Unmanned aerial vehicle hangar scheduling method based on heuristic algorithm

Abstract

The invention discloses an unmanned aerial vehicle hangar scheduling method based on a heuristic algorithm, and relates to the technical field of unmanned aerial vehicles. Screening a machine library capable of being scheduled, and constructing an alternative task set based on the machine library; deleting the task data with the route data; performing initialization grouping on the alternative task set; in the same group of data, calculating the sum dist of distances between each task data and other task data, and the average value Dist of the dist; for one alternative task set, respectively selecting one task data of the two groups of data; calculating the Dist sum of the two selected groups of data before and after task data exchange, namely Dist1 and Dist2; only when Dist1gt; when the Dest2 is detected, exchange is executed; when a preset target is reached, switching the scheduling of the alternative task set; and repeating the above steps to complete task allocation of all the machine libraries capable of being scheduled. Finally, the scheduling efficiency and task execution efficiency of the unmanned aerial vehicle can be effectively improved, and cost consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and particularly to a scheduling method for UAV hangars based on a heuristic algorithm. Background Art

[0002] With the vigorous development of the low-altitude economy in China, UAV technology has been widely applied in various industries due to its core advantages of low cost, high efficiency, and strong adaptability. The capabilities of UAVs such as real-time data collection, remote control, and autonomous navigation enable them to complete complex tasks in various environments. With the increase in the number of UAV tasks, many high-tech enterprises providing UAV services have built hangars at key nodes in the city, enabling UAVs to automatically execute tasks according to pre-deployed plans. However, how to efficiently and intelligently allocate tasks and execute plans for UAVs in each hangar has become a key issue in improving the task execution efficiency of UAVs and reducing the execution cost. In response to this problem, intelligent scheduling methods provide an effective solution idea, and at the same time can enable enterprises to reduce costs and increase efficiency, and reduce the work difficulty of scheduling personnel.

[0003] In the prior art, the problem of UAV task scheduling executed by hangars mainly involves allocating tasks for UAVs in multiple hangars according to time periods, while satisfying various constraint conditions and optimizing specific objectives. Its main challenges include:

[0004] (1) Task cost: The cost of UAVs executing tasks is proportional to the total flight distance. The tasks in each time period of scheduling need to satisfy the proximity of distances, which requires high professional requirements for relevant personnel;

[0005] (2) High dynamicity: UAV tasks are usually in a dynamically changing environment, and task requirements and UAV states may change in real time. The scheduling plan needs to have the ability to quickly adjust;

[0006] (3) Timeliness: When the number of UAV tasks is very large, traditional methods based on human experience can no longer make quick responses, and the program needs to quickly give strategies. Summary of the Invention

[0007] In view of the problems existing in the prior art, the present invention provides a scheduling method for UAV hangars based on a heuristic algorithm. On the basis of fully considering costs and multiple constraint conditions, through comprehensive analysis of airspace, UAV tasks, hangar information, task information, etc., combined with the heuristic algorithm, automatic intelligent scheduling is performed on all hangars, which can effectively improve the UAV scheduling efficiency, improve the task execution efficiency of UAVs, and reduce the cost consumption of UAVs executing tasks.

[0008] The technical solution of the present invention is implemented as follows:

[0009] A method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm, comprising the following steps:

[0010] T1. Collect data, including airspace data, hangar data, and mission data; the airspace data includes legal airspace ranges; the hangar data includes hangar coordinates and mission executable radii; the mission data includes mission coordinates; wherein, some or all of the mission data includes route data;

[0011] T2. Screen the hangars whose hangar coordinates are within the airspace range as schedulable hangars; the range within the mission executable radius centered on the hangar coordinates of any one of the schedulable hangars is the coverage range; several pieces of the mission data whose mission coordinates are within the coverage range constitute an alternative mission set; one schedulable hangar corresponds to one alternative mission set;

[0012] T3. For an alternative mission set, delete the mission data with the route data; the missions with existing route data are regarded as having been determined to be arranged; the remaining mission data does not include route data;

