Inspection unmanned aerial vehicle scheduling method and device, electronic equipment, storage medium and program product

Abstract

The invention discloses an inspection unmanned aerial vehicle scheduling method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of rail transit. The method comprises the following steps: dividing each candidate inspection task to obtain a candidate static inspection task and a candidate dynamic inspection task; for each candidate static inspection task, obtaining a candidate static inspection route of each candidate static inspection task; based on a task scheduling model, screening each target static inspection task from each candidate static inspection task, and screening a target unmanned aerial vehicle executing each target static inspection task from each candidate unmanned aerial vehicle; and for each candidate dynamic inspection task, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, determining the candidate dynamic inspection task as a target dynamic inspection task, and pausing the current dynamic inspection task. According to the technical scheme of the embodiment of the invention, the inspection efficiency and the inspection safety of the unmanned aerial vehicle inspection of the railway marshalling station are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit, and particularly relates to a patrol unmanned aerial vehicle scheduling method, device, electronic device, storage medium and program product. Background Art

[0002] At present, as the core component of a railway hub, a railway marshalling yard undertakes important tasks such as train disintegration and marshalling. Due to the wide patrol area of the marshalling yard, the intricate layout of the tracks in the yard, and the large number of equipment and facilities in the yard, the traditional manual patrol method has problems such as low patrol efficiency and high safety risks for roof patrol. Summary of the Invention

[0003] The present invention provides a patrol unmanned aerial vehicle scheduling method, device, electronic device, storage medium and program product, which improves the patrol efficiency and patrol safety of the unmanned aerial vehicle patrol in the railway marshalling yard.

[0004] According to one aspect of the present invention, there is provided a patrol unmanned aerial vehicle scheduling method, the method comprising:

[0005] Obtaining the scheduling time, route attributes, patrol targets, patrol plans, task locations, task priorities and task resource requirements of each candidate patrol task in the railway marshalling yard, as well as the unmanned aerial vehicle locations, unmanned aerial vehicle flight speeds, unmanned aerial vehicle endurance times and unmanned aerial vehicle resource totals of each candidate unmanned aerial vehicle;

[0006] Dividing each of the candidate patrol tasks according to the scheduling time and route attributes of each of the candidate patrol tasks to obtain candidate static patrol tasks and candidate dynamic patrol tasks;

[0007] For each of the candidate static patrol tasks, obtaining the candidate static patrol routes of each of the candidate static patrol tasks according to the patrol targets and patrol plans of each of the candidate static patrol tasks;

[0008] For each of the candidate static patrol tasks, based on a task scheduling model, according to the task locations, task priorities and task resource requirements of each of the candidate static patrol tasks, and the unmanned aerial vehicle locations, unmanned aerial vehicle flight speeds, unmanned aerial vehicle endurance times and unmanned aerial vehicle resource totals of each of the candidate unmanned aerial vehicles, screening each target static patrol task from each of the candidate static patrol tasks, and screening the target unmanned aerial vehicles for executing each of the target static patrol tasks from each of the candidate unmanned aerial vehicles, so that each of the target unmanned aerial vehicles executes the target static patrol tasks along the target static patrol routes corresponding to the target static patrol tasks;

[0009] For each of the candidate dynamic inspection tasks, obtain the current unmanned aerial vehicles (UAVs) for executing the current dynamic inspection tasks, the task priorities of the current dynamic inspection tasks, and the candidate dynamic inspection routes of the candidate dynamic inspection tasks, and compare the task priorities of the candidate dynamic inspection tasks with those of the current dynamic inspection tasks;

[0010] For each of the candidate dynamic inspection tasks, when the task priority of the candidate dynamic inspection task is higher than that of the current dynamic inspection task, determine the candidate dynamic inspection task as the target dynamic inspection task, and pause the current dynamic inspection task, so that each of the current UAVs executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

[0011] According to another aspect of the present invention, there is provided an inspection UAV scheduling device, the device includes:

[0012] An inspection task acquisition module, configured to acquire the scheduling time, route attributes, inspection targets, inspection plans, task locations, task priorities, and task resource requirements of each candidate inspection task of a railway marshalling yard, as well as the UAV locations, UAV flight speeds, UAV endurance times, and total UAV resources of each candidate UAV;

[0013] An inspection task division module, configured to divide each of the candidate inspection tasks according to the scheduling time and route attributes of each of the candidate inspection tasks, to obtain candidate static inspection tasks and candidate dynamic inspection tasks;

[0014] A static inspection task route planning module, configured to, for each of the candidate static inspection tasks, obtain the candidate static inspection routes of each of the candidate static inspection tasks according to the inspection targets and inspection plans of each of the candidate static inspection tasks;

[0015] A static inspection task scheduling module, configured to, for each of the candidate static inspection tasks, based on a task scheduling model, screen each target static inspection task from the candidate static inspection tasks, and screen target UAVs for executing each of the target static inspection tasks from the candidate UAVs, so that each of the target UAVs executes the target static inspection task along the target static inspection route corresponding to the target static inspection task;

[0016] A dynamic inspection task priority comparison module, configured to obtain, for each of the candidate dynamic inspection tasks, the current unmanned aerial vehicle (UAV) for executing each current dynamic inspection task, the task priority of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and compare the task priority of each candidate dynamic inspection task with the task priority of each current dynamic inspection task;

[0017] A dynamic inspection task scheduling module, configured to, for each of the candidate dynamic inspection tasks, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, determine the candidate dynamic inspection task as a target dynamic inspection task, and pause the current dynamic inspection task, so that each current UAV executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

[0018] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0019] At least one processor; and

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the inspection UAV scheduling method according to any embodiment of the present invention.

[0022] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions, and when the computer instructions are executed by a processor, the inspection UAV scheduling method according to any embodiment of the present invention is implemented.

[0023] According to another aspect of the present invention, there is provided a computer program product including a computer program, and when the computer program is executed by a processor, the inspection UAV scheduling method according to any embodiment of the present invention is implemented.

[0024] The technical solution of the embodiment of the present invention divides each candidate inspection task into a candidate static inspection task and a candidate dynamic inspection task through the scheduling opportunity and route attributes of each candidate inspection task, improving the flexibility and accuracy of task scheduling in a railway marshalling station. For each candidate static inspection task, according to the inspection target and inspection plan of each candidate static inspection task, a candidate static inspection route for each candidate static inspection task is obtained. Based on the task scheduling model, according to the task location, task priority, and task resource demand of each candidate static inspection task, as well as the drone location, drone flight speed, drone battery life, and total drone resources of each candidate drone, each target static inspection task is screened from the candidate static inspection tasks, and the target drone for executing each target static inspection task is screened from the candidate drones, so that each target drone executes the target static inspection task along the target static inspection route corresponding to the target static inspection task, realizing the task scheduling of the candidate static inspection tasks. Through the task scheduling model, the scheduling efficiency and accuracy of the static inspection tasks can be improved. For each candidate dynamic inspection task, by obtaining the current drone executing each current dynamic inspection task, the task priority of each current dynamic inspection task, and the candidate dynamic inspection route of each candidate dynamic inspection task, and comparing the task priority of each candidate dynamic inspection task with the task priority of each current dynamic inspection task, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, the candidate dynamic inspection task is determined as the target dynamic inspection task, and the current dynamic inspection task is paused, so that each current drone executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task, realizing the task scheduling of the candidate dynamic inspection tasks. Through the comparison of the task priority of the candidate dynamic inspection task and the task priority of the current dynamic inspection task, the resource preemption of the dynamic inspection task with a high task priority is realized, and during the execution of the dynamic inspection task, the task priority is ensured, which can improve the scheduling efficiency and accuracy of the dynamic inspection task, and thus improve the inspection efficiency and inspection safety of the drone inspection in the railway marshalling station.

[0025] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Brief Description of the Drawings

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0027] Figure 1 It is a flowchart of an inspection UAV scheduling method provided according to Embodiment 1 of the present invention;

[0028] Figure 2 It is a flowchart of an inspection UAV scheduling method provided according to Embodiment 2 of the present invention;

[0029] Figure 3 It is a classification diagram of candidate inspection tasks provided according to Embodiment 2 of the present invention;

[0030] Figure 4 It is a schematic diagram of a candidate static inspection route provided according to Embodiment 2 of the present invention;

[0031] Figure 5 It is a schematic diagram of a UAV task operation diagram provided according to Embodiment 2 of the present invention;

[0032] Figure 6 It is a schematic structural diagram of an inspection UAV scheduling device provided according to Embodiment 3 of the present invention;

[0033] Figure 7 It is a schematic structural diagram of an electronic device for implementing the inspection UAV scheduling method of the embodiments of the present invention. Detailed implementation manners

[0034] In order to enable those skilled in the art to better understand the solution of the present invention, 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 of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0036] Embodiment 1

[0037] Figure 1The figure is a flowchart of an inspection UAV scheduling method provided in the first embodiment of the present invention. The embodiments of the present invention are applicable to the situation of scheduling inspection UAVs for railway marshalling yards. This method can be executed by an inspection UAV scheduling device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device with inspection UAV scheduling functions.

