Automatic inspection method and system and computer readable storage medium

Abstract

The invention provides an automatic inspection method and system and a computer readable storage medium, at least one candidate driving path of an inspection vehicle is generated according to inspection task parameters, an inspection blind area of the vehicle in the inspection process is determined based on the candidate driving path, and inspection work of the inspection blind area is automatically completed by an unmanned aerial vehicle step by step. And cooperative inspection work of the vehicle and the unmanned aerial vehicle is realized. And planning an unmanned aerial vehicle inspection sub-task for the inspection blind area, associating the position of the unmanned aerial vehicle take-off and landing point with the candidate driving path, generating a target driving path of the inspection vehicle, and ensuring that the whole inspection process is globally optimal and the path is safe and reliable. Meanwhile, the target driving path passes through the unmanned aerial vehicle take-off and landing point, so that the unmanned aerial vehicle inspection subtask is triggered and executed when the inspection vehicle drives to the unmanned aerial vehicle take-off and landing point along the target driving path, seamless cooperative operation of the inspection vehicle and the unmanned aerial vehicle in a complex environment is realized, manual intervention is reduced, and the inspection efficiency and safety are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to an automatic inspection method, system and computer readable storage medium. BACKGROUND

[0002] In traditional inspection operations (such as traffic facilities, park security, etc.), manual inspection or single-device automatic inspection is usually adopted.

[0003] The traditional inspection method has some limitations, such as low efficiency, high cost, safety risks, and is greatly affected by weather, terrain and other factors. When a vehicle is used for inspection, there are visual blind spots in the vehicle camera, especially in complex environments such as high-rise buildings, bridges and pipelines, it is difficult to achieve accurate and comprehensive inspection. SUMMARY

[0004] To overcome the problems in the related art, the present specification provides an automatic inspection method, system and computer readable storage medium.

[0005] In a first aspect, an automatic inspection method is provided, the method comprising:

[0006] obtaining an inspection task parameter, the task parameter comprising an inspection area;

[0007] generating at least one candidate driving path of an inspection vehicle based on the inspection area;

[0008] in response to determining that there is at least one inspection blind spot formed by the vehicle camera visual angle limitation on the candidate driving path, planning a UAV inspection sub-task for the inspection blind spot, the UAV inspection sub-task being associated with a UAV take-off and landing point;

[0009] associating the position of the UAV take-off and landing point with the candidate driving path to generate a target driving path of the inspection vehicle, the target driving path passing through the UAV take-off and landing point, so that the inspection vehicle triggers the execution of the UAV inspection sub-task when driving to the UAV take-off and landing point along the target driving path.

[0010] According to the automatic inspection method provided by the present application, the association of the position of the UAV take-off and landing point with the candidate driving path to generate a target driving path of the inspection vehicle comprises:

[0011] combining at least one candidate driving path and at least one candidate inspection route of the UAV inspection sub-task to generate a plurality of candidate task execution schemes for executing the inspection task, the candidate inspection route being determined according to the positions of a plurality of UAV take-off and landing points in the inspection blind spot for the UAV to take off and land;

[0012] determining a total time required for execution of each of the candidate task execution schemes;

[0013] selecting a candidate task execution scheme with the shortest total time as a target task execution scheme, and a candidate travel path in the target task execution scheme as the target travel path of the inspection vehicle.

[0014] According to the automatic inspection method provided in the present application, the method further comprises:

[0015] determining a plurality of unmanned aerial vehicle landing points in each of the inspection blind areas for takeoff and landing of the unmanned aerial vehicle, and generating a plurality of candidate inspection flight paths of the unmanned aerial vehicle inspection subtask containing the unmanned aerial vehicle landing points;

[0016] combining the plurality of candidate inspection flight paths to obtain a plurality of initial inspection flight paths;

[0017] The combination of at least one candidate travel path and the candidate inspection flight path of at least one unmanned aerial vehicle inspection subtask to generate a plurality of candidate task execution schemes for executing the inspection task comprises:

[0018] Combining at least one candidate travel path and a plurality of initial inspection flight paths to generate a plurality of candidate task execution schemes for executing the inspection task, wherein the candidate task execution scheme comprises one candidate travel path and one initial inspection flight path associated therewith.