[0013] T4. Initialize the grouping of an alternative mission set, divide several pieces of the mission data into several groups of data; for the data in the same group, calculate the distance between any one piece of the mission data and other pieces of the mission data, and the sum of the distances is denoted as dist; the average value of dist of the data in the same group is denoted as Dist;

[0014] T5. For an alternative mission set, select two groups of data based on Dist; for the two selected groups of data, select one piece of the mission data from one group of data based on dist;

[0015] Calculate the sum of Dist of the two selected groups of data, denoted as Dist1; for the two selected groups of data, exchange the positions of the two selected pieces of mission data and calculate the sum of Dist of the two groups of data, denoted as Dist2; only when Dist1>Dist2, perform the exchange of the mission data; otherwise, do not perform the exchange;

[0016] T6. Repeat T4 and T5 until a preset target is reached, and complete the task scheduling assignment for one schedulable hangar;

[0017] T7. Repeat T3 - T6 to complete the task assignment for all schedulable hangars.

[0018] The legal airspace ranges mainly include flyable airspace and approved controlled airspace.

[0019] An air route is the specific path that an aircraft flies in the air, usually formed by connecting a series of navigation points (such as VOR, NDB, GPS points).

[0020] By continuously optimizing the Dist value of each group of data through dist and Dist, that is, the sum of distances, the task coordinates of the same group of task data can be made relatively more aggregated and closer, enabling the drone to execute as many tasks as possible in a shorter time, greatly reducing the time cost of task execution and the energy consumption cost of the drone.

[0021] As a further optimization of the above solution, the preset goal is: when executing step T6, the number of consecutive non-execution of exchanges reaches a preset number.

[0022] As a further optimization of the above solution, in step T5, for one of the alternative task sets, a first probability θ1 is assigned to each group of data according to Dist, that is:

[0023] where i represents the i-th group of data;

[0024] Two groups of data are selected from one of the alternative task sets according to the first probability.

[0025] As a further optimization of the above solution, in step T5, for a group of data, a second probability θ2 is assigned to each task data according to dist, that is:

[0026] where j represents the j-th task data;

[0027] One task data is selected from a group of data according to the second probability.

[0028] As a further optimization of the above solution, the initialization grouping is to evenly divide the area according to the abscissa or ordinate of the coverage range, or divide the grid based on the coordinates to obtain the evenly divided area; finally, a preset number of evenly divided areas are obtained; the task data located within the evenly divided area constitutes the group of data.

[0029] As a further optimization of the above solution, the initialization grouping is to evenly divide the task data according to the abscissa or ordinate of the task coordinates, or perform grid equal division through the task coordinates; finally, a corresponding preset number of groups of data are obtained.

[0030] As a further optimization of the above solution, multiple time periods are set for one hangar; one group of data corresponds to one time period; within one time period, the drone task is executed according to the task data corresponding to one group of data.

[0031] As a further optimization of the above solution, a hangar is also provided with a task execution times threshold; the task execution times threshold corresponds to the maximum number of drone tasks that the hangar can execute within a time period.

[0032] In the initialization grouping, the task data is evenly grouped according to the task execution times threshold.

[0033] As a further optimization of the above solution, in step T2, in an alternative task set, for task data with route data, it is classified according to the route data, and task data on the same route executes drone tasks within the same time period.

[0034] As a further optimization of the above solution, in step T1, the collected data is judged for legality to obtain legal data and illegal data; the legal data is preprocessed, and the illegal data is given a warning; the data preprocessing includes deleting invalid data and redundant data; the illegal data includes data with values exceeding a preset range and data with data types not meeting requirements.

[0035] Some terms in the data that are irrelevant and useless for scheduling are deleted to relieve memory pressure. Or for example, there may be multiple name tags in a piece of data, and only one unique name tag needs to be saved.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] (1) The present invention provides a method for scheduling drone hangars based on a heuristic algorithm. Through the heuristic algorithm, each hangar is automatically and intelligently scheduled. On the basis of meeting airspace conditions and reducing task execution costs, a scheduling plan can be quickly and automatically given, effectively improving the scheduling efficiency of drone tasks in the hangar.