[0038] See Figure 1 The inspection UAV scheduling method shown includes:

[0039] S110. Obtain the scheduling time, route attributes, inspection targets, inspection plans, task locations, task priorities, and task resource requirements of each candidate inspection task in the railway marshalling yard, as well as the UAV locations, UAV flight speeds, UAV endurance times, and total UAV resources of each candidate UAV.

[0040] Railway marshalling yards are characterized by a wide inspection area, intricate in-station track layouts, and numerous in-station equipment and facilities. Candidate inspection tasks can be unexecuted inspection tasks in railway marshalling yards. The scheduling time can be used to characterize the scheduling situation of candidate inspection tasks. Exemplarily, the scheduling time can include fixed scheduling time and non-fixed scheduling time. The route attributes of candidate inspection tasks can be used to characterize the route situations of candidate inspection tasks. Exemplarily, the route attributes of candidate inspection tasks can include fixed routes and non-fixed routes. The inspection targets of candidate inspection tasks can be used to characterize the inspection objects of candidate inspection tasks. Exemplarily, the inspection targets of candidate inspection tasks can include catenaries, trains, railway tracks, operating personnel, or goods, etc. The inspection plans of candidate inspection tasks can be used to characterize the execution situations of candidate inspection tasks. Exemplarily, the inspection plans of candidate inspection tasks can include the inspection times of candidate inspection tasks, each inspection point, the inspection sequence of each inspection point, and the inspection requirements of each inspection point, etc. The task locations of candidate inspection tasks can be the locations of each inspection point of candidate inspection tasks. The task priorities of candidate inspection tasks can be used to characterize the urgency or importance of candidate inspection tasks. The task resource requirements of candidate inspection tasks can be used to characterize the amount of resources that need to be consumed by candidate UAVs to execute candidate inspection tasks. Candidate UAVs can be used as optional UAVs for executing candidate inspection tasks. The UAV locations of candidate UAVs can be the locations where candidate UAVs are located. The UAV flight speeds of candidate UAVs can be used to characterize the speeds of candidate UAVs for executing candidate inspection tasks. The UAV endurance times of candidate UAVs can be used to characterize the task-executable times of candidate UAVs. The total UAV resources of candidate UAVs can be used to characterize the remaining resource situations of candidate UAVs.

[0041] Specifically, each candidate inspection task and each candidate UAV of the railway marshalling yard can be obtained. Each candidate inspection task can be detected to determine the scheduling time, route attributes, inspection targets, inspection plans, task locations, task priorities, and task resource requirements of each candidate inspection task in the railway marshalling yard. Each candidate UAV can be detected to determine the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV.

[0042] S120. According to the scheduling time and route attributes of each candidate inspection task, each candidate inspection task is divided to obtain candidate static inspection tasks and candidate dynamic inspection tasks.

[0043] Candidate static inspection tasks can be candidate inspection tasks with fixed scheduling times and fixed routes. Exemplarily, candidate static inspection tasks can include daily inspection tasks, on-track inspection tasks for personnel, train status inspection tasks, fixed-point inspection tasks, and hump shunting operation inspection tasks, etc. Daily inspection tasks can be used for daily inspections of the railway marshalling yard, such as daily inspections of the catenary or railway tracks in the railway marshalling yard. On-track inspection tasks for personnel can be used to inspect the on-track operation conditions of the operating personnel in the railway marshalling yard. Among them, the operating personnel in the railway marshalling yard can be the operation and maintenance personnel in the railway marshalling yard. Train status inspection tasks can be used to inspect the train status. For example, train status inspection tasks can include train skylight identification tasks or train integrity identification tasks, etc. Fixed-point inspection tasks can be used to inspect fixed positions in the railway marshalling yard. Hump shunting operation inspection tasks can be used to inspect hump shunting operations.

[0044] Candidate dynamic inspection tasks can be candidate inspection tasks with non-fixed scheduling times and / or non-fixed routes. Exemplarily, candidate static inspection tasks can include special cargo escort inspection tasks, emergency inspection tasks, and abnormal inspection tasks, etc. Special cargo escort inspection tasks can be used to escort special cargo. Emergency inspection tasks can be inspection tasks in case of emergencies, such as safety critical equipment inspection tasks. Abnormal inspection tasks can be used to inspect abnormal situations in the railway marshalling yard, such as emergency fault detection tasks. Compared with candidate static inspection tasks, candidate dynamic inspection tasks have a certain degree of temporariness and randomness.

[0045] Specifically, the scheduling time and route attributes of each candidate inspection task can be detected. When the candidate inspection task has a fixed scheduling time and a fixed route, it can be determined that the candidate inspection task is a candidate static inspection task; when the candidate inspection task has a non-fixed scheduling time and / or a non-fixed route, it can be determined that the candidate inspection task is a candidate dynamic inspection task.

[0046] S130. For each candidate static inspection task, according to the inspection objectives and inspection plans of each candidate static inspection task, obtain the candidate static inspection routes of each candidate static inspection task.

[0047] The candidate static inspection route can be the inspection route of the candidate static inspection task. For a candidate static inspection task, while determining the inspection objectives and inspection plans of the candidate inspection task, the candidate static inspection route of the candidate static inspection task can be directly determined based on the inspection objectives and inspection plans of the candidate inspection task.

[0048] Specifically, for each candidate static inspection task, for each inspection point in the inspection objectives and inspection plans of each candidate static inspection task, generate the candidate static routes of each candidate static inspection task.

[0049] Optionally, the no-fly zone range of the railway marshalling yard can be obtained, and according to the overlap between the candidate static inspection routes of each candidate static inspection task and the no-fly zone range, adjust the candidate static inspection routes of each candidate static inspection task so that the candidate static inspection routes of the candidate static inspection tasks avoid the no-fly zone range of the railway marshalling yard.

[0050] Among them, the no-fly zone range can be the area above the railway marshalling yard where unmanned aerial vehicles are prohibited from flying. The overlap between the candidate static inspection route of the candidate static inspection task and the no-fly zone range can be used to characterize the situation where the candidate static inspection route falls into the no-fly zone range.

[0051] Specifically, the pre-determined no-fly zone range of the railway marshalling yard can be obtained. The overlap between the candidate static inspection routes of each candidate static inspection task and the no-fly zone range can be detected. Adjust the candidate static inspection routes of the overlapping parts of the candidate static inspection routes of each candidate static inspection task and the no-fly zone range so that the candidate static inspection routes of the candidate static inspection tasks avoid the no-fly zone range of the railway marshalling yard.

[0052] This solution introduces the no-fly zone range of the railway marshalling yard, and adjusts the candidate static inspection routes of the candidate static inspection tasks through the no-fly zone range of the railway marshalling yard, ensuring the availability of the candidate static inspection routes and improving the feasibility of the candidate static inspection task scheduling.

[0053] In an optional embodiment of the present invention, after obtaining the candidate static inspection routes of each candidate static inspection task according to the inspection objectives and inspection plans of each candidate static inspection task, it further includes: obtaining the restricted flight zone range of the railway marshalling yard, and determining the flight altitude of each candidate static inspection waypoint in each candidate static inspection route according to the overlap between the candidate static inspection routes of each candidate static inspection task and the restricted flight zone range.

[0054] The no-fly zone can be an area above a railway marshalling yard that restricts the flight altitude of drones. The overlapping situation between the candidate static inspection route of the candidate static inspection task and the no-fly zone can be used to characterize the situation where the candidate static inspection route falls within the no-fly zone. The candidate static inspection waypoint can be a point on the candidate static inspection route. The candidate static inspection route includes at least one candidate static inspection waypoint. The flight altitude of the candidate static inspection waypoint can be the maximum flight altitude of the drone at the candidate static inspection waypoint.

[0055] Specifically, the no-fly zone of the pre-determined railway marshalling yard can be obtained. The overlapping situation between the candidate static inspection route of each candidate static inspection task and the no-fly zone can be detected. The flight altitude of each candidate static inspection waypoint in the candidate static inspection route of the overlapping part between the candidate static inspection route of each candidate static inspection task and the no-fly zone can be adjusted.

[0056] This solution introduces the no-fly zone of the railway marshalling yard. Through the no-fly zone of the railway marshalling yard, the flight altitude of each candidate static inspection waypoint in the candidate static inspection route of the candidate static inspection task is adjusted, ensuring the usability of the candidate static inspection route and improving the feasibility of the candidate static inspection task scheduling.

[0057] S140. For each candidate static inspection task, based on the task scheduling model, according to the task location, task priority, and task resource requirements of each candidate static inspection task, as well as the drone location, drone flight speed, drone endurance time, and total drone resources of each candidate drone, screen out each target static inspection task among the candidate static inspection tasks, and screen out the target drones for executing each target static inspection task among the candidate drones, so that each target drone executes the target static inspection task along the target static inspection route corresponding to the target static inspection task.

