Inspection route determination method and device and electronic equipment

Abstract

The invention discloses a method and a device for determining an inspection route and electronic equipment. The method comprises the following steps: receiving an inspection request of a target inspection area, the inspection request being a request for inspecting a plurality of inspection objects; in response to the inspection request, marking target frames respectively corresponding to the plurality of inspection objects in a three-dimensional space corresponding to the target inspection area; normal vectors of the target frames corresponding to the inspection objects are determined, and the normal vectors are perpendicular to the operation planes of the corresponding inspection objects; determining projection coordinates respectively corresponding to the plurality of normal vectors in the three-dimensional space; according to the plurality of projection coordinates, determining parking parameters respectively corresponding to the plurality of inspection objects; and determining a target inspection route according to the parking parameters corresponding to the plurality of inspection objects. The technical problems that the efficiency of determining the inspection route is low and the inspection requirement cannot be met due to the fact that the stop point and the inspection point are manually preset in the related technology are solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular, to a method, apparatus, and electronic device for determining an inspection route. Background Art

[0002] In industrial scenarios such as substations and petrochemical plants, inspection equipment plays an important role in safety inspections, maintenance monitoring, and environmental monitoring. Currently, the determination of the inspection route of inspection equipment mainly relies on manually presetting stopping points and inspection points. This method requires a large amount of manual calibration work to be completed in advance, and the efficiency of determining the inspection route is low, unable to meet the inspection requirements.

[0003] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] Embodiments of the present invention provide a method, apparatus, and electronic device for determining an inspection route, so as to at least solve the technical problem in the related art that manually presetting stopping points and inspection points results in low efficiency of determining the inspection route and inability to meet the inspection requirements.

[0005] According to one aspect of the embodiments of the present invention, there is provided a method for determining an inspection route, including: receiving an inspection request for a target inspection area, where the inspection request is a request for inspecting a plurality of inspection objects; in response to the inspection request, marking target boxes corresponding to the plurality of inspection objects in a three-dimensional space corresponding to the target inspection area; determining normal vectors of the target boxes corresponding to the plurality of inspection objects respectively, where the normal vector is perpendicular to the operation plane of the corresponding inspection object; determining projection coordinates corresponding to the plurality of normal vectors in the three-dimensional space respectively; determining docking parameters corresponding to the plurality of inspection objects respectively according to the plurality of projection coordinates; and determining a target inspection route according to the docking parameters corresponding to the plurality of inspection objects respectively.

[0006] Optionally, determining a target inspection route according to the docking parameters corresponding to the plurality of inspection objects respectively includes: determining inspection object coordinates corresponding to the plurality of inspection objects respectively according to the target boxes corresponding to the plurality of inspection objects respectively; determining an initial inspection route according to the inspection object coordinates corresponding to the plurality of inspection objects respectively; and determining the target inspection route according to the initial inspection route and the docking parameters corresponding to the plurality of inspection objects respectively.

[0007] Optionally, determining an initial inspection route according to the inspection object coordinates respectively corresponding to the multiple inspection objects includes: obtaining three-dimensional point cloud data corresponding to the target inspection area; determining a passable area corresponding to the target inspection area according to the three-dimensional point cloud data, where a vertical height of an obstacle voxel corresponding to the passable area is less than a height threshold, and the obstacle voxel is a voxel formed by data points representing obstacles; determining the initial inspection route according to the inspection object coordinates respectively corresponding to the multiple inspection objects and the passable area corresponding to the target inspection area.

[0008] Optionally, after determining the target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects, it further includes: when the target inspection route includes the target inspection positions respectively corresponding to the multiple inspection objects, controlling the inspection device to sequentially reach the target inspection positions respectively corresponding to the multiple inspection objects according to the target inspection route; determining adjustment parameters corresponding to the inspection device at the corresponding target inspection positions; determining target pose information corresponding to the inspection device when operating the corresponding inspection object according to the adjustment parameters of the inspection device at the corresponding target inspection positions; controlling the inspection device to operate the corresponding inspection object in the pose corresponding to the target pose information.

[0009] Optionally, determining the adjustment parameters corresponding to the inspection device includes: when the target inspection route includes the initial pose information corresponding to the inspection device when operating the corresponding inspection object, obtaining the inspection object coordinates respectively corresponding to the corresponding inspection object; determining the object projection coordinates corresponding to the corresponding inspection object according to the inspection object coordinates, and determining the device projection coordinates corresponding to the inspection device when operating the corresponding inspection object according to the initial pose information; determining the projection coordinate difference corresponding to the corresponding inspection object according to the object projection coordinates and the device projection coordinates; determining the adjustment parameters according to the projection coordinate difference and the initial pose information.

[0010] Optionally, after controlling the inspection device to operate on the corresponding inspection object in the pose corresponding to the target pose information, it further includes: when the operation is a shooting operation, and controlling the inspection device to operate on the corresponding inspection object in the pose corresponding to the target pose information is to control the inspection device to shoot the corresponding inspection object in the pose corresponding to the target pose information to obtain a corresponding first captured image, acquiring a regional image corresponding to the target inspection area; determining first feature points of the corresponding inspection object according to the first captured image; determining second feature points of the corresponding inspection object according to the regional image; determining a corresponding feature point deviation parameter according to the first feature points and the second feature points; when the feature point deviation parameter is greater than a predetermined threshold, adjusting the target pose information when the inspection device operates on the corresponding inspection object according to the corresponding feature point deviation parameter to obtain updated pose information when the inspection device operates on the corresponding inspection object; controlling the inspection device to shoot the corresponding inspection object in the pose corresponding to the updated pose information.

[0011] Optionally, in response to the inspection request, marking target frames respectively corresponding to the multiple inspection objects in the three-dimensional space corresponding to the target inspection area includes: in response to the inspection request, retrieving three-dimensional point cloud data corresponding to the target inspection area; determining a three-dimensional space corresponding to the target inspection area according to the three-dimensional point cloud data; marking target frames respectively corresponding to the multiple inspection objects in the three-dimensional space corresponding to the target inspection area.