[0019] According to the automatic inspection method provided in the present application, the method further comprises:

[0020] determining a plurality of candidate inspection flight paths corresponding to the initial inspection flight path in the target task execution scheme as target inspection flight paths for executing the unmanned aerial vehicle inspection subtask.

[0021] According to the automatic inspection method provided in the present application, the task parameters include an inspection time period; the method further comprises:

[0022] obtaining historical road condition information of the inspection time period;

[0023] In the process of planning the candidate travel path and / or the target travel path, the corresponding path and the unmanned aerial vehicle landing point associated with the path are adjusted based on the historical road condition information.

[0024] According to the automatic inspection method provided in the present application, the inspection vehicle triggers execution of the unmanned aerial vehicle inspection subtask when traveling along the target travel path to the unmanned aerial vehicle landing point, comprising:

[0025] When the inspection vehicle travels along the target travel path to the unmanned aerial vehicle landing point, the vehicle camera acquires surrounding environment information.

[0026] In a case where it is identified that the surrounding environment satisfies the safety condition for the UAV to take off and land, the UAV is controlled to take off and perform the UAV inspection sub-task, or the UAV is controlled to land.

[0027] According to the automatic inspection method provided in the application, the safety condition comprises a case where the surrounding environment has no obstacle;

[0028] The surrounding environment information is acquired through the vehicle camera, comprising:

[0029] The vehicle camera is adjusted to turn to a take-off and landing area of the UAV on the inspection vehicle, and an environment image is collected;

[0030] Obstacle detection is performed on the environment image to determine whether there is an obstacle in the surrounding environment.

[0031] In a second aspect, an automatic inspection system is provided, and the system comprises:

[0032] An inspection planning unit is configured to implement the automatic inspection method according to any one of the first aspect;

[0033] An inspection vehicle is configured to perform an inspection task based on a target driving path issued by the inspection planning unit, and when passing through a UAV take-off and landing point, trigger the UAV to synchronously perform a UAV inspection sub-task;

[0034] A UAV is configured to perform an inspection task based on a target inspection route issued by the inspection planning unit.

[0035] In a third aspect, an inspection planning unit is provided, and the inspection planning unit is configured to implement the automatic inspection method according to any one of the first aspect.

[0036] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores an automatic inspection program, and the automatic inspection program performs steps of any one of the automatic inspection methods.

[0037] The application further provides a computer program product comprising a computer program, and the computer program is executed by a processor to implement the automatic inspection method according to any one of the above.

[0038] The automatic inspection method, system and computer readable storage medium provided in the embodiments of the present application have the following beneficial effects:

[0039] According to the inspection task parameters, at least one candidate driving path of the inspection vehicle is generated, and a blind area of the vehicle in the inspection process is determined based on the candidate driving path. The blind area is completed by the unmanned aerial vehicle to realize the cooperative inspection operation of the vehicle and the unmanned aerial vehicle. The unmanned aerial vehicle inspection sub-task is planned for the blind area, the position of the unmanned aerial vehicle take-off and landing point is associated with the candidate driving path, the target driving path of the inspection vehicle is generated, and the whole inspection process is globally optimal, safe and reliable. At the same time, the target driving path passes through the unmanned aerial vehicle take-off and landing point, so that the inspection vehicle triggers the unmanned aerial vehicle inspection sub-task when driving to the unmanned aerial vehicle take-off and landing point along the target driving path, realizes seamless cooperative operation of the inspection vehicle and the unmanned aerial vehicle in a complex environment, reduces manual intervention, and improves inspection efficiency and safety.

