.

[0026] It needs to be explained that the neighborhood radius is a distance and time threshold used to determine whether task targets belong to the same neighborhood. It includes the spatial neighborhood radius and the temporal neighborhood radius. The average spatial distance between historical task targets is calculated, and the spatial neighborhood radius is set to 1 to 2 times the average distance. The average time window length of the task is calculated, and the temporal neighborhood radius is set to half the average time window length. The minimum sample size refers to the minimum number of task targets that must be included within the neighborhood radius of a certain task target to determine that point as a core point. Based on the average and standard deviation of the number of task targets in the historical candidate domain, the minimum sample size is set to the difference between the average and standard deviation of the number of task targets in the historical candidate domain.

[0027] In step S3, the aerial robots corresponding to each candidate marking region are integrated into an aerial robot list for each candidate marking region; Match the marked task target of each aerial robot in the aerial robot list of each marked candidate area with the historical task database to obtain the maximum motor temperature when each marked task target is executed; The current temperature of each aviation robot's motor is obtained by using the temperature sensor built into each aviation robot's motor. The maximum temperature of the motor when the marked task target is executed is subtracted from the current temperature of the corresponding aviation robot's motor, and the ratio of this to the corresponding expected execution time is used as the motor temperature rise gradient for each aviation robot in the list of aviation robots. The length, width, and height of each candidate label region are obtained by subtracting the maximum x-axis coordinate, maximum y-axis coordinate, and maximum z-axis coordinate from the minimum x-axis coordinate, minimum y-axis coordinate, and minimum z-axis coordinate, respectively. The spatial size of each candidate region is obtained by calculating its length, width, and height using the ellipsoid approximation algorithm. The space required for marking the target of each aerial robot is obtained by following the same steps as above; The ratio of the spatial size of each candidate region to the spatial size required by the target of the corresponding aerial robot in the aerial robot list of the corresponding candidate region is used as the airspace occupancy of each aerial robot in the aerial robot list. Repeat the above steps to obtain the airspace occupancy of each aerial robot in the list of aerial robots; Calculate the average airspace occupancy of all aerial robots within a candidate region for marking; The ratio of the airspace occupancy of each aerial robot corresponding to the candidate region to the average airspace occupancy of the candidate region is used as the track density feature of each aerial robot in the aerial robot list of the candidate region. Repeat the above steps to obtain the track density characteristics of each aerial robot in the list of aerial robots; It should be noted that the historical mission database is a data storage unit used to store historical mission execution records and related performance parameters of the aerial robot, providing data support for current mission decision-making; the built-in temperature sensors of each aerial robot motor are devices used to collect the real-time temperature of the motor windings or the casing; the ellipsoid approximation algorithm is a method used to estimate the size of a three-dimensional irregular spatial region. It determines the major axis, width axis, and height axis of the ellipsoid by the boundary range of the region on the three spatial axes, thereby approximately describing the spatial occupancy of the entire region.

[0028] The temperature rise gradient and track density characteristics of the motors of each aerial robot in each list of aerial robots are standardized to obtain the temperature rise gradient factor and track density factor. The optimal performance score for each aerial robot in each candidate marked region is calculated by combining the temperature rise gradient factor and the track density factor. The calculation formula is as follows: ,in, The temperature gradient factor. For track density factor, Optimal performance scores are given to each aerial robot in each candidate marked area; The larger the temperature gradient factor, the larger the track density factor, the faster the motor temperature rises, the more crowded the space, the greater the risk of performing the task, and the lower the performance optimization score of each aerial robot in each marked candidate area. The smaller the temperature gradient factor, the smaller the track density factor, the slower the motor temperature rise, the more open the space, the lower the risk of performing the task, and the higher the execution optimization score of each aerial robot in each marked candidate area.