[0012] In an embodiment of the present invention, a patrol request for a target patrol area is received, where the patrol request is a request for patrolling multiple patrol objects; in response to the patrol request, target boxes corresponding to the multiple patrol objects in a three-dimensional space corresponding to the target patrol area are marked; normal vectors of the target boxes corresponding to the multiple patrol objects are determined, where the normal vector is perpendicular to the operation plane of the corresponding patrol object; projection coordinates corresponding to the multiple normal vectors in the three-dimensional space are determined; docking parameters corresponding to the multiple patrol objects are determined based on the multiple projection coordinates; and a target patrol route is determined based on the docking parameters corresponding to the multiple patrol objects. By determining the normal vectors of the target boxes corresponding to the multiple patrol objects, the purpose of determining the docking parameters corresponding to the multiple patrol objects based on the projection coordinates corresponding to the multiple normal vectors and determining the target patrol route based on the docking parameters corresponding to the multiple patrol objects is achieved. Since the normal vectors of the target boxes corresponding to the multiple patrol objects can reflect the positions corresponding to the multiple patrol objects and the operation plane directions of the patrol objects, the docking parameters corresponding to the patrol objects can be determined based on the projections of the normal vectors, and the target patrol route of the patrol device can be determined, thereby solving the technical problem in the related art that manual presetting of docking points and inspection points results in low efficiency in determining the patrol route and inability to meet the inspection requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0014] Figure 1 is a flowchart of a method for determining a patrol route according to an embodiment of the present invention;

[0015] Figure 2 is a flowchart of a robot vision target autonomous positioning patrol method based on a point cloud panoramic view provided by an alternative embodiment of the present invention;

[0016] Figure 3 is a schematic diagram of a robot vision target autonomous positioning patrol based on a point cloud panoramic view provided by an alternative embodiment of the present invention;

[0017] Figure 4 is a structural block diagram of a device for determining a patrol route according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solution in the embodiments of the present invention 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. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0019] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned accompanying 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 here can be implemented in an order different from those illustrated or described here. 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 comprising 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.

[0020] Embodiment 1

[0021] According to an embodiment of the present invention, an embodiment of a method for determining an inspection route is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0022] Figure 1 is a flowchart of the method for determining an inspection route according to an embodiment of the present invention. As Figure 1 shown, the method includes the following steps:

[0023] Step S102, receiving an inspection request for a target inspection area, where the inspection request is a request for inspecting multiple inspection objects.

[0024] In step S102 provided in the present application, an inspection request for a target inspection area is received.

[0025] Among them, the target inspection area is involved. The target inspection area refers to a specific geographical or physical space area that needs to be inspected. For example, environments such as substations, power lines, industrial facilities, warehouses, petrochemical plants, etc. that need to be regularly inspected or monitored.

[0026] Among them, there is an inspection request. An inspection request refers to an instruction issued by an operator or a monitoring system, instructing the inspection device to inspect a specified inspection area. The request usually contains specific operation details, such as the type, quantity, location information of the inspection object, and the priority of the inspection.

[0027] Among them, there is an inspection object. An inspection object refers to the target object device that needs to be monitored or inspected within the target inspection area. For example, power equipment, building structures, mechanical equipment, stored items, etc.

[0028] In this step, an inspection request is received from an operator or a monitoring system. The inspection request requires the inspection device to perform inspection tasks on multiple predefined inspection objects, that is, multiple inspection objects, within a specific geographical area, namely the target inspection area. By receiving and processing inspection requests in real time, the inspection device can quickly respond to emergency or specific inspection requirements and provide immediate feedback on inspection results. At the same time, using the inspection device for automated inspection improves the safety of the inspection operation.

[0029] Step S104, in response to the inspection request, mark the target boxes corresponding to multiple inspection objects in the three-dimensional space corresponding to the target inspection area respectively.

[0030] In step S104 provided by this application, in response to the inspection request, the target boxes corresponding to multiple inspection objects are marked in the three-dimensional space corresponding to the target inspection area respectively.

[0031] Among them, there is a three-dimensional space. The three-dimensional space refers to a three-dimensional space model of the target inspection area established based on the three-dimensional point cloud data of the target inspection area.

[0032] Among them, there is a target box. A target box refers to a virtual bounding box that determines the position and shape of an inspection object, used to define the spatial position and range of the object. The target box can be a three-dimensional rectangle or other shapes, depending on the actual geometric characteristics of the object.

[0033] In this step, after receiving the inspection request, first determine the inspection objects that need to be inspected according to the content of the inspection request. Then, in the three-dimensional space model of the target inspection area, mark each object or location that needs to be inspected through vision or positioning technology, and enclose their spatial positions and ranges in the form of a three-dimensional bounding box.

[0034] Through this step, autonomously identify and locate the target in the three-dimensional space, mark the inspection object with a target box, which provides a basis for determining the corresponding docking points of the inspection object in the subsequent process. At the same time, using the target box to mark the inspection object helps the inspection device plan the optimal inspection path, avoid ineffective movement and repeated inspections, thereby improving the inspection efficiency and the automation level of the inspection task.

[0035] It should be noted that the target box can include the attribute information of the corresponding inspection object, such as the name of the inspection object, the last inspection time, etc. In addition, for dynamic inspection objects, such as moving personnel or vehicles, during the inspection process, the position of the target box can be updated in real time to cope with environmental changes or target movement, ensuring the safety of the inspection equipment during inspection while maintaining the accuracy and timeliness of the inspection.

[0036] Step S106: Determine the normal vectors of the target boxes corresponding to multiple inspection objects respectively, where the normal vector is perpendicular to the operation plane of the corresponding inspection object.

[0037] In step S106 provided in this application, the normal vectors of the target boxes corresponding to multiple inspection objects are determined respectively, where the normal vector is perpendicular to the operation plane of the corresponding inspection object.

[0038] Among them, the normal vector is involved. The normal vector refers to the unit vector perpendicular to the operation plane of the target box, which is used to indicate the direction of the operation plane of the target box.