[0040] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present specification. BRIEF DESCRIPTION OF DRAWINGS

[0041] The drawings incorporated into the specification and forming a part of the specification, show embodiments consistent with the present specification, and together with the specification, serve to explain the principles of the present specification.

[0042] Figure 1 is a flow diagram of an automatic inspection method according to an exemplary embodiment of the present specification;

[0043] Figure 2 is a schematic diagram of route planning according to an exemplary embodiment of the present specification;

[0044] Figure 3 is another schematic diagram of route planning according to an exemplary embodiment of the present specification;

[0045] Figure 4 is a variation schematic diagram of route planning according to an exemplary embodiment of the present specification;

[0046] Figure 5 is a schematic block diagram of an automatic inspection device according to an exemplary embodiment of the present specification. DETAILED DESCRIPTION

[0047] Hereinafter, the technical solutions in the embodiments (or "embodiments") of the present application will be described clearly and completely with reference to the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated.

[0048] If the application embodiments involve terms of direction indication or position relationship (for example, up, down, left, right, front, back, inner, outer, top, bottom, center, vertical, horizontal, longitudinal, transverse, length, width, counterclockwise, clockwise, axial, radial, circumferential, etc.), such terms are only used to explain the relative position relationship, motion condition, etc. between components in a certain specific posture (as shown in the drawings); if the specific posture changes, the direction indication or position relationship will also change accordingly. In addition, the terms "first", "second", etc. in the application embodiments are only used for convenience of description, and cannot be understood as indicating or implying relative importance.

[0049] The application provides an automatic inspection method, a system and a computer readable storage medium. The application will be described in detail below with reference to the drawings. The features in the following embodiments and implementation manners can be combined with each other.

[0050] In the traditional inspection operation, manual inspection or single-device automatic inspection (such as using an inspection vehicle) is usually adopted. However, the coverage of vehicle inspection is limited, and there is a visual blind area (such as a high place or a shielding area) for a camera, which undoubtedly reduces the comprehensiveness and accuracy of the inspection.

[0051] The flexibility and wide angle of view of a UAV can be used as an inspection method. However, the endurance of the UAV is limited, the take-off and landing of the UAV are limited by the environment, and the coordination ability of the UAV with ground equipment is weak. Therefore, how to realize the coordinated automatic inspection of the vehicle and the UAV in a complex environment is a difficult problem.

[0052] To solve the above technical problems, the present specification provides an automatic inspection method.

[0053] The present specification aims to divide the actual inspection task into a task of coordination of the vehicle and the UAV, and to associate and bind the take-off and landing points of the UAV and the driving path of the vehicle in time and space, so that the vehicle can automatically arrive at and support the take-off and landing of the UAV during driving according to the planned path, thereby jointly completing a global inspection task including a ground segment and an air segment. In this way, not only can the inspection of the area that cannot be covered by the vehicle inspection be realized by the UAV, but also the vehicle provides accurate take-off and landing point positions for the UAV, guarantees the efficient and long-time work of the UAV, realizes the task coordination of the vehicle and the UAV, and realizes efficient, full-coverage and automatic inspection.

[0054] The application provides an embodiment of an automatic inspection method, referring to Figure 1 , Figure 1 is a flowchart of an automatic inspection method provided by the embodiment of the present specification.

[0055] It should be noted that the automatic inspection scheme in this paper is applied to the inspection planning unit, which can be, but is not limited to, any device or system among terminals, vehicles, independent servers, server clusters, cloud computing, vehicle network platforms, IoT platforms, etc.

[0056] Specifically, this includes the following steps 101 to 104:

[0057] In step 101, the inspection task parameters are obtained, including the inspection area.

[0058] This paper describes a two-stage process for achieving task collaboration between vehicles and drones: a background preparation stage and a task execution stage.

[0059] During the background preparation phase, the inspection planning unit plans the inspection route based on the inspection task and its parameters.