[0029] It should be explained that the space required to mark the task target refers to the minimum three-dimensional space range that the aerial robot needs to occupy to safely perform the task. This value is preset in the task planning stage based on the task type, the physical size of the aerial robot itself, and the safety margin, and is stored in the task database. The standardization processing methods include, but are not limited to, standard linear transformation based on interval scaling, Z-Score standardization method based on statistics, or normalization method based on nonlinear mapping function. The application methods of standardization processing will not be elaborated here. Furthermore, detecting the motor temperature rise gradient relies on real-time hardware data from the temperature sensor built into the aerospace robot motor. Its core function is to monitor whether the motor will exceed the safe temperature threshold during task execution (e.g., an excessively high temperature rise gradient indicates that the motor is prone to overheating). This is a dynamic control of the robot's hardware status. Calculating airspace occupancy and track density characteristics involves quantifying the proportion of physical airspace occupied to avoid multiple robots overlapping in the same airspace. Essentially, this is an active control of airspace conflicts. Together, they constitute the execution optimization score. The score directly corresponds to the control priority of the robot's safe task execution, rather than simply ranking tasks by efficiency.

[0030] In step S4, the aerial robots in each candidate marking region are sorted according to their performance optimization scores. The sorting rules are as follows: Compare aerial robots within the same candidate marker region; The aerial robots in the same candidate region are sorted in descending order based on their performance preference scores. The sorting results of each marked candidate region are imported into the hierarchical analysis algorithm to generate a candidate list of aerial robots.

[0031] The specific generation process is as follows: The sorting results formed by each marked candidate region are input into the hierarchical analysis algorithm model. The aerial robot is used as the scheme layer, the optimization scoring is used as the criterion layer, and the candidate aerial robot list generation is used as the target layer. The sorting results of each marked candidate region are converted into relative weights by linear mapping and normalized to obtain the comprehensive weight of each aerial robot in the overall candidate list. Perform a consistency check on the obtained weight vector to ensure that the weight logic is reasonable across regions and criteria, and avoid weight contradictions or inconsistencies. After the weights are verified for consistency, the comprehensive priority index of each aerial robot is output, which is used to generate the final list of candidate aerial robots.

[0032] It needs to be explained that linear mapping is a simple mathematical method used to linearly transform numbers in one range to another while maintaining the proportional relationship between the numbers; relative weight is to convert the ranking results into calculable continuous values ​​through linear mapping, which are used to represent the importance or priority difference of each candidate region relative to other regions. Furthermore, the generated candidate aerial robot list is not simply a task allocation list, but a control priority list for the safe execution of robot commands. The robots ranked at the top of the list all meet the physical control conditions of sufficient energy, controllable motor temperature rise, and reasonable airspace occupation. Subsequently, execution commands can be directly issued to the robots based on this list, ensuring that when the commands are implemented, the robot's hardware status and airspace environment meet the control safety requirements, thus realizing a control closed loop from algorithm optimization results to physical control implementation.

[0033] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0034] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0035] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0036] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0037] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for dynamic task allocation and optimization of aerial robots, characterized in that: Includes the following steps: Step S1: Obtain the current task list corresponding to each aerial robot through the task database, retrieve the task level of the task target in the current task list, collect the current remaining battery power of the aerial robot to analyze the endurance time, and filter and mark the task targets in the current task list in combination with the task level. Step S2: Import the location coordinates of the marked task target and the available time of the task into the clustering analysis algorithm to generate candidate regions. Mark the overlapping regions in the candidate regions corresponding to different aerial robots. Step S3: Obtain a list of aerial robots with marked candidate regions, detect the motor temperature rise gradient and airspace occupancy of each aerial robot in the list, analyze the trajectory density characteristics using the airspace occupancy, and evaluate the execution optimization score of the aerial robot in combination with the motor temperature rise gradient. Step S4: Sort the aerial robots in each marked candidate region using the performance optimization score, and import the sorting results of each marked candidate region into the hierarchical analysis algorithm to generate a candidate aerial robot list.

2. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S1, the identifier ID of each aerial robot is matched with the task database to obtain the current task target corresponding to each aerial robot, and then integrated into the current task list of the aerial robots. Match the task ID of each task target in the current task list of the aerial robot with the task database, and calculate the hierarchical depth of the task target as the hierarchical level of the task target; If the task level of the task objective is lower than the preset task level threshold, the task objective is determined to be of high priority. If the task level of the task objective is greater than or equal to the preset task level threshold, the task objective is determined to be of low priority.

3. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S1, the remaining power of each aerial robot is read by calling the BMS interface of each aerial robot; The historical power consumption of each aerial robot was retrieved, and the average historical power consumption of the same model of robot performing similar tasks was taken as the average power consumption of each aerial robot. The remaining power of each aerial robot is obtained by multiplying the remaining power of each aerial robot by the nominal battery capacity. The remaining power of each aerial robot is then divided by the average power consumption to obtain the flight time of each aerial robot. Iterate through the current task list corresponding to each aerial robot and read the estimated execution time of each task objective; If the priority of the task objective is high and the expected execution time of the task objective is less than the endurance of the corresponding aerial robot, then the task objective is marked. Conversely, the task target is not marked.

4. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S2, each marked task target is matched with the task database to obtain the location coordinates and available time of each marked task target; The location coordinates of each marked task target and the available time of the task are standardized and integrated into a spatiotemporal feature vector set; The standardized spatiotemporal feature vector set is input into the clustering algorithm model, and judgment is made based on pre-defined logic. Assign a region number to each marked task target that conforms to the predefined logic; Marked task targets with the same region number are grouped together to form a candidate region.

5. The method for dynamic task allocation and optimization of aerial robots according to claim 4, characterized in that: In step S2, if a candidate region has some location coordinates that are the same as another candidate region, and the available time of the candidate region has overlapping time periods, then it is determined that the space where the candidate region has the same location coordinates as another candidate region is marked to obtain the marked candidate region. Conversely, if the condition is not met, then no marking is performed.

6. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S3, the aerial robots corresponding to each candidate marking region are integrated into an aerial robot list for each candidate marking region; Match the marked task target of each aerial robot in the aerial robot list of each marked candidate area with the historical task database to obtain the maximum motor temperature when each marked task target is executed; The current temperature of each aviation robot's motor is obtained by using the temperature sensor built into each aviation robot's motor. The ratio of the maximum motor temperature when the marked task target is executed to the current motor temperature of the corresponding aerial robot, and then to the corresponding expected execution time, is used as the motor temperature rise gradient for each aerial robot in the aerial robot list.

7. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S3, the maximum x-axis coordinate, maximum y-axis coordinate, and maximum z-axis coordinate of each candidate marker region are subtracted from the minimum x-axis coordinate, minimum y-axis coordinate, and minimum z-axis coordinate, respectively, to obtain the length, width, and height of each candidate marker region. The spatial size of each candidate region is obtained by calculating its length, width, and height using the ellipsoid approximation algorithm. The ratio of the spatial size of each candidate marking region to the spatial size required by the marking task target of each aerial robot in the aerial robot list corresponding to the candidate marking region is used as the airspace occupancy of each aerial robot in the aerial robot list.

8. The method for dynamic task allocation and optimization of aerial robots according to claim 7, characterized in that: In step S3, the average airspace occupancy of all aerial robots within a candidate region is calculated; The ratio of the airspace occupancy of each aerial robot corresponding to the candidate region to the average airspace occupancy of the candidate region is used as the track density feature of each aerial robot in the aerial robot list of the candidate region. The temperature rise gradient and track density characteristics of the motors of each aerial robot in each list of aerial robots are standardized to obtain the temperature rise gradient factor and track density factor. The optimal performance score for each aerial robot in each candidate marked region is calculated by combining the temperature gradient factor and the track density factor. ,in, The temperature gradient factor. For track density factor, The best performance score is given to each aerial robot in each marked candidate area.

9. The method for dynamic task allocation and optimization of aerial robots according to claim 1, characterized in that: In step S4, the aerial robots in the same candidate region are sorted in descending order based on their performance preference scores. The sorting results formed by each marked candidate region are input into the hierarchical analysis algorithm model. The aerial robot is used as the scheme layer, the optimization scoring is used as the criterion layer, and the candidate aerial robot list generation is used as the target layer. The sorting results of each marked candidate region are converted into relative weights by linear mapping and normalized to obtain the comprehensive weight of each aerial robot in the overall candidate list. After the overall weights are verified for consistency, the overall priority index of each aerial robot is output, which is used to generate the final list of candidate aerial robots.

Images