[0038] (2) By continuously optimizing the Dist value of each group of data through dist and Dist, that is, the sum of distances, the task coordinates of the same group of task data can be made relatively more aggregated and closer, enabling the drone to execute as many tasks as possible in a short time, greatly reducing the time cost of task execution and the energy consumption cost of the drone, and effectively improving the execution efficiency of the drone for tasks.

[0039] (3) The present invention can avoid the errors generated by traditional scheduling strategies and does not require the executor to have rich prior knowledge of drone task scheduling rules and strategies, greatly reducing the labor cost and experience cost of scheduling. Description of the Drawings

[0040] Figure 1It is a schematic flowchart of a method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm provided by an embodiment of the present invention;

[0041] Figure 2 It is a schematic diagram of the hangar coordinates of a schedulable hangar and the task coordinates of each task data within the coverage range provided by an embodiment of the present invention;

[0042] Figure 3 It is Figure 2 A schematic diagram obtained after initial grouping;

[0043] Figure 4 It is a data table of the first group of data provided by an embodiment of the present invention;

[0044] Figure 5 It is a data table of the second group of data provided by an embodiment of the present invention;

[0045] Figure 6 It is a data table of the seventh group of data provided by an embodiment of the present invention;

[0046] Figure 7 It is Figure 3 A schematic diagram obtained after scheduling. Detailed implementation manners

[0047] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0048] As Figures 1 to 7 shown, this embodiment provides a method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm, including the following steps:

[0049] T1. Collect data, including airspace data, hangar data, and task data; the airspace data includes the legal airspace range; the hangar data includes the hangar coordinates and the task executable radius; the task data includes the task coordinates; among them, part of the task data includes route data;

[0050] In this embodiment, the collected data is judged for legality to obtain legal data and illegal data; the legal data is preprocessed, and the illegal data is given a warning; the data preprocessing includes deleting invalid data and redundant data; the illegal data includes data with values exceeding the preset range and data with data types not meeting the requirements. Some terms in the data that are irrelevant and useless for scheduling are deleted to relieve the memory pressure. For example, there may be multiple name tags in a piece of data, and only one unique name tag needs to be saved.

[0051] T2. Screen the hangars whose coordinates are within the airspace as schedulable hangars; the range within the task executable radius centered on the hangar coordinates of any one schedulable hangar is the coverage area; several task data with task coordinates within the coverage area form an alternative task set; one schedulable hangar corresponds to one alternative task set; in this embodiment, multiple time periods are set for one hangar; one hangar is also set with a task execution times threshold; the task execution times threshold corresponds to the maximum number of drone tasks that the hangar can execute within one time period. In this embodiment, one hangar can execute 10 tasks within one time period.

[0052] In this embodiment, in step T2, in an alternative task set, for the task data with route data, classification is performed according to the route data, and the task data on the same route executes drone tasks within the same time period.

[0053] T3. For an alternative task set, delete the task data with route data; the task that already has route data is regarded as having been determined to be arranged; the remaining task data does not include route data; in this embodiment, the obtained alternative task set includes 69 pieces of task data.

[0054] T4. Initialize the grouping for an alternative task set. In this embodiment, the initialization grouping is to evenly group the task data according to the ordinate of the task coordinates, and finally obtain several groups of data corresponding to the preset quantity. The maximum quantity of task data in each group is the task execution times threshold. One group of data corresponds to one time period; within one time period, drone tasks are executed according to the task data corresponding to one group of data. In this example, the 69 tasks in the alternative task set are divided into 7 groups in the form of 10 tasks per group based on the dimension, and are identified with different colors and shapes (as Figure 3 shown, from top to bottom, including "light-colored triangle", "light-colored square", "light-colored circle", "dark-colored triangle", "dark-colored square", "dark-colored circle" and "black triangle".); among them, the last 9 pieces of task data form one group of data.