[0060] First, the user determines the inspection task parameters on the platform where the inspection planning unit is located.

[0061] As an example, the inspection task parameters include, but are not limited to, the start point, end point, inspection time period, and inspection area of ​​the inspection task.

[0062] During the route planning process, the inspection time period is used to obtain historical road condition data for the same time period and to predict road congestion during vehicle inspection.

[0063] In step 102, at least one candidate driving path for the inspection vehicle is generated based on the inspection area.

[0064] Based on the inspection task parameters, at least one candidate driving path is planned for the inspection vehicle, covering the inspection area and connecting the starting point and the terminal. The candidate driving path serves as the basic driving route and provides a basis for subsequent path optimization.

[0065] In step 103, in response to determining that there is at least one inspection blind spot on the candidate driving path formed by the limited field of view of the vehicle camera, a drone inspection sub-task is planned for the inspection blind spot, and the drone inspection sub-task is associated with the drone take-off and landing point.

[0066] For areas that cannot be covered by the vehicle's camera in terms of focal length and angle when the vehicle is performing an inspection task on a candidate driving path, a drone inspection sub-task is inserted, and the drone automatically takes over to complete the inspection task.

[0067] Using a 3D city model or geographic information system, the theoretical field of view of the vehicle's camera along the route is automatically analyzed. Any key areas outside the field of view are marked as blind spots. For each blind spot, an independent drone inspection sub-task is generated. For example... Figure 2 As shown,Figure 2 Area A is the vehicle inspection area, and Area B is the inspection blind spot, which is the area where the drone inspection sub-task is performed.

[0068] In some embodiments, multiple drone take-off and landing points for the drone to take off and land in each of the inspection blind spots are determined, and multiple candidate inspection routes containing the drone take-off and landing points are generated corresponding to the drone inspection sub-task.

[0069] Multiple candidate inspection routes are combined to obtain multiple initial inspection routes.

[0070] When planning drone inspection routes, the planning materials within the scope of the inspection task are retrieved, including but not limited to satellite images, architectural drawings, and 3D models.

[0071] Using high-precision satellite images and building outline maps, obstacles (such as trees and tall buildings) are identified, and open, flat areas without high-voltage lines or tree obstructions are searched around the blind spots of the inspection to determine multiple candidate take-off and landing points for unobstructed drones.

[0072] Based on a 3D model of the region (including but not limited to information such as building height and tree height), a flight path is planned that can completely cover the inspection blind spots and avoid all obstacles; this is the candidate inspection route. The flight altitude and inspection shooting angle are precisely set to ensure that the acquired images are clear and complete.

[0073] In some embodiments, if there are multiple drone take-off and landing points identified in the inspection blind zone, there will also be multiple candidate inspection routes that include both take-off and landing points. Subsequently, by combining these with the candidate travel paths of the inspection vehicle, an optimal inspection plan is determined. The candidate inspection route corresponding to the optimal inspection plan serves as the optimal path for the drone to perform its inspection sub-task in the current inspection blind zone.

[0074] In some embodiments, if there are multiple blind spots and multiple corresponding drone inspection sub-tasks, then the various drone inspection sub-tasks are associated to form a drone inspection task.

[0075] In this process, candidate inspection routes for each UAV inspection sub-task are combined to obtain multiple initial inspection routes for the UAV inspection task. Subsequently, the optimal inspection plan is determined by combining the candidate travel paths of the inspection vehicle.

[0076] In step 104, the location of the UAV take-off and landing point is associated with the candidate driving path to generate the target driving path of the inspection vehicle. The target driving path passes through the UAV take-off and landing point, so that when the inspection vehicle travels along the target driving path to the UAV take-off and landing point, the UAV inspection sub-task is triggered.

[0077] At this point, the inspection task has one main task (vehicle inspection task) and multiple branch tasks (drone inspection sub-tasks). The branch tasks are inserted into the timeline of the main task, and the optimal inspection plan is determined based on the optimization objective. This means that multiple vehicle inspection routes and drone inspection routes are combined and integrated into a single unmanned inspection task execution plan.