[0039] Among them, the operation plane is involved. The operation plane refers to the specific plane on the inspection object for performing inspections or operations. For example, when inspecting the panel of an electrical equipment, the operation plane is the plane where the panel is located.

[0040] In this step, first determine the operation planes corresponding to multiple inspection objects respectively. Then, determine the normal vectors corresponding to multiple target boxes respectively, and ensure that this normal vector is perpendicular to the operation plane of the corresponding inspection object. Through this step, the normal vector perpendicular to the operation plane of the inspection object is determined, providing a basis for the subsequent inspection equipment to accurately align with the operation plane during the inspection operation of the inspection object, thereby improving the accuracy and efficiency of the inspection.

[0041] It should be noted that during the inspection process, if the inspection object or the environment changes, the normal vector of the target box can be updated in real time to adapt to the new situation and maintain the accuracy of data collection.

[0042] Step S108: Determine the projection coordinates corresponding to the multiple normal vectors in the three-dimensional space respectively.

[0043] In step S108 provided in this application, the projection coordinates corresponding to the multiple normal vectors in the three-dimensional space are determined respectively.

[0044] Among them, the projection coordinates are involved. The projection coordinates refer to the coordinates of the projection points where the normal vectors in the three-dimensional space are projected onto the ground, which are used to help the inspection equipment determine its target position and direction on the two-dimensional map.

[0045] In this step, the coordinate values of the projection points of the normal vectors corresponding to multiple inspection objects on the ground are determined. Since the inspection device usually navigates and locates based on a two-dimensional map, by determining the projection coordinates of the normal vectors, the inspection device can accurately locate the positions of the inspection objects and the operation plane directions on the two-dimensional map.

[0046] It should be noted that the conversion from the three-dimensional spatial position of the normal vector to the projection coordinates on the two-dimensional map can be completed through mathematical transformations, such as projection matrices, to ensure the accuracy and reliability of the conversion.

[0047] Step S110: Determine the docking parameters corresponding to multiple inspection objects according to multiple projection coordinates.

[0048] In step S110 provided in this application, the docking parameters corresponding to multiple inspection objects are determined according to multiple projection coordinates.

[0049] Among them, docking parameters are involved. Docking parameters refer to the parameter information of the position and attitude where the inspection device needs to dock when inspecting the inspection objects during the inspection task of the inspection device.

[0050] In this step, after determining the projection coordinates of each inspection object on the ground, the docking parameters corresponding to the inspection device when inspecting each object are determined based on the projection coordinate information. The docking parameters can determine where the inspection device should stop, that is, the docking point coordinates, and the attitude corresponding to the inspection device when performing inspection operations on the inspection objects.

[0051] Step S112: Determine the target inspection route according to the docking parameters corresponding to multiple inspection objects.

[0052] In step S112 provided in this application, the target inspection route is determined according to the docking parameters corresponding to multiple inspection objects.

[0053] Among them, the target inspection route is involved. The target inspection route refers to the path or trajectory that the inspection device needs to follow during the inspection process. The target inspection route is formed by connecting multiple docking points, and each docking point corresponding to an inspection object is determined according to the corresponding docking parameters.

[0054] In this step, first, the docking points corresponding to multiple inspection objects are determined according to multiple docking parameters to ensure that at each docking point, the inspection device can accurately face the operation plane of the inspection object. Then, based on multiple docking points, the optimal path for the inspection device to perform the entire inspection task is planned to determine the target inspection path. The entire route planning process is automated, reducing the need for manual intervention and improving the intelligence and automation level of the inspection task.

[0055] It should be noted that the inspection route can be adjusted according to real-time docking parameters and the status of the inspection objects, enabling the inspection equipment to adapt to emergencies such as target movement and environmental changes. At the same time, according to the urgency and importance of the inspection objects, a higher-priority inspection route can be planned for the inspection equipment to ensure that key facilities and problems are inspected in a timely manner. In addition, in large-scale inspection tasks, the inspection routes of multiple inspection equipment can be coordinated to avoid collisions between the inspection equipment and improve the overall inspection efficiency and coverage rate.

[0056] Through the above steps S102 - S112, receive the inspection request for the target inspection area, where the inspection request is a request to inspect multiple inspection objects; in response to the inspection request, mark the target boxes corresponding to the multiple inspection objects in the three-dimensional space corresponding to the target inspection area respectively; determine the normal vectors of the target boxes corresponding to the multiple inspection objects respectively, where the normal vector is perpendicular to the operation plane of the corresponding inspection object; determine the projection coordinates of the multiple normal vectors in the three-dimensional space respectively; determine the docking parameters corresponding to the multiple inspection objects respectively based on the multiple projection coordinates; determine the target inspection route based on the docking parameters corresponding to the multiple inspection objects respectively. By determining the normal vectors of the target boxes corresponding to the multiple inspection objects respectively, the purpose of determining the docking parameters corresponding to the multiple inspection objects respectively based on the projection coordinates corresponding to the multiple normal vectors and determining the target inspection route based on the docking parameters corresponding to the multiple inspection objects respectively is achieved. Since the normal vectors of the target boxes corresponding to the multiple inspection objects can reflect the positions corresponding to the multiple inspection objects and the operation plane directions of the inspection objects, the docking parameters corresponding to the inspection objects can be determined based on the projections of the normal vectors, and the target inspection route of the inspection equipment can be determined, thus solving the technical problem in the related art that the efficiency of determining the inspection route is low due to the artificial presetting of docking points and inspection points and cannot meet the inspection requirements.

[0057] As an optional embodiment, determining the target inspection route based on the docking parameters corresponding to the multiple inspection objects respectively includes: determining the inspection object coordinates corresponding to the multiple inspection objects respectively based on the target boxes corresponding to the multiple inspection objects respectively; determining the initial inspection route based on the inspection object coordinates corresponding to the multiple inspection objects respectively; determining the target inspection route based on the initial inspection route and the docking parameters corresponding to the multiple inspection objects respectively.

[0058] In this embodiment, the specific steps of determining the target inspection route based on the docking parameters corresponding to the multiple inspection objects respectively are described.