[0058] The task scheduling model can be used to schedule the candidate static inspection tasks to determine the target static inspection tasks to be executed and the target drones for executing the target static inspection tasks. Optionally, the task scheduling model can be a machine learning model. The input data of the task scheduling model can be the task location, task priority, and task resource requirements of each candidate static inspection task, as well as the drone location, drone flight speed, drone endurance time, and total drone resources of each candidate drone. The output result of the task scheduling model can be the target static inspection tasks and the target drones for executing each target static inspection task. The target static inspection task can be a candidate static inspection task that needs to be executed. The target drone can be a candidate drone for executing the target static inspection task. The target static inspection route can be the candidate static inspection route corresponding to the target static inspection task.

[0059] Specifically, for each candidate static inspection task, the task location, task priority, and task resource requirement of each candidate static inspection task, as well as the drone location, drone flight speed, drone battery life, and total drone resources of each candidate drone, can be input into the task scheduling model to screen out each target static inspection task among the candidate static inspection tasks and screen out the target drones for executing each target static inspection task among the candidate drones, so that each target drone executes the target static inspection task along the target static inspection route corresponding to the target static inspection task.

[0060] In an alternative embodiment of the present invention, for each candidate static inspection task, based on the task scheduling model, according to the task location, task priority, and task resource requirement of each candidate static inspection task, as well as the drone location, drone flight speed, drone battery life, and total drone resources of each candidate drone, screening out each target static inspection task among the candidate static inspection tasks and screening out the target drones for executing each target static inspection task among the candidate drones includes: for each candidate static inspection task, obtaining the inspection accuracy of each candidate drone and determining the inspection accuracy required for each candidate static inspection task according to the inspection plan of each candidate static inspection task; for each candidate static inspection task, matching each candidate drone and each candidate static inspection task according to the inspection accuracy of each candidate drone and the inspection accuracy required for each candidate static inspection task, so that the inspection accuracy of the candidate drone meets the inspection accuracy required for the candidate static inspection task; for each matched candidate drone and candidate static inspection task, based on the task scheduling model, according to the task location, task priority, and task resource requirement of each candidate static inspection task, as well as the drone location, drone flight speed, drone battery life, and total drone resources of each candidate drone, screening out each target static inspection task among the candidate static inspection tasks and screening out the target drones for executing each target static inspection task among the candidate drones.

[0061] The inspection accuracy of the candidate drone can be the inspection accuracy of the inspection equipment of the candidate drone. Exemplarily, the inspection equipment of the candidate drone can be an image acquisition device or a lidar device, etc. The inspection accuracy required for the candidate static inspection task can be the inspection accuracy of the drone required for executing the candidate static inspection task. Matching each candidate drone and each candidate static inspection task enables a candidate drone with high accuracy to execute a candidate static inspection task with high accuracy, that is, enables the candidate drone to meet the inspection accuracy requirement of the candidate static inspection task.

[0062] Specifically, for each candidate static inspection task, the inspection accuracy of each candidate UAV can be detected. The inspection accuracy required for each candidate static inspection task can be extracted according to the inspection plan of each candidate static inspection task. For each candidate static inspection task, each candidate UAV is matched with each candidate static inspection task so that the inspection accuracy of the candidate UAV can meet the inspection accuracy required for the matched candidate static inspection task. For each matched candidate UAV and each candidate static inspection task, the task location, task priority, and task resource demand of each candidate static inspection task, as well as the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV, are input into the task scheduling model. Each target static inspection task is screened from each candidate static inspection task, and the target UAVs for executing each target static inspection task are screened from each candidate UAV, so that each target UAV executes the target static inspection task along the target static inspection route corresponding to the target static inspection task.

[0063] This solution introduces the inspection accuracy of each candidate UAV and the inspection accuracy required for each candidate static inspection task. By matching the inspection accuracy of each candidate UAV with the inspection accuracy required for each candidate static inspection task, it can be ensured that the inspection accuracy of the candidate UAV can meet the inspection accuracy required for the candidate static inspection task, thus ensuring the effectiveness of the completion of the candidate static inspection task.

[0064] In an alternative embodiment of the present invention, after screening each target static inspection task from each candidate static inspection task and screening the target UAVs for executing each target static inspection task from each candidate UAV, it further includes: generating a UAV task operation map according to the target static inspection route of each target static inspection task and the flight altitude of each target static inspection waypoint in each target static inspection route; based on the intersection waypoints in the UAV task operation map, adjusting the flight altitude of each target static inspection route at each intersection waypoint to avoid collisions between different target UAVs at the same intersection waypoint.

[0065] The UAV task operation map can be used to visually display the target static inspection route corresponding to the target UAV. Exemplarily, the abscissa of the UAV task operation map can be time, and the ordinate can be location. The intersection waypoints in the UAV task operation map can be the route intersection points of different target static inspection routes. Optionally, the corresponding flight altitude of each target static inspection route can be marked at the intersection waypoints in the UAV task operation map.

[0066] Specifically, after screening out each target static inspection task from each candidate static inspection task and screening out the target drones for executing each target static inspection task from each candidate drone, a drone task operation map is generated with time as the abscissa and the positions of each target static inspection waypoint in each target static inspection route as the ordinate according to the target static inspection routes of each target static inspection task and the flight altitudes of each target static inspection waypoint in each target static inspection route. The intersection waypoints in the drone task operation map are detected, the flight altitudes of different target static inspection routes at the same intersection waypoint are compared, and when the flight altitudes of different target static inspection routes at the same intersection waypoint are the same, the flight altitude of one of the target static inspection routes is adjusted to avoid collision between different target drones at the same intersection waypoint.

[0067] Through generating the drone task operation map, this solution can more intuitively display the target static inspection routes when the target drones execute the target static inspection tasks. By adjusting the flight altitudes of the target static inspection routes at each intersection waypoint, collision between different target drones at the same intersection waypoint can be avoided, improving the safety during the execution of the target static inspection tasks.

[0068] S150. For each candidate dynamic inspection task, obtain the current drones for executing each current dynamic inspection task, the task priorities of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and compare the task priorities of each candidate dynamic inspection task with the task priorities of each current dynamic inspection task.

[0069] The current dynamic inspection task can be a dynamic inspection task being executed at the current moment. The current drone can be the drone for executing the current dynamic inspection task. The task priority of the current dynamic inspection task can be used to represent the importance or urgency of the current dynamic inspection task.

[0070] Specifically, for each candidate dynamic inspection task, each current dynamic inspection task can be detected to determine the task priorities of each current dynamic inspection task, the candidate dynamic inspection routes of each candidate dynamic inspection task, and the current drones for executing each current dynamic inspection task. The task priorities of each candidate dynamic inspection task can be compared with the task priorities of each current dynamic inspection task.

[0071] S160. For each candidate dynamic inspection task, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, determine the candidate dynamic inspection task as the target dynamic inspection task, and pause the current dynamic inspection task so that each current drone executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

[0072] The target dynamic inspection task may be a candidate dynamic inspection task whose task priority is higher than that of the current dynamic inspection task. The target dynamic inspection route may be a candidate dynamic inspection route corresponding to the target dynamic inspection task.

[0073] Specifically, for each candidate dynamic inspection task, when it is detected that the task priority of the candidate dynamic inspection task is higher than that of the current dynamic inspection task, the candidate dynamic inspection task is determined as the target dynamic inspection task, and the current dynamic inspection task is paused, so that each current unmanned aerial vehicle executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task. Optionally, according to the task requirements of the target dynamic inspection task, the target dynamic inspection route of the target dynamic inspection task, as well as the flight speed and flight altitude of each target dynamic inspection waypoint in the target dynamic inspection route, can be determined. According to the target dynamic inspection route of the target dynamic inspection task, as well as the flight speed and flight altitude of each target dynamic inspection waypoint in the target dynamic inspection route, the current dynamic inspection route of the current unmanned aerial vehicle and the flight speed and flight altitude of each current dynamic inspection waypoint in the current dynamic inspection route are adjusted, so that the current unmanned aerial vehicle can meet the task requirements of the target static inspection task.

[0074] Optionally, when it is detected that the task priority of the candidate dynamic inspection task is lower than or equal to that of the current dynamic inspection task, the current dynamic inspection task is continued, so that each current unmanned aerial vehicle continues to execute the current dynamic inspection task along the current dynamic inspection route corresponding to the current dynamic inspection task.

[0075] The technical solution of the embodiment of the present invention divides each candidate inspection task into a candidate static inspection task and a candidate dynamic inspection task through the scheduling time and route attributes of each candidate inspection task, improving the flexibility and accuracy of task scheduling in a railway marshalling yard. For each candidate static inspection task, according to the inspection target and inspection plan of each candidate static inspection task, the candidate static inspection route of each candidate static inspection task is obtained. Based on the task scheduling model, according to the task location, task priority, and task resource demand of each candidate static inspection task, as well as the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV, the target static inspection tasks are screened from each candidate static inspection task, and the target UAVs for executing each target static inspection task are screened from each candidate UAV, so that each target UAV executes the target static inspection task along the target static inspection route corresponding to the target static inspection task, realizing the task scheduling of the candidate static inspection task. Through the task scheduling model, the scheduling efficiency and accuracy of the static inspection task can be improved. For each candidate dynamic inspection task, by obtaining the current UAVs executing each current dynamic inspection task, the task priorities of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and comparing the task priorities of each candidate dynamic inspection task with the task priorities of each current dynamic inspection task, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, the candidate dynamic inspection task is determined as the target dynamic inspection task, and the current dynamic inspection task is paused, so that each current UAV executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task, realizing the task scheduling of the candidate dynamic inspection task. Through the comparison of the task priority of the candidate dynamic inspection task and the task priority of the current dynamic inspection task, the resource preemption of the dynamic inspection task with a high task priority is realized, and during the execution of the dynamic inspection task, the task priority is ensured, which can improve the scheduling efficiency and accuracy of the dynamic inspection task, thereby improving the inspection efficiency and inspection safety of the UAV inspection in the railway marshalling yard.