[0078] In some embodiments, the optimal inspection scheme is determined based on actual needs, and may be, but is not limited to, based on the shortest task execution time.

[0079] As an example, the step of associating the location of the UAV take-off and landing point with the candidate driving path to generate the target driving path of the inspection vehicle includes:

[0080] The candidate driving path and the candidate inspection route of the drone inspection sub-task are combined to generate multiple candidate task execution schemes for performing the inspection task. The candidate inspection route is determined based on the location of multiple drone take-off and landing points in the inspection blind zone for the drone to take off and land.

[0081] Determine the total time required to execute each of the candidate task execution schemes;

[0082] The candidate task execution plan with the shortest total time is selected as the target task execution plan, and the candidate driving path in the target task execution plan is the target driving path of the inspection vehicle.

[0083] Reference Figure 3 The inspection area includes area A, area B and area C, of ​​which area B is the drone inspection area.

[0084] Assume that the inspection vehicle has two inspection paths when performing an inspection task in area A: route A1 and route A2. The time taken to inspect area A based on route A1 is T. A1 The time taken to inspect area A based on route A2 is T. A2 The inspection vehicle has two routes when performing an inspection task in area B: route B1 and route B2. The time taken to inspect area B based on route B1 is T. B1 The time taken to inspect area B based on route B2 is T. B2 Both routes B1 and B2 allow for drone takeoffs and landings along their entire length. The inspection paths for the inspection vehicle performing its inspection mission in area C include routes C1 and C2. The time taken to inspect area C using route C1 is T. C1 The time taken to inspect area C based on route C2 is T. C2The starting points of routes C1 and C2 are located around the drone flight path, and drones can also take off or land at the starting points.

[0085] When a drone performs an inspection mission in area B, the available inspection paths include candidate inspection route B, and the required inspection time is T. B It should be noted that only one candidate inspection route for the UAV is set in the example shown in the illustration. In reality, there may be multiple routes, but the implementation examples are basically the same, and will not be described in detail here.

[0086] The inspection routes A1 and A2, B1 and B2, and C1 and C2 of the inspection vehicle are combined to generate multiple candidate driving routes from the starting point to the end point.

[0087] The candidate driving paths and the candidate inspection route B of the UAV are arranged and combined to obtain multiple candidate task execution schemes for performing inspection tasks.

[0088] Next, iterate through each candidate task execution plan to determine the total execution time required for each candidate task execution plan.

[0089] It should be noted that the inspection vehicle is performing its own inspection tasks while the drone is carrying out its own. Therefore, the total time refers to the time it takes for the inspection vehicle to travel from the starting point to the destination and receive the drone.

[0090] Finally, the candidate task execution plan with the shortest total time is selected as the final target task execution plan. The candidate travel path in the target task execution plan is the target travel path of the inspection vehicle.

[0091] In some embodiments, the task parameters include an inspection time period; the method further includes:

[0092] Obtain historical road condition information for the aforementioned inspection period;

[0093] During the planning of the candidate driving path and / or target driving path, the corresponding path and the UAV take-off and landing point associated with the path are adjusted based on the historical road condition information.

[0094] During the inspection route planning process, historical traffic information for the inspection period is obtained. If historical traffic information predicts that a certain section of road will be extremely congested, preventing vehicles from arriving at the designated drone landing point on time, the plan is dynamically adjusted. Of course, in other examples, congestion conditions on candidate driving routes can also be predicted based on real-time traffic information.

[0095] As an example, a longer flight path could be redesigned for the drone, increasing its flight time to match the vehicle's delayed arrival time. Another example is to relocate the drone to a more distant takeoff and landing point that is easier for vehicles to reach.