[0059] Among them, the inspection object coordinates are involved. The inspection object coordinates refer to the precise position coordinates representing the inspection object in the three-dimensional space extracted from the target box.

[0060] Among them, the initial inspection route is involved. The initial inspection route refers to the inspection path of the inspection equipment directly planned based on the coordinates of the inspection objects.

[0061] In this step, first, according to the bounding boxes of each inspection object, their precise coordinates in the three-dimensional space, that is, the coordinates of the inspection objects, are extracted. Then, based on the coordinates of the inspection objects, a preliminary inspection route, that is, the initial inspection route, is determined. This route only considers the positions of the inspection objects and does not include the information about the docking points of the inspection equipment when inspecting the inspection objects. Finally, the initial route is combined with the docking parameters of each inspection object to optimize the initial inspection route and determine the final target inspection route. Through this step, by combining the docking parameters and the coordinates of the inspection objects, the shortest target inspection route is planned, saving inspection time and energy and improving inspection efficiency.

[0062] It should be noted that when planning the route, the energy consumption of the inspection equipment can be considered, and the path with the lowest energy consumption can be selected to extend the continuous operation time of the inspection equipment and improve the energy utilization efficiency.

[0063] As an optional embodiment, to determine the initial inspection route according to the coordinates of the inspection objects corresponding to multiple inspection objects, it includes: obtaining the three-dimensional point cloud data corresponding to the target inspection area; according to the three-dimensional point cloud data, determining the passable area corresponding to the target inspection area, where the vertical height of the obstacle voxels corresponding to the passable area is less than the height threshold, and the obstacle voxels are voxels composed of data points representing obstacles; determining the initial inspection route according to the coordinates of the inspection objects corresponding to multiple inspection objects and the passable area corresponding to the target inspection area.

[0064] In this embodiment, the specific steps of determining the initial inspection route according to the coordinates of the inspection objects corresponding to multiple inspection objects are described.

[0065] Among them, the three-dimensional point cloud data is involved. The three-dimensional point cloud data refers to a set composed of a large number of discrete points in the three-dimensional space. These points usually contain position information and attribute information, such as color, reflectivity, etc. In the inspection scenario, the three-dimensional point cloud data of the target inspection area is collected through sensors such as lidar and stereo vision cameras to construct a three-dimensional model of the target inspection area.

[0066] Among them, the passable area is involved. The passable area refers to the area where the inspection equipment can pass safely.

[0067] Among them, the obstacle voxels are involved. The obstacle voxels refer to the voxels composed of the three-dimensional point cloud data corresponding to the objects existing in the target inspection area that affect the passage of the inspection equipment.

[0068] Among them, the height threshold is involved. The height threshold refers to a preset height value used to determine the passage area within the target inspection area.

[0069] In this step, the 3D point cloud data of the target inspection area is first acquired. Then, obstacle pixels within the target inspection area are identified based on this 3D point cloud data. The vertical heights of these obstacle pixels are compared with a height threshold. The areas corresponding to obstacles below the height threshold are designated as areas where the inspection equipment can move freely, known as the pass zone. Finally, combining the coordinates of the inspection objects and the pass zone information, a preliminary route, known as the initial inspection route, is planned, connecting all inspection objects and avoiding obstacles. This ensures a safe and smooth inspection process for the inspection equipment.

[0070] Through this step, the passage area in the target inspection area is determined, so that the inspection equipment can effectively identify and avoid obstacles during the inspection process, ensuring the safe passage of the inspection equipment in complex environments.

[0071] It's important to note that when determining the initial inspection route, the system updates 3D point cloud data in real time, dynamically identifying traffic zones and obstacles and adapting to environmental changes such as new facilities and moving obstacles. By integrating safety algorithms and rules, the system's planned routes avoid potential danger zones and operational restrictions, ensuring the safety of inspection equipment and personnel.

[0072] As an optional embodiment, after determining the target inspection route based on the docking parameters corresponding to multiple inspection objects, it also includes: when the target inspection route includes target inspection positions corresponding to multiple inspection objects, controlling the inspection equipment to arrive at the target inspection positions corresponding to the multiple inspection objects in sequence according to the target inspection route; determining the adjustment parameters corresponding to the inspection equipment at the corresponding target inspection positions; determining the corresponding target posture information when the inspection equipment operates the corresponding inspection object based on the adjustment parameters of the inspection equipment at the corresponding target inspection position; and controlling the inspection equipment to operate the corresponding inspection object with the posture corresponding to the target posture information.

[0073] In this embodiment, specific steps are described for controlling the inspection device to operate the corresponding inspection object in a posture corresponding to the target posture information when the target inspection route includes target inspection positions corresponding to multiple inspection objects.

[0074] Among them, inspection equipment is involved. Inspection equipment refers to automated or semi-automated equipment that performs inspection tasks, such as robots, drones, pan-tilt cameras, etc., which are used to collect status information of inspection objects.

[0075] Among them, the target inspection position is involved. The target inspection position refers to the point that the inspection equipment needs to reach accurately when inspecting the corresponding inspection object.

[0076] Among them, adjustment parameters are involved. The adjustment parameters refer to the angle and position information that the inspection device needs to adjust when reaching the target inspection position, so as to ensure that the inspection device can perform inspection operations on the inspection object.

[0077] Among them, target pose information is involved. The target pose information refers to the precise position and pose information of the inspection device when performing specific inspection operations.

[0078] In this step, after determining the target inspection route, when the target inspection route includes target inspection positions corresponding to multiple inspection objects respectively, control the inspection device to sequentially reach the target inspection positions of each inspection object according to the target inspection route. Then, at each target position, determine the adjustment parameters of the inspection device. Based on the adjustment parameters, further determine the target pose information of the inspection device when operating each inspection object to ensure that the inspection device can face the corresponding inspection object. Finally, the inspection device will adjust its own position and pose according to the target pose information and perform the inspection operation on the inspection object.