[0076] In an alternative embodiment of the present invention, while pausing the current dynamic inspection task, it further includes: obtaining the current dynamic inspection route, task progress, and collected inspection data of the current dynamic inspection task, as well as the UAV position, UAV flight speed, and UAV flight altitude of the current UAV executing the current dynamic inspection task; after determining the candidate dynamic inspection task as the target dynamic inspection task, it further includes: after the target dynamic inspection task is completed, detecting the execution environment of the current dynamic inspection task according to the UAV position of the current UAV of the current dynamic inspection task; when the execution environment of the current dynamic inspection task remains unchanged, resuming the current dynamic inspection task according to the current dynamic inspection route, task progress, and collected inspection data of the current dynamic inspection task, as well as the UAV position, UAV flight speed, and UAV flight altitude of the current UAV executing the current dynamic inspection task; when the execution environment of the current dynamic inspection task changes, adjusting the resumption execution time of the current dynamic inspection task.

[0077] The current dynamic inspection route can be the dynamic inspection route of the current dynamic inspection task. The task progress of the current dynamic inspection task can be used to characterize the task completion situation of the current dynamic inspection task. The collected inspection data of the current dynamic inspection task can be the inspection data that has been collected for the current dynamic inspection task. The collected data of the current dynamic inspection task can reflect the task completion situation of the current dynamic inspection task from another perspective. The UAV position of the current UAV can be the position where the current UAV is located when pausing the current dynamic inspection task, that is, the pause position of the current dynamic inspection task. The UAV flight speed of the current UAV can be the UAV flight speed of the current UAV when pausing the current dynamic inspection task. The UAV flight altitude of the current UAV can be the UAV flight altitude of the current UAV when pausing the current dynamic inspection task. The execution environment of the current dynamic inspection task can be the environment where the current UAV is located when executing the current inspection task. Exemplarily, the execution environment of the current dynamic inspection task can include the lighting conditions and weather conditions where the current dynamic inspection task is located, etc. The execution environment of the current dynamic inspection task affects whether the current dynamic inspection task can be executed. The resumption execution time of the current dynamic inspection task can be the time to resume the execution of the current dynamic inspection task.

[0078] Specifically, while pausing the current dynamic inspection task, the current dynamic inspection route, task progress, and collected inspection data of the current dynamic inspection task can be obtained and stored, as well as the drone position, drone flight speed, and drone flight altitude of the current drone executing the current dynamic inspection task. After determining the candidate dynamic inspection task as the target dynamic inspection task, after detecting that the target dynamic inspection task is completed, the execution environment of the current dynamic inspection task can be detected according to the drone position of the current drone of the current dynamic inspection task, using the corresponding sensors. When it is detected that the execution environment of the current dynamic inspection task has not changed, the current dynamic inspection task can be resumed according to the current dynamic inspection route, task progress, and collected inspection data of the current dynamic inspection task, as well as the drone position, drone flight speed, and drone flight altitude of the current drone executing the current dynamic inspection task. When it is detected that the execution environment of the current dynamic inspection task has changed, the resume execution time of the current dynamic inspection task is adjusted so that the execution environment of the current dynamic inspection task can meet the normal execution of the current dynamic inspection task.

[0079] In this solution, when pausing the current dynamic inspection task, the current dynamic inspection route, task progress, and collected inspection data of the current dynamic inspection task, as well as the drone position, drone flight speed, and drone flight altitude of the current drone executing the current dynamic inspection task are obtained and saved, facilitating the quick resume of the current dynamic inspection task; after determining the candidate dynamic inspection task as the target dynamic inspection task, after the target dynamic inspection task is completed, the execution environment of the current dynamic inspection task is detected, and when the execution environment of the current dynamic inspection task has changed, the resume execution time of the current dynamic inspection task is adjusted, ensuring the smooth completion of the current dynamic inspection task, improving the task execution effectiveness of the current dynamic inspection task, and further improving the fault tolerance of the task scheduling process.

[0080] Embodiment 2

[0081] Figure 2The flowchart of a patrol UAV scheduling method provided by the second embodiment of the present invention. On the basis of the above embodiment, the present invention embodiment concretizes "based on the task scheduling model, according to the task location, task priority and task resource demand of each candidate static patrol task, as well as the UAV location, UAV flight speed, UAV endurance time and UAV total resource of each candidate UAV, screening each target static patrol task from each candidate static patrol task, and screening the target UAV for executing each target static patrol task from each candidate UAV" into "calculating the UAV task distance between each candidate UAV and each candidate static patrol task according to the task location of each candidate static patrol task and the UAV location of each candidate UAV; determining the UAV task completion time between each candidate UAV and each candidate static patrol task according to the UAV flight speed of each candidate UAV and the UAV task distance between each candidate UAV and each candidate static patrol task; using the task priority constraint, UAV endurance time constraint, UAV total resource constraint, task assignment constraint, minimizing the total task completion time constraint and minimizing the total resource consumption constraint, according to the task location, task priority and task resource demand of each candidate static patrol task, as well as the UAV location, UAV flight speed, UAV endurance time and UAV total resource of each candidate UAV, screening each target static patrol task from each candidate static patrol task, and screening the target UAV for executing each target static patrol task from each candidate UAV". Through the task priority constraint, UAV endurance time constraint, UAV total resource constraint, task assignment constraint, minimizing the total task completion time constraint and minimizing the total resource consumption constraint, the scheduling efficiency of the static patrol task scheduling can be improved. It should be noted that for the parts not detailed in the embodiments of the present invention, reference can be made to the descriptions of other embodiments.

[0082] See Figure 2 The patrol UAV scheduling method shown in the figure includes:

[0083] S210. Obtain the scheduling opportunity, route attribute, patrol target, patrol plan, task location, task priority and task resource demand of each candidate patrol task of the railway marshalling yard, as well as the UAV location, UAV flight speed, UAV endurance time and UAV total resource of each candidate UAV.

[0084] S220. Divide each candidate patrol task according to the scheduling opportunity and route attribute of each candidate patrol task to obtain candidate static patrol tasks and candidate dynamic patrol tasks.

[0085] S230. For each candidate static patrol task, obtain the candidate static patrol route of each candidate static patrol task according to the patrol target and patrol plan of each candidate static patrol task.

[0086] S240. For each candidate static inspection task, calculate the UAV-task distance between each candidate UAV and each candidate static inspection task according to the task location of each candidate static inspection task and the UAV location of each candidate UAV.

[0087] The UAV-task distance can be the distance between the candidate UAV and the candidate static inspection task.

[0088] Specifically, for each candidate static inspection task, the distance between the task location of each candidate static inspection task and the UAV location of each candidate UAV can be calculated to obtain the UAV-task distance between each candidate UAV and each candidate static inspection task.

[0089] S250. For each candidate static inspection task, determine the UAV-task completion time for each candidate UAV to execute each candidate static inspection task according to the UAV flight speed of each candidate UAV and the UAV-task distances between each candidate UAV and each candidate static inspection task.

[0090] The UAV-task completion time can be the ratio of the UAV-task distance to the UAV flight speed of a single candidate UAV.

[0091] Specifically, for each candidate static inspection task, the ratio of the UAV-task distances between each candidate UAV and each candidate static inspection task to the UAV flight speed of each candidate UAV can be calculated to determine the UAV-task completion time for each candidate UAV to execute each candidate static inspection task.

[0092] S260. For each candidate static inspection task, using task priority constraints, UAV endurance constraints, total UAV resource constraints, task assignment constraints, minimizing the total task completion time constraints, and minimizing the total resource consumption constraints, screen the target static inspection tasks from each candidate static inspection task, and screen the target UAVs for executing each target static inspection task from each candidate UAV according to the task location, task priority, and task resource requirements of each candidate static inspection task, and the UAV location, UAV flight speed, UAV endurance, and total UAV resources of each candidate UAV.

[0093] The task priority constraint can give priority to the execution of candidate static inspection tasks with high task priorities. The UAV endurance time constraint can ensure that the task completion time of the target UAV for the target static inspection task does not exceed the UAV endurance time. The total UAV resource constraint can ensure that the required task resources consumed by the target UAV for the target static inspection task do not exceed the total UAV resources. The task assignment constraint can ensure that a single target static inspection task is executed by a single target UAV. The constraint of minimizing the total task completion time can minimize the total task completion time of each target UAV for the corresponding target static inspection task. The constraint of minimizing the total resource consumption can minimize the total resource consumption of each target UAV for the corresponding target static inspection task.