[0096] Reference Figure 4 Adjust the take-off and landing points of drones during area inspections, and accordingly... Figure 4 The candidate inspection route for the drone in (a) has been adjusted to Figure 4 The candidate inspection routes in (b) are then used to make a decision on the target mission execution plan.

[0097] In some embodiments, combining at least one candidate driving path and at least one candidate inspection route of the UAV inspection sub-task to generate multiple candidate task execution schemes for performing the inspection task includes:

[0098] At least one of the candidate driving paths and multiple initial inspection routes are combined to generate multiple candidate task execution schemes for performing the inspection task. Each candidate task execution scheme includes a candidate driving path and an associated initial inspection route.

[0099] Continuing with the aforementioned embodiments, if there are multiple blind spots requiring drone-assisted operations in an inspection task, there are corresponding multiple drone inspection sub-tasks, and each drone inspection sub-task corresponds to one or more candidate inspection routes. The multiple candidate inspection routes for each inspection sub-task are combined to obtain multiple initial inspection routes for executing the inspection tasks in chronological order. For example, referring to... Figure 3 Area A is also a blind spot for inspection. There are multiple candidate inspection routes in both Area A and Area B, which are determined based on multiple UAV take-off and landing points. By combining the candidate inspection routes in the two areas, multiple initial inspection routes for UAVs to inspect from Area A to Area B are obtained.

[0100] Next, the candidate travel paths are combined with the initial inspection route of the UAV to obtain multiple candidate task execution schemes for performing the inspection task. Then, based on the candidate task execution schemes, the target task execution scheme is determined. This means that the multiple candidate inspection routes corresponding to the initial inspection route in the target task execution scheme are the target inspection routes for performing the UAV inspection sub-task.

[0101] Through the above embodiments, a planning algorithm was used to achieve global optimization of time cost and path resources. This fundamentally solves the visual blind spot problem of a single inspection platform, realizing integrated ground-air three-dimensional full-coverage inspection.

[0102] After completing the inspection route planning, the task execution phase begins.

[0103] In some embodiments, when the inspection vehicle travels along the target travel path to the UAV take-off and landing point, it triggers the execution of the UAV inspection sub-task, including:

[0104] When the inspection vehicle travels along the target driving path to the drone take-off and landing point, it acquires surrounding environmental information through the vehicle's camera.

[0105] If the surrounding environment is found to meet the safety conditions for the take-off and landing of the drone, the drone is controlled to take off and perform the drone inspection sub-task, or the drone is controlled to land.

[0106] The backend platform sets the target driving route for the inspection vehicles and the direction of vehicle inspection for each section. It also determines the fixed location of the inspection vehicles, the flight path of the drones, and the shooting direction of the cameras on the inspection vehicles.

[0107] Once the inspection mission is initiated, the inspection vehicle will patrol along the preset target route and record video along the way.

[0108] Centimeter-level high-precision positioning is achieved through the GPS / BeiDou (BDS) satellite positioning system, ensuring accurate location identification even at complex intersections. When the inspection vehicle reaches the aforementioned designated take-off and landing point of the drone, satellite positioning is triggered, and the vehicle automatically comes to a stop.

[0109] The vehicle's camera acquires information about the surrounding environment to ensure that the environment for drone take-off and landing is safe.

[0110] As an example, the safety conditions include a situation where the surrounding environment is free of obstacles; acquiring surrounding environment information through the vehicle's camera includes:

[0111] Adjust the vehicle's camera to face the take-off and landing area of ​​the drone on the inspection vehicle and collect environmental images;

[0112] Obstacle detection is performed on the environmental image to determine whether there are obstacles in the surrounding environment.

[0113] After the inspection vehicle comes to a complete stop, its camera turns towards the drone's takeoff and landing area (including but not limited to the drone's landing pad and top position) to capture video. The video is aggregated in the backend, where an algorithm is used for obstacle detection. If no obstacles are detected, the drone takes off to perform its mission. If an obstacle is detected, an alarm is triggered, and the backend will alert the user for remote intervention (such as manual takeoff via remote control) or on-site manual intervention. This significantly improves system safety and prevents drone collisions with obstacles.