[0079] Through this step, the target pose information corresponding to the inspection device when performing inspection operations on multiple inspection objects is determined. By controlling the inspection device to operate according to the target pose information, it is ensured that the device can accurately align with the target object, improving the accuracy of inspection and the quality of data.

[0080] As an optional embodiment, determining the adjustment parameters corresponding to the inspection device includes: when the target inspection route includes the initial pose information corresponding to the inspection device when operating the corresponding inspection object, obtaining the inspection object coordinates corresponding to the corresponding inspection objects respectively; according to the inspection object coordinates, determining the object projection coordinates corresponding to the corresponding inspection objects, and according to the initial pose information, determining the device projection coordinates corresponding to the inspection device when operating the corresponding inspection object; according to the object projection coordinates and the device projection coordinates, determining the projection coordinate difference corresponding to the corresponding inspection object; according to the projection coordinate difference and the initial pose information, determining the adjustment parameters.

[0081] In this embodiment, the specific steps of determining the adjustment parameters are described when the target inspection route includes the initial pose information corresponding to the inspection device when operating the corresponding inspection object.

[0082] Among them, initial pose information is involved. The initial pose information refers to the initial position and pose information of the inspection device after reaching the target inspection position and before performing inspection operations on the inspection object, including the three-dimensional coordinates, direction and rotation angle of the device.

[0083] Among them, the object projection coordinates are involved. The object projection coordinates refer to the projection coordinates of the inspection object coordinates in the plane coordinate system corresponding to the inspection device when the inspection device performs an inspection operation on the inspection object.

[0084] Among them, the device projection coordinates are involved. The device projection coordinates refer to the projection coordinates of the device sensor that specifically performs the inspection operation on the inspection device in the plane coordinate system corresponding to the inspection device.

[0085] Among them, the projection coordinate difference is involved. The projection coordinate difference refers to the difference between the object projection coordinates and the device projection coordinates.

[0086] In this step, first, obtain the three-dimensional space coordinates of the inspection object, that is, the inspection object coordinates. Using the inspection object coordinate information, through coordinate transformation and projection algorithms, determine the projection coordinates of each inspection object in the plane coordinate system of the inspection device, that is, the object projection coordinates. At the same time, based on the initial pose information of the device, determine the projection coordinates of the device sensor in the plane coordinate system of the inspection device, that is, the device projection coordinates. Then, compare the object projection coordinates with the device projection coordinates to determine the difference between the two, that is, the projection coordinate difference. Finally, based on the projection coordinate difference and the initial pose information of the device, determine the adjustment parameters of the device sensor to achieve the effect of accurately aligning with the target object.

[0087] Through this step, determine the difference between the device projection coordinates of the inspection device and the object projection coordinates of the inspection object. Based on the projection coordinates and the initial pose information, determine the adjustment parameters, which provides a basis for determining the target pose information according to the adjustment parameters later, ensuring that the device sensor can accurately align with the target object and improving the accuracy of inspection and the reliability of data collection.

[0088] As an optional embodiment, after controlling the inspection device to operate on the corresponding inspection object in the pose corresponding to the target pose information, it further includes: when the operation is a shooting operation, and controlling the inspection device to operate on the corresponding inspection object in the pose corresponding to the target pose information means controlling the inspection device to shoot the corresponding inspection object in the pose corresponding to the target pose information to obtain the corresponding first captured image, obtain the area image corresponding to the target inspection area; determine the first feature points of the corresponding inspection object according to the first captured image; determine the second feature points of the corresponding inspection object according to the area image; determine the corresponding feature point deviation parameter according to the first feature points and the second feature points; when the feature point deviation parameter is greater than the predetermined threshold, adjust the target pose information when the inspection device operates on the corresponding inspection object according to the corresponding feature point deviation parameter to obtain the updated pose information when the inspection device operates on the corresponding inspection object; control the inspection device to shoot the corresponding inspection object in the pose corresponding to the updated pose information.

[0089] In this embodiment, the specific steps of determining the updated pose information and controlling the inspection device to pose corresponding to the updated pose information to capture the corresponding inspection object are described when the operation is a shooting operation, the inspection device is controlled to pose corresponding to the target pose information, and the corresponding inspection object for the operation is to control the inspection device to pose corresponding to the target pose information to capture the corresponding inspection object, thereby obtaining the corresponding first captured image.

[0090] Among them, the first captured image is involved. The first captured image refers to the image obtained by capturing the inspection object after the inspection device is adjusted to the target pose information.

[0091] Among them, the area image is involved. The area image refers to the panoramic or comprehensive image of the target inspection area, which usually contains the visual information of multiple inspection objects and is used to provide a more comprehensive environmental background.

[0092] Among them, the first feature point is involved. The first feature point refers to the feature point corresponding to the inspection object extracted from the first captured image.

[0093] Among them, the second feature point is involved. The second feature point refers to the feature point corresponding to the inspection object extracted from the area image.

[0094] Among them, the feature point deviation parameter is involved. The feature point deviation parameter refers to the difference value between the first feature point and the second feature point, and is used to evaluate the position and pose accuracy of the inspection device during shooting.

[0095] Among them, the predetermined threshold is involved. The predetermined threshold refers to the preset value used to determine whether the feature point deviation parameter is within an acceptable range. If the feature point deviation parameter is greater than the predetermined threshold, it is considered that the shooting result is not accurate enough and needs to be further adjusted.

[0096] Among them, the updated pose information is involved. The updated pose information refers to the more accurate pose information adjusted according to the deviation parameter when the feature point deviation parameter exceeds the predetermined threshold, and is used to guide the inspection device to reposition and adjust the pose to improve the shooting accuracy.

[0097] In this step, first, the inspection device aligns with the inspection object according to the target pose information and takes a picture to obtain the first captured image. Then, the first feature points of the inspection object are identified and extracted from the first captured image, and at the same time, the corresponding second feature points are identified and extracted from the regional image. By comparing the positions of the first feature points and the second feature points, the feature point deviation parameter is determined to evaluate the position and attitude accuracy of the inspection device during shooting. If the feature point deviation parameter is greater than the predetermined threshold, it indicates that the shooting is not precise enough. The pose information of the inspection device is readjusted according to the deviation parameter to obtain the updated pose information. Finally, the inspection device performs repositioning and attitude adjustment according to the updated pose information and takes pictures of the inspection object again to ensure that the shooting results meet the accuracy requirements.