[0094] Specifically, for each candidate static inspection task, the task priority constraint and the task assignment constraint are adopted. According to the task location, task priority, and required task resources of each candidate static inspection task, as well as the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV, a record scheduling plan for each candidate UAV to execute each candidate static inspection task is generated. The UAV endurance time constraint, the total UAV resource constraint, the constraint of minimizing the total task completion time, and the constraint of minimizing the total resource consumption are used to compare each alternative scheduling plan to determine the target scheduling plan. According to the target scheduling plan, the corresponding target static inspection tasks are selected from each candidate static inspection task, and the target UAVs for executing each target static inspection task are selected from each candidate UAV.

[0095] In an alternative embodiment of the present invention, before selecting the target static inspection tasks from each candidate static inspection task and selecting the target UAVs for executing each target static inspection task from each candidate UAV, it further includes: screening each candidate static inspection task according to the candidate static inspection area of each candidate static inspection task to obtain the first candidate static inspection tasks and the corresponding first static inspection areas; dividing each first static inspection area to obtain each first static inspection sub-area; dividing each first candidate static inspection task according to each first static inspection sub-area to obtain each first candidate static inspection sub-task; removing the first candidate static inspection tasks from each candidate static inspection task and adding each first static inspection sub-task to each candidate static inspection task to update each candidate static inspection task.

[0096] The static inspection area can be the inspection area of the candidate static inspection tasks. The first static inspection area can be a static inspection area with a relatively large coverage range. The first candidate static inspection task can be the first candidate static inspection task corresponding to the first static inspection area. The first static inspection sub-area can be the division result of the first static inspection area. Optionally, the division criteria for the first static inspection sub-areas can be set and adjusted by technicians according to experience. The first candidate static inspection sub-task can be the candidate static inspection sub-task for inspecting the first static inspection sub-areas.

[0097] Specifically, before screening the target static inspection tasks from each candidate static inspection task and screening the target drones for executing each target static inspection task from each candidate drone, the coverage range of the candidate static inspection areas of each candidate static inspection task can be compared with the preset coverage range. The candidate static inspection areas larger than the preset coverage range can be determined as the first static inspection areas, and the candidate static inspection tasks corresponding to the first static inspection areas can be determined as the first candidate static inspection tasks. Each first static inspection area can be divided based on the division criteria of the first static inspection sub-areas pre-determined by technicians to obtain each first static inspection sub-area. Each first candidate static inspection task can be divided according to each first static inspection sub-area to obtain each first candidate static inspection sub-task. The first candidate static inspection tasks can be excluded from each candidate static inspection task, and each first candidate static inspection sub-task can be added to each candidate static inspection task to update each candidate static inspection task.

[0098] This solution introduces screening of each candidate static inspection task based on the candidate static inspection areas of each candidate static inspection task, determines the first candidate static inspection tasks with a relatively large coverage range and the corresponding first static inspection areas. By dividing each first static inspection area, each first static inspection sub-area is obtained. According to each first static inspection sub-area, each first candidate static inspection task is divided to obtain each first candidate static inspection sub-task. The first candidate static inspection tasks are excluded from each candidate static inspection task, and each first static inspection sub-task is added to each candidate static inspection task to update each candidate static inspection task, realizing the splitting of the first static inspection task, so that the target drone can simultaneously inspect the first candidate static inspection sub-tasks, improving the execution efficiency of the first candidate static inspection tasks.

[0099] S270. For each candidate dynamic inspection task, obtain the current drone for executing each current dynamic inspection task, the task priority of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and compare the task priority of each candidate dynamic inspection task with the task priority of each current dynamic inspection task.

[0100] S280. For each candidate dynamic inspection task, when the task priority of the candidate dynamic inspection task is higher than that of the current dynamic inspection task, determine the candidate dynamic inspection task as the target dynamic inspection task, and pause the current dynamic inspection task, so that each current unmanned aerial vehicle (UAV) executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

[0101] In the technical solution of the embodiment of the present invention, by calculating the UAV-task distances between each candidate UAV and each candidate static inspection task according to the task positions of each candidate static inspection task and the UAV positions of each candidate UAV, and determining the UAV task completion times for each candidate UAV to execute each candidate static inspection task according to the UAV flight speeds of each candidate UAV and the UAV-task distances between each candidate UAV and each candidate static inspection task, and adopting task priority constraints, UAV endurance time constraints, total UAV resource constraints, task assignment constraints, minimizing the total task completion time constraints, and minimizing the total resource consumption constraints, according to the task positions, task priorities, and task resource requirements of each candidate static inspection task, as well as the UAV positions, UAV flight speeds, UAV endurance times, and total UAV resources of each candidate UAV, screen each target static inspection task from each candidate static inspection task, and screen the target UAVs for executing each target static inspection task from each candidate UAV. Through the task priority constraints, UAV endurance time constraints, total UAV resource constraints, task assignment constraints, minimizing the total task completion time constraints, and minimizing the total resource consumption constraints, the scheduling efficiency of static inspection task scheduling can be improved.

[0102] The present invention proposes an inspection UAV scheduling method suitable for marshalling yard inspections. The method includes the following contents:

[0103] 1. As Figure 3 shown, the inspection UAV scheduling of the present invention can include static inspection task scheduling and dynamic inspection task scheduling. According to the different scheduling times and route attributes of the candidate inspection tasks, the candidate inspection tasks can be divided into two categories: candidate static inspection tasks and candidate dynamic inspection tasks.

[0104] Among them, the candidate static inspection tasks include daily inspection tasks (i.e., daily inspection function), personnel on-track inspection tasks (i.e., personnel on-track inspection function), train status inspection tasks (i.e., train status inspection function), fixed-point inspection tasks (i.e., fixed-point inspection function), and hump humping operation inspection tasks (i.e., hump humping operation inspection function). The candidate dynamic inspection tasks can include special cargo escort inspection tasks (i.e., special cargo escort function), emergency inspection tasks (i.e., emergency inspection function), and abnormal inspection tasks (i.e., abnormal inspection function).

[0105] 2. Adopt fixed route planning and task allocation to schedule candidate static inspection tasks.

[0106] First, according to the inspection objectives and inspection plans of the candidate static inspection tasks, obtain the corresponding candidate static inspection routes.

[0107] Among them, different types of inspection objectives can include catenaries, freight vehicles, railway tracks, and operating personnel, etc. Factors such as their shapes, sizes, and distribution characteristics determine the route planning method of the UAV. For example, catenary networks are usually linearly distributed. The candidate static inspection route of the UAV for this candidate static inspection task can be planned along the catenary line direction to ensure full coverage of each section of the line. Operating personnel may cross different railway tracks. It is necessary to plan a route that can cover the operation site and adapt to terrain changes according to the walking direction of the operating personnel and the surrounding environment in the inspection plan. At the same time, it is also necessary to strictly abide by the laws and regulations of railways and marshalling yards to ensure that the UAV flies in a legal airspace. That is, according to the no-fly zone range and restricted flight zone range, avoid route planning in the no-fly zone range and formulate a reasonable flight altitude for the restricted flight zone range.

[0108] Second, according to the task requirements and the performance characteristics of the UAV, allocate different candidate static inspection tasks to suitable candidate UAVs.

[0109] First, according to the coverage situation of the candidate static inspection tasks in the railway marshalling yard, combined with factors such as the task location, task priority, task resource demand of each candidate static inspection task, and the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV, allocate to suitable candidate UAVs according to the specific requirements of different candidate static inspection tasks, such as data acquisition accuracy requirements. For example, for the first static inspection area with a large coverage area, the first static inspection area can be divided into multiple first static inspection sub-areas, and the corresponding first candidate static inspection sub-tasks can be determined, and multiple candidate UAVs can be allocated to conduct inspections simultaneously to improve the task completion efficiency. For candidate static inspection tasks that require high-precision detection, they can be allocated to candidate UAVs equipped with high-resolution sensors. At the same time, a task scheduling model can be established, adopting task priority constraints, UAV endurance time constraints, total UAV resource constraints, task allocation constraints, minimizing the total task completion time constraints, and minimizing the total resource consumption constraints. According to the task location, task priority, and task resource demand of each candidate static inspection task, and the UAV location, UAV flight speed, UAV endurance time, and total UAV resources of each candidate UAV, screen the target static inspection tasks among the candidate static inspection tasks, and screen the target UAVs for executing each target static inspection task among the candidate UAVs.

[0110] Exemplarily, assume there are n candidate drones and m candidate static inspection tasks. Each candidate drone has different performance parameters and resource limitations, and each candidate static inspection task also has different task requirements and task priorities. At any given moment, new candidate static inspection tasks may emerge or the status of existing candidate static inspection tasks may change, requiring dynamic scheduling of the candidate drones. The goal is to minimize the total task completion time constraint and the total resource consumption constraint, while satisfying the task priority constraint, the drone endurance time constraint, the total drone resource constraint, and the task assignment constraint.