[0114] After the drone takes off, it sends a "mission started" or "takeoff completed" signal to the vehicle for synchronization. Upon receiving the signal, the vehicle continues its patrol along the target path until it reaches the next mission point, which is the landing location where the drone mission ends.

[0115] After completing its mission, the drone flies to the designated landing point. The vehicle, like other vehicles, undergoes a process of positioning, visual verification, and AI recognition to ensure the drone's safe recovery. Specifically, before the drone lands, the vehicle, after coming to a complete stop, directs its inspection camera towards the drone's landing site and top to capture video. The video is then compiled into a backend system, where algorithms are used for obstacle detection. If no obstacles are detected, the drone lands. If obstacles are detected, the backend system will alert staff for remote or on-site manual intervention.

[0116] The above embodiments maximize the advantages of both vehicles and drones, achieving full automation from task planning to execution and recovery, greatly reducing human intervention and improving overall inspection efficiency.

[0117] The automatic inspection method, system, and computer-readable storage medium presented in this paper generate at least one candidate driving path for the inspection vehicle based on inspection task parameters. Based on the candidate driving paths, blind spots in the inspection process are identified, and these blind spots are automatically inspected by a drone, enabling collaborative inspection operations between the vehicle and the drone. A drone inspection sub-task is planned for the blind spot, associating the drone's take-off and landing point with the candidate driving path to generate the target driving path for the inspection vehicle, ensuring global optimization and path safety throughout the entire inspection process. Simultaneously, the target driving path passes through the drone's take-off and landing point, triggering the execution of the drone inspection sub-task when the inspection vehicle reaches the drone's take-off and landing point along the target driving path. This achieves seamless collaborative operation between the inspection vehicle and the drone in complex environments, reducing human intervention and improving inspection efficiency and safety.

[0118] Based on the same application concept as the above method, this application also proposes a motor control system.

[0119] The automated inspection system includes an inspection planning unit, inspection vehicles, and drones.

[0120] An inspection planning unit is used to implement the automatic inspection method as described in any of the preceding items;

[0121] The inspection vehicle is used to respond to the inspection task issued by the inspection planning unit. The task parameters include the target driving path. The vehicle performs the inspection task based on the target driving path and triggers the UAV to synchronously execute the UAV inspection sub-task when passing the UAV take-off and landing point.

[0122] The drone is used to respond to the inspection tasks issued by the inspection planning unit, the task parameters of which include the target inspection route; and to perform the inspection tasks based on the target inspection route.

[0123] The implementation process of the functions and roles of each unit in the above system is detailed in the implementation process of the corresponding steps in the above method, which can achieve the same technical effect, and will not be repeated here.

[0124] This application also provides an inspection planning unit, which is used to implement the automatic inspection method described above.

[0125] Figure 5 An example is a schematic diagram of the physical structure of an automatic inspection device, such as... Figure 5 As shown, the automatic inspection device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840. The processor 810, communication interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions from the memory 830 to execute the automatic inspection method.

[0126] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0127] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the automatic inspection method provided by the above methods.

[0128] In another aspect, this application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the automatic inspection methods provided by the above methods.

[0129] It should be noted that the technical solutions or features described in the above embodiments can be combined or supplemented with each other without conflict. The scope of protection of this application is not limited to the precise structures described in the above embodiments and shown in the accompanying drawings; all modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An automatic inspection method, characterized in that, The method includes: Obtain inspection task parameters, including the inspection area; Based on the inspection area, at least one candidate driving path for the inspection vehicle is generated. In response to determining that there is at least one inspection blind spot on the candidate driving path formed by the limited field of view of the vehicle camera, a drone inspection sub-task is planned for the inspection blind spot, and the drone inspection sub-task is associated with the drone take-off and landing point; The location of the UAV take-off and landing point is associated with the candidate driving path to generate the target driving path of the inspection vehicle. The target driving path passes through the UAV take-off and landing point, so that when the inspection vehicle travels along the target driving path to the UAV take-off and landing point, the UAV inspection sub-task is triggered.