[0055] For the same set of data, calculate the distance between any one task data and other task data, and record the sum of the distances as dist; the average value of dist for the same set of data is recorded as Dist; in this example, the dist values of the 10 tasks in the first group are: 19, 16, 17, 22, 21, 16, 15, 18, 22, 29, and Dist is 19.8.

[0056] T5. For an alternative task set, select two sets of data based on Dist; specifically, assign the first probability θ1 to each set of data according to Dist, that is:

[0057] where i represents the i-th set of data and n is the number of data sets in the current alternative task set;

[0058] In this embodiment, the Dist values of the 7 sets of data are [19.8, 24.4, 18.8, 21.3, 20.2, 17.6, 16.4], and the converted θ1 values are respectively

[0059] [0.143, 0.176, 0.136, 0.154, 0.146, 0.127, 0.118].

[0060] Select two sets of data from an alternative task set according to the first probability. Suppose the 1st and 2nd sets are selected.

[0061] For the two selected sets of data, select one task data from each set based on dist; specifically, assign the second probability θ2 to each task data according to dist, that is:

[0062] where j represents the j-th task data and m is the number of task data in the current set of data; taking the 1st set as an example, dist is [19, 16, 17, 22, 21, 16, 15, 18, 22, 29], and the converted θ2 values are respectively

[0063] [0.10, 0.08, 0.09, 0.11, 0.11, 0.08, 0.08, 0.09, 0.11, 0.15].

[0064] Select one task data from a set of data according to the second probability. Suppose task 10 in the 1st set and task 1 in the 2nd set are selected.

[0065] Calculate the sum of Dist of the two selected sets of data, and record it as Dist1; for the two selected sets of data, swap the positions of the two selected task data and calculate the sum of Dist of the two sets of data, and record it as Dist2; only when Dist1 > Dist2, perform the swap of the task data; otherwise, do not perform the swap.

[0066] In a specific implementation, the changed part of each set of data is only the distance of the swapped task relative to other tasks. Therefore, when comparing Dist1 and Dist2, only the change value of the distance needs to be compared.

[0067] Before the task data is swapped, the sum of the distances dist of task 10 in the first group relative to other tasks is 29, and the sum of the distances dist of task 1 in the second group relative to other tasks is 24.

[0068] Suppose after two tasks are swapped, the distances of task 1 in the second group relative to tasks 1 - 9 in the first group become [2, 2, 3, 4, 1, 1, 2, 3, 4], then the dist of "task 10" in the first group is 22.

[0069] Similarly, the distances of task 10 in the first group relative to tasks 2 - 10 in the second group become [2, 3, 4, 5, 1, 1, 3, 6, 7], then the dist of "task 1" in the second group is 32.

[0070] Therefore, after the task swap, the comparison of Dist1 and Dist2 is transformed into the comparison of (29 + 24 = 53) and (22 + 32 = 54). Thus, it is obtained that Dist1 < Dist2, and this swap is not executed.

[0071] T6. Repeat T4 and T5 until a preset goal is reached to complete the task scheduling and allocation of a schedulable hangar; in this embodiment, the preset goal is: for an alternative task set, the number of consecutive non - executed swaps reaches a preset number.

[0072] T7. Repeat T3 - T6 to complete the task allocation of all schedulable hangars.

[0073] The legal airspace range mainly includes flyable airspace and approved controlled airspace.

[0074] A flight route is the specific path of an aircraft flying in the air, usually connected by a series of navigation points (such as VOR, NDB, GPS points).

[0075] By continuously optimizing the Dist value, that is, the sum of distances, of each set of data through dist and Dist, the task coordinates of the same set of task data can be made relatively more aggregated and closer, enabling the UAV to execute as many tasks as possible in a shorter time, greatly reducing the time cost of task execution and the energy consumption cost of the UAV.

[0076] Based on the disclosure and teachings of the above specification, those skilled in the art to which the present invention pertains can also make changes and modifications to the above embodiments. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and some modifications and changes to the present invention should also fall within the protection scope of the claims of the present invention. In addition, although some specific terms are used in this specification, these terms are only for convenience of description and do not constitute any limitation to the present invention.