[0111] The task scheduling model is defined as follows:

[0112] U = {u 1 , u 2 , … u n}: Represents the set of candidate drones.

[0113] T = {t 1 , t 2 , … t m}: Represents the set of candidate static inspection tasks.

[0114] d ij : Represents the drone task distance from candidate drone u i to candidate static inspection task t j .

[0115] v i : Represents the flight speed of candidate drone u i .

[0116] e i : Represents the endurance time of candidate drone u i .

[0117] p j : Represents the task priority of candidate static inspection task t j .

[0118] r ij : Represents the required task resource demand for candidate drone u i to execute candidate static inspection task t j .

[0119] R i : Represents the total drone resources of candidate drone u i .

[0120] t ij : Represents the required drone task completion time for candidate drone u i to execute candidate static inspection task t j .

[0121] x ij : represents a decision variable. If candidate drone u i executes candidate static inspection task t j , then x ij = 1, otherwise x ij = 0.

[0122] Total task completion time minimization constraint:

[0123] Total resource consumption minimization constraint:

[0124] Task assignment constraint: Each task can only be executed by one drone:

[0125] Total drone resource constraint: The amount of resources required for a drone to execute a task cannot exceed its total resources:

[0126]

[0127] Task priority constraint: High-priority tasks must be executed first: If p j > p k , then t ij + t ik ≤ t kj + t ki .

[0128] The above task scheduling model aims to minimize the total task completion time constraint and the total resource consumption minimization constraint, while satisfying the task priority constraint, the drone endurance time constraint, the total drone resource constraint, and the task assignment constraint. By solving the task scheduling model using the integer programming algorithm, an optimal task scheduling scheme can be obtained, thereby improving the utilization efficiency of candidate drones and the quality of task completion.

[0129] For example, assume there are 3 candidate drones and 5 candidate static inspection tasks, and the drone performance parameters and task requirements are shown in Table 1 and Table 2 below.

[0130] Table 1 Drone Performance Parameter Table

[0131]

[0132] Table 2 Drone Task Requirement Table

[0133]

[0134]

[0135] Based on the data in Table 1 and Table 2 above, the UAV task distance between the candidate UAVs and the candidate static inspection tasks, and the UAV task completion time required for the candidate UAVs to execute the candidate static inspection tasks can be calculated.

[0136] Table 3 UAV Task Distance Table

[0137]

[0138] Table 4 UAV Task Completion Time Table

[0139]

[0140] Then, substitute the above data into the task scheduling model and use the integer programming algorithm to solve it to obtain the data in Table 5 and Table 6.

[0141] Table 5 Total Task Completion Time Table

[0142]

[0143]

[0144] Table 6 Total Resource Consumption Table

[0145]

[0146]

[0147] With the goal of minimizing the total task completion time constraint and the total resource consumption constraint, while satisfying the task priority constraint, the UAV endurance time constraint, the total UAV resource constraint, and the task assignment constraint, the inspection UAV scheduling results in Table 7 are obtained.

[0148] Table 7 Inspection UAV Scheduling Results Table

[0149] Candidate UAV Task Assignment Execution Time (s) Resource Consumption <![CDATA[u 1 > <![CDATA[t 4 > 204.12 25 <![CDATA[u 2 > <![CDATA[t 5 > 90.14 10 <![CDATA[u 3 > <![CDATA[t 1 > 130.17 20

[0150] As shown in Table 7, normalize and synthesize the total task completion time and the total resource consumption to obtain the final inspection UAV scheduling results, that is, select the target static inspection tasks from each candidate static inspection task, and select the target UAVs to execute each target static inspection task from each candidate UAV, so that each target UAV executes the target static inspection task along the target static inspection route corresponding to the target static inspection task.

[0151] After obtaining the final inspection UAV scheduling result, a UAV mission operation map can be drawn, which can more intuitively display the target position and mission progress when the target UAV conducts inspections according to the target static inspection route. That is, the task assignment situation and task execution situation are displayed in the form of a UAV mission operation map, so that the target static inspection route matches the time. The inspection data collected by the target UAV can be processed and analyzed to identify problems such as equipment failures and defects in the railway marshalling yard. At the same time, the inspection data can be statistically analyzed to generate reports and charts, providing data support for decision-making.

[0152] 3. Adopt a preemptive strategy to achieve the scheduling of candidate dynamic inspection tasks.

[0153] The preemptive strategy means that during the process of the current UAV executing the current dynamic inspection task, when a candidate dynamic inspection task with a higher task priority appears, the current dynamic inspection task is suspended, and resources are preferentially allocated to the new target dynamic inspection task (i.e., the candidate dynamic inspection task with a higher task priority). After the new target dynamic inspection task is completed, the suspended current dynamic inspection task is continued. Its principle is to dynamically adjust the dynamic inspection tasks based on task priorities to ensure that the most important dynamic inspection tasks can be processed in a timely manner, thereby improving the efficiency and adaptability of the UAV inspection system.

[0154] First is the priority setting.

[0155] The priority can be divided into task type priority, time urgency priority, and resource requirement priority. Among them, the task type priority sets different task type priorities according to the nature and importance of the task. For example, the priority of the emergency fault detection task is higher than that of the regular inspection task; the priority of the inspection task involving safety-critical equipment is higher than that of the general equipment inspection task. The time urgency priority assigns a higher priority to tasks with strict time requirements, such as tasks to be completed within a specific time window or time-sensitive emergency rescue tasks. The resource requirement priority considers the degree of resource requirements of the task, and sets tasks with large resource requirements and significant impacts on system operation as higher priorities. For example, tasks that require a large amount of data transmission and processing may occupy more communication and computing resources, and such tasks can be given higher priorities during resource allocation. For the task type priority, time urgency priority, and resource requirement priority, the task priority decreases.

[0156] Secondly is the resource reallocation.

[0157] When task preemption occurs, first adjust the flight resources of the current UAV. According to the task requirements of the new target dynamic inspection task, re-plan parameters such as the current dynamic inspection route, UAV flight speed, and UAV flight altitude of the current UAV. If the new target dynamic inspection task requires a faster response time, the current UAV can increase the flight speed or select a more direct target dynamic inspection route; if the new target dynamic inspection task has special requirements for the UAV flight altitude, the flight altitude of the current UAV will be adjusted to meet the task requirements of the target dynamic inspection task.

[0158] Finally, it is task recovery and continued execution.

[0159] Because when the task is preempted, the status information of the currently paused dynamic inspection task will be saved, including the task progress, the collected inspection data, and the UAV flight parameters of the current UAV, etc. After the new target dynamic inspection task is completed, the preempted task will be restored according to the saved status information of the current dynamic inspection task and continue to execute. When restoring the task, it will be checked whether the execution environment of the current dynamic inspection task has changed. If there is a change, the recovery execution time of the current dynamic inspection task will be appropriately adjusted according to the new execution environment.

[0160] The present invention designs an intelligent scheduling method for UAVs suitable for inspection in marshalling yards, which completes the unified command and scheduling of multiple UAVs, has functions such as task planning, route setting, and UAV status monitoring. Through an intelligent task allocation algorithm, the inspection tasks are reasonably arranged to ensure the efficient and orderly progress of the inspection work of the entire railway marshalling yard.

[0161] Embodiment III

[0162] Figure 6 It is a schematic structural diagram of an inspection UAV scheduling device provided in Embodiment III of the present invention. The embodiment of the present invention is applicable to the situation of scheduling inspection UAVs in railway marshalling yards. This device can execute the inspection UAV scheduling method. This device can be implemented in the form of hardware and / or software, and this device can be configured in an electronic device carrying the inspection UAV scheduling function.