[0098] Through this step, it is analyzed whether the first captured image taken by the inspection device according to the target pose information meets the shooting accuracy requirements, thereby determining the accuracy of the pose of the inspection object. And based on the updated pose information determined by the feature point deviation parameter, necessary adjustments are made to the inspection device to improve the shooting accuracy and ensure the acquisition of high-quality inspection images.

[0099] As an optional embodiment, in response to the inspection request, target boxes corresponding to multiple inspection objects in the three-dimensional space corresponding to the target inspection area are marked, including: in response to the inspection request, the three-dimensional point cloud data corresponding to the target inspection area is retrieved; based on the three-dimensional point cloud data, the three-dimensional space corresponding to the target inspection area is determined; target boxes corresponding to multiple inspection objects in the three-dimensional space corresponding to the target inspection area are marked respectively.

[0100] In this embodiment, the specific steps of marking the target boxes corresponding to multiple inspection objects in the three-dimensional space corresponding to the target inspection area in response to the inspection request are described.

[0101] In this step, when the inspection request is received, the three-dimensional point cloud data related to the target inspection area will be retrieved, and these data provide detailed three-dimensional environment information within the target area. Using the three-dimensional point cloud data, a three-dimensional space model of the target inspection area is constructed. In the three-dimensional space model, a three-dimensional target box is marked for each inspection object that needs to be inspected. This box locates the precise spatial position and range of the inspection object and provides an important reference for subsequent positioning of the inspection device and performing inspection operations.

[0102] Through this step, target boxes corresponding to the inspection objects are marked in the three-dimensional space corresponding to the target inspection area, providing a basis for subsequent determination of the target inspection route, ensuring that the inspection device can accurately find the inspection objects in the complex three-dimensional space, and reducing the ineffective movement of the inspection device.

[0103] It should be noted that based on the above embodiments and optional embodiments, an optional implementation manner is provided, which is specifically described below.

[0104] Intelligent inspection robots are widely used in various scenarios such as substations, power plants, oil, and chemical industries. Especially in those places where regular safety inspections, maintenance monitoring, and environmental monitoring are required. The inspection robot can effectively replace manual labor in performing highly repetitive, labor-intensive, or potentially dangerous tasks, thereby improving work efficiency, reducing human errors, and ensuring personnel safety.

[0105] In the related art, it is necessary to manually set docking points and preset marked inspection points in advance according to the inspection objects, thus requiring a large amount of manual calibration work. At the same time, in case of temporarily or urgently needing to view unpreset points, it is impossible to quickly call the robot for inspection, and it is also necessary to reach the designated location after calibration, which cannot meet the flexible inspection requirements.

[0106] In view of this, in an alternative embodiment of the present invention, a robot vision target autonomous positioning inspection method based on a point cloud panoramic image is provided. Figure 2 It is a flowchart of the robot vision target autonomous positioning inspection method based on a point cloud panoramic image provided by an alternative embodiment of the present invention. As Figure 2 shown, in an alternative embodiment of the present invention, the robot can be quickly called for inspection in the case of unpreset points, and at the same time, the workload of manually marking preset points in advance is reduced.

[0107] The following will explain in detail the method steps provided by an alternative embodiment of the present invention.

[0108] S1. Receive an inspection request for the target inspection area.

[0109] S2. In response to the inspection request, mark the target boxes respectively corresponding to multiple inspection objects in the three-dimensional space corresponding to the target inspection area, and determine the normal vectors of the target boxes respectively corresponding to the multiple inspection objects.

[0110] Based on the three-dimensional point cloud data and the panoramic image (the same as the above regional image), mark multiple target objects to be inspected (the same as the above inspection objects). By marking in the three-dimensional point cloud space in the form of a rectangular box, the three-dimensional space marking coordinates are generated: . At the same time, the marked box (the same as the above target box) is mapped to the corresponding panoramic image to form a rectangular marked box in the panoramic image, where the standard panoramic image is used as the reference picture for the registration of the later autonomous positioning captured images.

[0111] After the marking is completed, the normal vector of the marked box is automatically generated, the points are named and saved, and the interactive objectified data parameters are formed, including the point identifier (ID), point name, point space coordinates, the normal vector of the marked box (the same as the above normal vector), and the panoramic image corresponding to the point.

[0112] Based on multiple inspected points completed with annotations, inspection instructions are issued. Objectified data parameters are transmitted to the inspection robot in real time through the Hypertext Transfer Protocol (HTTP interface).

[0113] S3. Determine the projection coordinates corresponding to multiple normal vectors in three-dimensional space respectively, and determine the docking parameters corresponding to multiple inspection objects based on the multiple projection coordinates.

[0114] In an alternative embodiment of the present invention, the determination of the inspection path and the calculation of the docking points adopt the normal vector projection orthogonality determination method, that is, the projection line of the normal vector line of each target inspection point is projected onto the ground, and the coincidence point of the projection line and the inspection route (the same as the above initial inspection route) is the docking point (the same as the above target inspection position).

[0115] Among them, the specific steps for determining the inspection route are as follows.

[0116] A1. Obtain the three-dimensional point cloud data corresponding to the target inspection area.

[0117] A2. Determine the passage area corresponding to the target inspection area based on the three-dimensional point cloud data.

[0118] The simultaneous localization and mapping (SLAM) navigation of the inspection robot Figure 1 Generally uses a two-dimensional grid map. The conversion of three-dimensional point cloud data into a two-dimensional grid map includes processes such as data preprocessing (filtering, downsampling), coordinate alignment, two-dimensional projection, grid map creation, occupancy grid update, and grid map generation.

[0119] In an alternative embodiment of the present invention, the Gaussian filtering algorithm is used to remove image noise, which has good smoothing effect and less negative impact, and is simple and fast in calculation.