2. The automatic inspection method as described in claim 1, characterized in that, The step of associating the location of the UAV take-off and landing point with the candidate driving path to generate the target driving path for the inspection vehicle includes: The candidate driving path and the candidate inspection route of the drone inspection sub-task are combined to generate multiple candidate task execution schemes for performing the inspection task. The candidate inspection route is determined based on the location of multiple drone take-off and landing points in the inspection blind zone for the drone to take off and land. Determine the total time required to execute each of the candidate task execution schemes; The candidate task execution plan with the shortest total time is selected as the target task execution plan, and the candidate driving path in the target task execution plan is the target driving path of the inspection vehicle.

3. The automatic inspection method as described in claim 2, characterized in that, The method further includes: Determine multiple drone take-off and landing points in each of the inspection blind spots for the drone to take off and land, and generate multiple candidate inspection routes corresponding to the drone inspection sub-task that include the drone take-off and landing points; Multiple candidate inspection routes are combined to obtain multiple initial inspection routes; The step of combining at least one of the candidate driving paths and at least one of the candidate inspection routes of the UAV inspection sub-task to generate multiple candidate task execution schemes for performing the inspection task includes: At least one of the candidate driving paths and multiple initial inspection routes are combined to generate multiple candidate task execution schemes for performing the inspection task. Each candidate task execution scheme includes a candidate driving path and an associated initial inspection route.

4. The automatic inspection method as described in claim 3, characterized in that, The method further includes: Multiple candidate inspection routes corresponding to the initial inspection route in the target task execution plan are determined as the target inspection routes for executing the UAV inspection sub-task.

5. The automatic inspection method as described in claim 1, characterized in that, The task parameters include the inspection time period; the method further includes: Obtain historical road condition information for the aforementioned inspection period; During the planning of the candidate driving path and / or target driving path, the corresponding path and the UAV take-off and landing point associated with the path are adjusted based on the historical road condition information.

6. The automatic inspection method as described in claim 1, characterized in that, When the inspection vehicle travels along the target travel path to the drone take-off and landing point, it triggers the execution of the drone inspection sub-task, including: When the inspection vehicle travels along the target driving path to the drone take-off and landing point, it acquires surrounding environmental information through the vehicle's camera. If the surrounding environment is found to meet the safety conditions for the take-off and landing of the drone, the drone is controlled to take off and perform the drone inspection sub-task, or the drone is controlled to land.

7. The automatic inspection method as described in claim 6, characterized in that, The safety conditions include the absence of obstacles in the surrounding environment; Information about the surrounding environment is acquired through the vehicle's camera, including: Adjust the vehicle's camera to face the take-off and landing area of ​​the drone on the inspection vehicle and collect environmental images; Obstacle detection is performed on the environmental image to determine whether there are obstacles in the surrounding environment.

8. An automatic inspection system, characterized in that, The system includes: An inspection planning unit is used to implement the automatic inspection method as described in any one of claims 1 to 7; The inspection vehicle is used to respond to the inspection task issued by the inspection planning unit. The task parameters include the target driving path. The vehicle performs the inspection task based on the target driving path and triggers the UAV to synchronously execute the UAV inspection sub-task when passing the UAV take-off and landing point. The drone is used to respond to the inspection tasks issued by the inspection planning unit, the task parameters of which include the target inspection route; and to perform the inspection tasks based on the target inspection route.

9. An inspection planning unit, characterized in that, Used to implement the automatic inspection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an automatic inspection program, which, when executed, implements the steps of the automatic inspection method as described in any one of claims 1-7.

Images