[0163] See Figure 6The shown inspection UAV scheduling device includes: an inspection task acquisition module 610, an inspection task division module 620, a static inspection task route planning module 630, a static inspection task scheduling module 640, a dynamic inspection task priority comparison module 650, and a dynamic inspection task scheduling module 660. Among them, the inspection task acquisition module 610 is used to acquire the scheduling time, route attributes, inspection targets, inspection plans, task locations, task priorities, and task resource requirements of each candidate inspection task in the railway marshalling yard, as well as the UAV locations, UAV flight speeds, UAV endurance times, and total UAV resources of each candidate UAV; the inspection task division module 620 is used to divide each of the candidate inspection tasks according to the scheduling time and route attributes of each candidate inspection task to obtain candidate static inspection tasks and candidate dynamic inspection tasks; the static inspection task route planning module 630 is used to obtain candidate static inspection routes for each of the candidate static inspection tasks according to the inspection targets and inspection plans of each candidate static inspection task; the static inspection task scheduling module 640 is used to, for each of the candidate static inspection tasks, based on the task scheduling model, according to the task locations, task priorities, and task resource requirements of each candidate static inspection task, as well as the UAV locations, UAV flight speeds, UAV endurance times, and total UAV resources of each candidate UAV, screen out target static inspection tasks from each of the candidate static inspection tasks, and screen out target UAVs for executing each of the target static inspection tasks from each of the candidate UAVs, so that each of the target UAVs executes the target static inspection task along the target static inspection route corresponding to the target static inspection task; the dynamic inspection task priority comparison module 650 is used to, for each of the candidate dynamic inspection tasks, obtain the current UAVs for executing each current dynamic inspection task, the task priorities of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and compare the task priorities of each candidate dynamic inspection task with the task priorities of each current dynamic inspection task; the dynamic inspection task scheduling module 660 is used to, for each of the candidate dynamic inspection tasks, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, determine the candidate dynamic inspection task as the target dynamic inspection task and pause the current dynamic inspection task, so that each of the current UAVs executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

[0164] The technical solution of the embodiment of the present invention divides each candidate inspection task according to the scheduling time and route attributes of each candidate inspection task, obtaining candidate static inspection tasks and candidate dynamic inspection tasks, improving the flexibility and accuracy of task scheduling in a railway marshalling station. For each candidate static inspection task, according to the inspection objectives and inspection plans of each candidate static inspection task, candidate static inspection routes of each candidate static inspection task are obtained. Based on the task scheduling model, according to the task positions, task priorities, and task resource requirements of each candidate static inspection task, as well as the drone positions, drone flight speeds, drone endurance times, and total drone resources of each candidate drone, target static inspection tasks are screened from each candidate static inspection task, and target drones for executing each target static inspection task are screened from each candidate drone, so that each target drone executes the target static inspection task along the target static inspection route corresponding to the target static inspection task, realizing the task scheduling of candidate static inspection tasks. Through the task scheduling model, the scheduling efficiency and accuracy of static inspection tasks can be improved. For each candidate dynamic inspection task, by obtaining the current drones executing each current dynamic inspection task, the task priorities of each current dynamic inspection task, and the candidate dynamic inspection routes of each candidate dynamic inspection task, and comparing the task priorities of each candidate dynamic inspection task with the task priorities of each current dynamic inspection task, when the task priority of a candidate dynamic inspection task is higher than the task priority of a current dynamic inspection task, the candidate dynamic inspection task is determined as a target dynamic inspection task, and the current dynamic inspection task is paused, so that each current drone executes the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task, realizing the task scheduling of candidate dynamic inspection tasks. Through the comparison of the task priorities of candidate dynamic inspection tasks and current dynamic inspection tasks, resource preemption of dynamic inspection tasks with high task priorities is realized, and during the execution of dynamic inspection tasks, the task priority is guaranteed, which can improve the scheduling efficiency and accuracy of dynamic inspection tasks, thereby improving the inspection efficiency and inspection safety of drone inspections in railway marshalling stations.

[0165] In an alternative embodiment of the present invention, the static inspection task scheduling module 640 includes: a UAV task distance calculation unit for calculating the UAV task distances between each of the candidate UAVs and each of the candidate static inspection tasks according to the task positions of each of the candidate static inspection tasks and the UAV positions of each of the candidate UAVs; a UAV task completion time calculation unit for determining the UAV task completion times of each of the candidate UAVs for executing each of the candidate static inspection tasks according to the UAV flight speeds of each of the candidate UAVs and the UAV task distances between each of the candidate UAVs and each of the candidate static inspection tasks; a first static inspection task scheduling unit for screening each target static inspection task from each of the candidate static inspection tasks and screening the target UAVs for executing each of the target static inspection tasks from each of the candidate UAVs by using task priority constraints, UAV endurance time constraints, total UAV resource constraints, task assignment constraints, minimizing the total task completion time constraint, and minimizing the total resource consumption constraint, according to the task positions, task priorities, and task resource requirements of each of the candidate static inspection tasks, and the UAV positions, UAV flight speeds, UAV endurance times, and total UAV resources of each of the candidate UAVs.

[0166] In an alternative embodiment of the present invention, the static inspection task scheduling module 640 further includes: a candidate static inspection task screening unit for screening each target static inspection task from each of the candidate static inspection tasks and screening the target UAVs for executing each of the target static inspection tasks from each of the candidate UAVs, and before that, screening each of the candidate static inspection tasks according to the candidate static inspection areas of each of the candidate static inspection tasks to obtain the first candidate static inspection tasks and the corresponding first static inspection areas; a first static inspection area division unit for dividing each of the first static inspection areas to obtain each first static inspection sub-area; a first candidate static inspection task division unit for dividing each of the first candidate static inspection tasks according to each of the first static inspection sub-areas to obtain each first candidate static inspection sub-task; a candidate static inspection task update unit for removing the first candidate static inspection tasks from each of the candidate static inspection tasks and adding each of the first candidate static inspection sub-tasks to each of the candidate static inspection tasks to update each of the candidate static inspection tasks.

[0167] In an alternative embodiment of the present invention, the device further includes: a no-fly zone range acquisition module, configured to, after obtaining the candidate static inspection routes of each of the candidate static inspection tasks according to the inspection targets and inspection plans of each of the candidate static inspection tasks, obtain the no-fly zone range of the railway marshalling yard, and determine the flight altitude of each candidate static inspection waypoint in each candidate static inspection route according to the overlapping situation between the candidate static inspection routes of each of the candidate static inspection tasks and the no-fly zone range; a drone task operation map generation module, configured to, after screening out the target static inspection tasks from each of the candidate static inspection tasks and screening out the target drones for executing each of the target static inspection tasks from each of the candidate drones, generate a drone task operation map according to the target static inspection routes of each of the target static inspection tasks and the flight altitude of each target static inspection waypoint in each of the target static inspection routes; a cross-waypoint flight altitude adjustment module, configured to adjust the flight altitude of each of the target static inspection routes at each of the cross-waypoints based on the cross-waypoints in the drone task operation map, so as to avoid collisions between different target drones at the same cross-waypoint.

[0168] In an alternative embodiment of the present invention, the static inspection task scheduling module 640 includes: an inspection accuracy acquisition unit, configured to, for each of the candidate static inspection tasks, obtain the inspection accuracy of each of the candidate drones, and determine the inspection accuracy required for each of the candidate static inspection tasks according to the inspection plan of each of the candidate static inspection tasks; a static inspection task matching unit, configured to, for each of the candidate static inspection tasks, match each of the candidate drones and each of the candidate static inspection tasks according to the inspection accuracy of each of the candidate drones and the inspection accuracy required for each of the candidate static inspection tasks, so that the inspection accuracy of the candidate drone meets the inspection accuracy required for the candidate static inspection task; a second static inspection task scheduling unit, configured to, for each of the candidate drones and each of the candidate static inspection tasks that are matched, based on a task scheduling model, screen out the target static inspection tasks from each of the candidate static inspection tasks and screen out the target drones for executing each of the target static inspection tasks from each of the candidate drones according to the task location, task priority, and task resource demand of each of the candidate static inspection tasks and the drone location, drone flight speed, drone endurance time, and total drone resources of each of the candidate drones.

[0169] In an alternative embodiment of the present invention, the device further includes: a current dynamic inspection data acquisition module, configured to acquire the current dynamic inspection route, task progress, and acquired inspection data of the current dynamic inspection task, as well as the UAV position, UAV flight speed, and UAV flight altitude of the current UAV executing the current dynamic inspection task while pausing the current dynamic inspection task; an execution environment detection module, configured to detect the execution environment of the current dynamic inspection task according to the UAV position of the current UAV of the current dynamic inspection task after determining the candidate dynamic inspection task as the target dynamic inspection task and after the target dynamic inspection task is completed; a current dynamic inspection task resumption module, configured to resume the current dynamic inspection task according to the current dynamic inspection route, task progress, and acquired inspection data of the current dynamic inspection task, as well as the UAV position, UAV flight speed, and UAV flight altitude of the current UAV executing the current dynamic inspection task when the execution environment of the current dynamic inspection task remains unchanged; a resumed execution time adjustment module, configured to adjust the resumed execution time of the current dynamic inspection task when the execution environment of the current dynamic inspection task changes.

[0170] The inspection UAV scheduling device provided by the embodiments of the present invention can execute the inspection UAV scheduling method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0171] In the technical solution of the embodiments of the present invention, the collected information is information and data authorized by the user or fully authorized by all parties, and the processing of relevant data such as collection, storage, use, processing, transmission, provision, disclosure, and application complies with relevant laws, regulations, and standards of relevant countries and regions, takes necessary confidentiality measures, does not violate public order and good customs, and provides a corresponding operation entry for the user to choose to authorize or reject.

[0172] Embodiment 4

[0173] Figure 7 FIG. shows a schematic structural diagram of an electronic device 700 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0174] As shown Figure 7 in FIG. 1, the electronic device 700 includes at least one processor 701 and a memory communicatively connected to the at least one processor 701, such as a read-only memory (ROM) 702, a random access memory (RAM) 703, etc. The memory stores a computer program executable by the at least one processor. The processor 701 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 702 or the computer program loaded from the storage unit 708 into the random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 can also be stored. The processor 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0175] A plurality of components in the electronic device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the electronic device 700 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0176] The processor 701 can be various general-purpose and / or dedicated processing components with processing and computing capabilities. Some examples of the processor 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 701 executes the various methods and processes described above, such as the inspection UAV scheduling method.