[0120] Gaussian filtering formula As follows:[[]]END]]

[0121]

[0122] Among them, represents the distance offset from the center point, represents the standard deviation, is a constant.

[0123] When is larger, the curve is flatter; when is smaller, the curve is steeper.

[0124] The point cloud data is divided into voxels of a certain size (the same as the above obstacle voxels) through a voxel grid sampling algorithm, and then a representative point is selected for sampling in each voxel, thereby reducing the number of points. Voxel grid sampling can preserve the shape information of the point cloud, and the sampled points are uniform. A voxel can be regarded as a cube, representing uniformly spaced unit samples on a three-dimensional grid. Usually, the voxel value is mapped to 0 or 1, where 0 represents an empty voxel and 1 represents the presence of depth points in the voxel.

[0125] By setting a height threshold, the points above the threshold are regarded as obstacles, and those below the height threshold are regarded as non-obstacles. The grid map is created by initializing the grid with values according to the grid size to create two-dimensional data, and each element represents a grid. The occupancy state within the grid is determined by the projected obstacles. White represents the free and passable area, and the stored value is 0; black represents the occupied and non-passable area, and the stored value is 100; gray represents the unknown area where it is uncertain whether it can be passed, and the stored value is -1.

[0126] A3. Determine the initial inspection route based on the passable area.

[0127] S4. Determine the target inspection route according to the docking parameters corresponding to multiple inspection objects respectively.

[0128] After forming multiple docking points, they are sorted in turn according to the distance from the current coordinates of the robot and docked for inspection in order. The inspection path of the robot adopts a preset method, that is, the starting and ending coordinates and width of the feasible path are clearly defined in advance, and the roads that do not meet the inspection requirements have been avoided in advance. The inspection robot automatically determines the target inspection route according to the starting and ending points of the road and the road center line.

[0129] S5. Control the inspection device to operate the corresponding inspection object in the pose corresponding to the target pose information.

[0130] Figure 3 It is a schematic diagram of the autonomous positioning inspection of the robot vision target based on the point cloud panorama provided by an optional embodiment of the present invention, as Figure 3 shown, control the inspection device to reach the docking point according to the robot navigation path (the same as the above target inspection route), and use the pan-tilt camera to perform shooting operations on the inspection object.

[0131] After the inspection robot reaches the docking point of the inspection target, it obtains the current self-pose information and the spatial parameter information of the target object.

[0132] Since the camera coordinate system is the coordinate system of the camera itself, usually with the camera lens as the origin, the z-axis is perpendicular to the lens plane and points outwards, the x-axis is parallel to the lens plane and points to the right, and the y-axis is parallel to the lens plane and points downwards. The robot coordinate system (the same as the above inspection equipment coordinate system) is the coordinate system of the robot itself, usually with the center of the robot as the origin, where the x-axis points in the forward direction of the robot, the y-axis points to the left of the robot, and the z-axis points above the robot.

[0133] Therefore, it is necessary to transform between the camera coordinate system and the robot coordinate system. The robot pose information includes the coordinates of the current docking point of the robot. , after the coordinate transformation, since the initial coordinates of the camera can be calculated through the height of the robot body and the installation position of the pan-tilt camera (the same as the above equipment projection coordinates, ). Transform the camera coordinates and the image coordinates. The center point of the target detection object (the same as the above inspection object coordinates, ), and the projection point on the imaging plane is (the same as the above object projection coordinates, ).

[0134] Calculate the initial coordinates of the camera and the deviation value (the same as the above projection coordinate difference) as the basis for adjusting the omnidirectional movement, lens zooming, and zoom control (PTZ) parameters of the pan-tilt camera.

[0135] Call the camera according to the obtained deviation for proportional-integral-derivative (PID) fuzzy control adjustment.

[0136] The automatic deviation adjustment formula of the proportional-integral-derivative (PID) algorithm is:

[0137]

[0138] Among them, is the output, is the error of the controlled object at time t, is the error change rate of the controlled object at time t, are the proportional gain, integral gain, and derivative gain; the proportional gain makes the system respond sensitively and quickly adjusts the system error; the integral gain gradually eliminates the steady-state error of the system; the derivative gain predicts the change trend of the system error in advance, thereby eliminating the error and suppressing the oscillation generated during the adjustment process.

[0139] S6. Control the inspection equipment to update the pose corresponding to the pose information and photograph the corresponding inspection object.

[0140] After the inspection robot performs PID fuzzy control adjustment, it takes pictures of the target object, and matches them with the panoramic image to accurately confirm whether the visual requirements are met.

[0141] In an alternative embodiment of the present invention, the Scale-Invariant Feature Transform (SIFT) algorithm is used to implement feature extraction, and the Euclidean distance is used as the similarity criterion for multi-dimensional vectors. After the feature points between the captured image and the panoramic image are paired, a set of feature point pairs P is obtained:

[0142]

[0143] Wherein, is the nth set of feature point pairs.

[0144] The conversion relationship between pixel coordinates and camera coordinates is:

[0145]

[0146] Wherein, is the conversion matrix of the initial azimuth angle α, horizontal angle β, and pitch angle θ of the camera, is the coordinate before image offset, is the coordinate after image offset. The deviation of the general feature point set is calculated to obtain the deviation parameters of the robot pan-tilt camera, and the adjustment is made until the registration meets the requirements.

[0147] Through the above alternative embodiments, at least the following beneficial effects can be achieved:

[0148] (1) The robot vision target autonomous positioning and inspection method based on the point cloud panoramic map can quickly call the robot for inspection without preset points, reducing the workload of a large number of manual annotations of preset points in advance;

[0149] (2) By generating three-dimensional space annotation coordinates in the three-dimensional point cloud space and mapping the annotation box to the corresponding panoramic image, a rectangular annotation box is formed in the panoramic image. Combining the standard panoramic image as the reference picture for the later registration of the autonomously positioned captured image. Adopting a scheme that combines the coordinate deviation algorithm and the image registration algorithm can improve the visual accuracy of the inspection robot target and reduce the workload of manual repeated correction.