Claims

1. A scheduling method for an unmanned aerial vehicle hangar based on a heuristic algorithm, characterized in that, Including the following steps: T1. Collect data, including airspace data, hangar data, and mission data; the airspace data includes legal airspace ranges; the hangar data includes hangar coordinates and mission execution radii; the mission data includes mission coordinates; wherein, some or all of the mission data includes route data; T2. Screen the hangars whose hangar coordinates are within the airspace range as schedulable hangars; the range within the mission execution radius centered on the hangar coordinates of any one of the schedulable hangars is the coverage range; several pieces of the mission data whose mission coordinates are within the coverage range form an alternative mission set; one schedulable hangar corresponds to one alternative mission set; T3. For an alternative mission set, delete the mission data with the route data; T4. Initialize the grouping for an alternative mission set, divide several pieces of the mission data into several groups of data; for the same group of data, calculate the distance between any one piece of the mission data and other pieces of the mission data, and the sum of the distances is denoted as dist; the average value of dist for the same group of data is denoted as Dist; T5. For an alternative mission set, select two groups of data based on Dist; for the two selected groups of data, select one piece of the mission data from one group of data respectively based on dist; Calculate the sum of Dist of the two selected groups of data, denoted as Dist1; for the two selected groups of data, exchange the positions of the two selected pieces of mission data and calculate the sum of Dist of the two groups of data, denoted as Dist2; only when Dist1>Dist2, perform the exchange of the mission data; otherwise, do not perform the exchange; T6. Repeat T4 and T5 until a preset goal is reached to complete the mission scheduling assignment for one schedulable hangar; T7. Repeat T3 - T6 to complete the mission assignment for all schedulable hangars.

2. The method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, wherein The preset goal is: when performing step T6, the number of consecutive times without performing the exchange reaches a preset number of times.

3. A method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, characterized in that, In step T5, for one of the alternative task sets, a first probability θ1 is assigned to each group of data according to Dist, that is: Wherein, i represents the i-th group of data; Select two groups of data from an alternative mission set according to the first probability.

4. A method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, characterized in that In step T5, for a set of data, the second probability θ2 is assigned to each task data according to dist, that is: Wherein, j represents the j-th piece of mission data; Select one piece of mission data from a group of data according to the second probability.

5. A method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, characterized in that, The initialization grouping is to evenly divide the area according to the abscissa or ordinate of the coverage range, or divide the grid based on the coordinates to obtain the evenly divided area; finally, obtain a preset number of evenly divided areas; the mission data located within the evenly divided area forms the group of data.

6. The method for scheduling the drone hangar based on the heuristic algorithm according to claim 1, wherein, The initialization grouping is to evenly group the mission data according to the abscissa or ordinate of the mission coordinates, or perform grid equal division through the mission coordinates; finally, obtain several groups of data corresponding to a preset number.

7. A scheduling method for an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, characterized in that Multiple time periods are set for one hangar; one group of data corresponds to one time period; within one time period, perform the UAV mission according to the mission data corresponding to one group of data.

8. A method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 7, characterized in that, A hangar is also provided with a task execution times threshold; the task execution times threshold corresponds to the maximum number of drone tasks that the hangar can execute within a said time period. In the said initialization grouping, the task data is evenly grouped according to the task execution times threshold.

9. The method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 7, wherein In step T2, in a said alternative task set, for task data with route data, classification is performed according to the route data, and task data located on the same route executes drone tasks within the same time period.

10. The method for scheduling an unmanned aerial vehicle hangar based on a heuristic algorithm according to claim 1, characterized in that, In step T1, the collected data is judged for legality to obtain legal data and illegal data; the legal data is subjected to data preprocessing, and the illegal data is subjected to warning processing; the data preprocessing includes deleting invalid data and redundant data; the illegal data includes data with values exceeding a preset range and data with data types not meeting the requirements.

Images