[0177] In some embodiments, the inspection UAV scheduling method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the processor 701, one or more steps of the inspection UAV scheduling method described above can be executed. Alternatively, in other embodiments, the processor 701 can be configured to execute the inspection UAV scheduling method by any other suitable means (e.g., by means of firmware).

[0178] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0179] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0180] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0181] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0182] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0183] The computing system can include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS (Virtual Private Server) services.

[0184] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0185] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for dispatching inspection drones, characterized in that: The method comprises: Obtain the scheduling timing, route attributes, inspection targets, inspection plans, task locations, task priorities, and task resource requirements of each candidate inspection task at the railway marshaling yard, as well as the drone location, drone flight speed, drone endurance time, and total amount of drone resources of each candidate drone; According to the scheduling timing and route attributes of each candidate inspection task, each candidate inspection task is divided to obtain candidate static inspection tasks and candidate dynamic inspection tasks; For each of the candidate static inspection tasks, according to the inspection target and inspection plan of each of the candidate static inspection tasks, a candidate static inspection route of each of the candidate static inspection tasks is obtained; For each of the candidate static inspection tasks, based on the task scheduling model, according to the task position, task priority and task resource requirement of each of the candidate static inspection tasks and the drone position, drone flight speed, drone endurance and total drone resource of each of the candidate drones, each target static inspection task is screened from each of the candidate static inspection tasks, and a target drone that performs each of the target static inspection tasks is screened from each of the candidate drones, so that each of the target drones performs the target static inspection task along the target static inspection route corresponding to the target static inspection task; For each of the candidate dynamic inspection tasks, obtain the current drone that executes each current dynamic inspection task, the task priority of each current dynamic inspection task, and the candidate dynamic inspection route of each of the candidate dynamic inspection tasks, and compare the task priority of each of the candidate dynamic inspection tasks with the task priority of each current dynamic inspection task; For each of the candidate dynamic inspection tasks, when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, the candidate dynamic inspection task is determined as the target dynamic inspection task, and the current dynamic inspection task is paused so that each of the current UAVs performs the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

2. The method according to claim 1, characterized in that The task scheduling model is based on the task location, task priority and task resource requirement of each candidate static inspection task and the drone location, drone flight speed, drone endurance and total drone resource of each candidate drone, screening each target static inspection task from each candidate static inspection task, and screening the target drone to perform each target static inspection task from each candidate drone, including: Calculating the drone task distance between each of the candidate drones and each of the candidate static inspection tasks according to the task location of each of the candidate static inspection tasks and the drone location of each of the candidate drones; Determine the completion time of each UAV task between each candidate UAV performing each candidate static inspection task according to the UAV flight speed of each candidate UAV and each UAV task distance between each candidate UAV and each candidate static inspection task; By adopting task priority constraints, UAV endurance time constraints, total UAV resource constraints, task allocation constraints, minimization of total task completion time constraints and minimization of total resource consumption constraints, each target static inspection task is screened from each of the candidate static inspection tasks, and a target UAV that executes each of the target static inspection tasks is screened from each of the candidate UAVs according to the task position, task priority and task resource requirement of each of the candidate static inspection tasks and the UAV position, UAV flight speed, UAV endurance time and total UAV resource of each of the candidate UAVs.

3. The method according to claim 2, characterized in that Before selecting each target static inspection task from each candidate static inspection task and selecting a target drone to perform each target static inspection task from each candidate drone, the method further includes: According to the candidate static inspection areas of the candidate static inspection tasks, the candidate static inspection tasks are screened to obtain a first candidate static inspection task and a corresponding first static inspection area; Dividing each of the first static inspection areas to obtain each of the first static inspection sub-areas; Dividing each of the first candidate static inspection tasks according to each of the first static inspection sub-areas to obtain each of the first candidate static inspection sub-tasks; The first candidate static inspection task is eliminated from each of the candidate static inspection tasks, and each of the first candidate static inspection subtasks is added to each of the candidate static inspection tasks to update each of the candidate static inspection tasks.

4. The method according to claim 1, characterized in that: After acquiring the candidate static inspection routes of each candidate static inspection task according to the inspection target and inspection plan of each candidate static inspection task, the method further includes: Obtaining the restricted flight zone range of the railway marshalling yard, and determining the flight altitude of each candidate static inspection waypoint in each candidate static inspection route according to the overlap between the candidate static inspection route of each candidate static inspection task and the restricted flight zone range; After selecting each target static inspection task from each candidate static inspection task, and selecting a target drone to perform each target static inspection task from each candidate drone, the method further includes: Generate a UAV mission operation diagram according to the target static inspection route of each target static inspection task and the flight altitude of each target static inspection waypoint in each target static inspection route; Based on the intersection waypoints in the UAV mission operation diagram, the flight altitude of each target static inspection route at each intersection waypoint is adjusted to avoid collision between different target UAVs at the same intersection waypoint.

5. The method according to claim 1, characterized in that The method of selecting target static inspection tasks from among the candidate static inspection tasks, and selecting target drones to perform the target static inspection tasks from among the candidate drones based on the task scheduling model and according to the task location, task priority, and task resource requirement of each candidate static inspection task, as well as the drone location, drone flight speed, drone endurance, and total drone resource of each candidate drone, includes: For each of the candidate static inspection tasks, the inspection accuracy of each of the candidate UAVs is obtained, and according to the inspection plan of each of the candidate static inspection tasks, the inspection accuracy required for each of the candidate static inspection tasks is determined; For each of the candidate static inspection tasks, according to the inspection accuracy of each of the candidate drones and the inspection accuracy required for each of the candidate static inspection tasks, each of the candidate drones is matched with each of the candidate static inspection tasks, so that the inspection accuracy of the candidate drone meets the inspection accuracy required for the candidate static inspection tasks; For each matched candidate UAV and each candidate static inspection task, based on the task scheduling model, according to the task position, task priority and task resource requirement of each candidate static inspection task as well as the UAV position, UAV flight speed, UAV endurance time and total UAV resources of each candidate UAV, each target static inspection task is screened from each candidate static inspection task, and the target UAV to execute each target static inspection task is screened from each candidate UAV.

6. The method according to claim 1, characterized in that While pausing the current dynamic inspection task, the method further includes: Obtaining the current dynamic inspection route, task progress and collected inspection data of the current dynamic inspection task, as well as the drone position, drone flight speed and drone flight altitude of the current drone performing the current dynamic inspection task; After determining the candidate dynamic inspection task as the target dynamic inspection task, the method further includes: After the target dynamic inspection task is executed, detecting the execution environment of the current dynamic inspection task according to the drone position of the current drone of the current dynamic inspection task; When the execution environment of the current dynamic inspection task has not changed, the current dynamic inspection task is restored according to the current dynamic inspection route, task progress and collected inspection data of the current dynamic inspection task, as well as the drone position, drone flight speed and drone flight altitude of the current drone executing the current dynamic inspection task; When the execution environment of the current dynamic inspection task changes, the recovery execution time of the current dynamic inspection task is adjusted.

7. A patrol drone dispatching device, characterized in that: The device comprises: The inspection task acquisition module is used to obtain the scheduling timing, route attributes, inspection targets, inspection plans, task locations, task priorities and task resource requirements of each candidate inspection task of the railway marshaling yard, as well as the drone location, drone flight speed, drone endurance time and total amount of drone resources of each candidate drone; An inspection task division module is used to divide each candidate inspection task according to the scheduling timing and route attributes of each candidate inspection task to obtain candidate static inspection tasks and candidate dynamic inspection tasks; A static inspection task route planning module is used to obtain a candidate static inspection route for each candidate static inspection task according to the inspection target and inspection plan of each candidate static inspection task; A static inspection task scheduling module is used for selecting each target static inspection task from each candidate static inspection task based on a task scheduling model and according to the task position, task priority and task resource requirement of each candidate static inspection task and the drone position, drone flight speed, drone endurance and total drone resource of each candidate drone, and selecting a target drone to perform each target static inspection task from each candidate drone, so that each target drone performs the target static inspection task along the target static inspection route corresponding to the target static inspection task; A dynamic inspection task priority comparison module is used to obtain, for each candidate dynamic inspection task, the current drone that executes each current dynamic inspection task, the task priority of each current dynamic inspection task, and the candidate dynamic inspection route of each candidate dynamic inspection task, and compare the task priority of each candidate dynamic inspection task with the task priority of each current dynamic inspection task; The dynamic inspection task scheduling module is used to determine each candidate dynamic inspection task as a target dynamic inspection task and suspend the current dynamic inspection task for each candidate dynamic inspection task when the task priority of the candidate dynamic inspection task is higher than the task priority of the current dynamic inspection task, so that each current UAV performs the target dynamic inspection task along the target dynamic inspection route corresponding to the target dynamic inspection task.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the inspection drone scheduling method described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the inspection drone scheduling method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the inspection drone scheduling method according to any one of claims 1 to 6.

Images