[0150] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0151] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to enable a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0152] Embodiment 2

[0153] According to an embodiment of the present invention, there is also provided a device for implementing the method for determining the above inspection route. Figure 4 It is a structural block diagram of the device for determining the inspection route according to an embodiment of the present invention, as Figure 4 shown. The device includes: a receiving module 402, a response module 404, a first determination module 406, a second determination module 408, a third determination module 410, and a fourth determination module 412. The device will be described in detail below.

[0154] The receiving module 402 is configured to receive an inspection request for a target inspection area, where the inspection request is a request for inspecting multiple inspection objects; the response module 404 is connected to the above receiving module 402 and is configured to, in response to the inspection request, mark the target frames respectively corresponding to the multiple inspection objects in the three-dimensional space corresponding to the target inspection area; the first determination module 406 is connected to the above response module 404 and is configured to determine the normal vectors of the target frames respectively corresponding to the multiple inspection objects, where the normal vector is perpendicular to the operation plane of the corresponding inspection object; the second determination module 408 is connected to the above first determination module 406 and is configured to determine the projection coordinates of the multiple normal vectors in the three-dimensional space respectively; the third determination module 410 is connected to the above second determination module 408 and is configured to determine the docking parameters respectively corresponding to the multiple inspection objects according to the multiple projection coordinates; the fourth determination module 412 is connected to the above third determination module 410 and is configured to determine the target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects.

[0155] It should be noted here that the above receiving module 402, response module 404, first determination module 406, second determination module 408, third determination module 410, and fourth determination module 412 correspond to steps S102 to S112 in the method for determining the inspection route. The instances and application scenarios implemented by the multiple modules and the corresponding steps are the same, but are not limited to the content disclosed in the above Embodiment 1.

[0156] Embodiment 3

[0157] According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including: a processor; a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement the inspection route determination method of any one of the above.

[0158] Embodiment 4

[0159] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by the processor of an electronic device, the electronic device can implement the inspection route determination method of any one of the above.

[0160] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0161] In the above embodiments of the present invention, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0162] In the several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of the units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the units or modules can be in an electrical or other form.

[0163] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0164] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0165] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0166] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for determining an inspection route, characterized in that, Including: Receiving an inspection request for a target inspection area, where the inspection request is a request for inspecting multiple inspection objects; In response to the inspection request, marking target boxes respectively corresponding to the multiple inspection objects in a three-dimensional space corresponding to the target inspection area; Determining normal vectors of the target boxes respectively corresponding to the multiple inspection objects, where the normal vectors are perpendicular to the operation planes of the corresponding inspection objects; Determining projection coordinates respectively corresponding to the multiple normal vectors in the three-dimensional space; Determining docking parameters respectively corresponding to the multiple inspection objects according to the multiple projection coordinates; Determining a target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects.

2. The method according to claim 1, characterized in that, Determining a target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects includes: Determining inspection object coordinates respectively corresponding to the multiple inspection objects according to the target boxes respectively corresponding to the multiple inspection objects; Determining an initial inspection route according to the inspection object coordinates respectively corresponding to the multiple inspection objects; Determining the target inspection route according to the initial inspection route and the docking parameters respectively corresponding to the multiple inspection objects.

3. The method according to claim 2, characterized in that, Determining an initial inspection route according to the inspection object coordinates respectively corresponding to the multiple inspection objects includes: Obtaining three-dimensional point cloud data corresponding to the target inspection area; Determining a passage area corresponding to the target inspection area according to the three-dimensional point cloud data, where a vertical height of an obstacle voxel corresponding to the passage area is less than a height threshold, and the obstacle voxel is a voxel composed of data points representing an obstacle; Determining an initial inspection route according to the inspection object coordinates respectively corresponding to the multiple inspection objects and the passage area corresponding to the target inspection area.

4. The method according to claim 1, wherein After determining the target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects, it further includes: When the target inspection route includes target inspection positions respectively corresponding to the multiple inspection objects, controlling an inspection device to sequentially reach the target inspection positions respectively corresponding to the multiple inspection objects according to the target inspection route; Determining adjustment parameters corresponding to the inspection device at a corresponding target inspection position; Determining target pose information corresponding to the inspection device when operating a corresponding inspection object according to the adjustment parameters of the inspection device at the corresponding target inspection position; Controlling the inspection device to operate the corresponding inspection object in a pose corresponding to the target pose information.

5. The method according to claim 4, characterized in that Determining the adjustment parameters corresponding to the inspection device includes: When the target inspection route includes initial pose information corresponding to the inspection device when operating a corresponding inspection object, obtaining inspection object coordinates respectively corresponding to the corresponding inspection object; Determining object projection coordinates corresponding to the corresponding inspection object according to the inspection object coordinates, and determining device projection coordinates corresponding to the inspection device when operating the corresponding inspection object according to the initial pose information; Determining a projection coordinate difference corresponding to the corresponding inspection object according to the object projection coordinates and the device projection coordinates; Determining the adjustment parameters according to the projection coordinate difference and the initial pose information.

6. The method according to claim 4, wherein After controlling the inspection device to be in the pose corresponding to the target pose information and operating the corresponding inspection object, the method further includes: When the operation is a photographing operation, and controlling the inspection device to be in the pose corresponding to the target pose information and operating the corresponding inspection object is to control the inspection device to be in the pose corresponding to the target pose information and photograph the corresponding inspection object to obtain a corresponding first photographed image, obtaining a regional image corresponding to the target inspection area; Determining first feature points of the corresponding inspection object according to the first photographed image; Determining second feature points of the corresponding inspection object according to the regional image; Determining corresponding feature point deviation parameters according to the first feature points and the second feature points; When the feature point deviation parameter is greater than a predetermined threshold, adjusting the target pose information when the inspection device operates the corresponding inspection object according to the corresponding feature point deviation parameter to obtain updated pose information when the inspection device operates the corresponding inspection object; Controlling the inspection device to photograph the corresponding inspection object in the pose corresponding to the updated pose information.

7. The method according to any one of claims 1 to 6, characterized in that, In response to the inspection request, marking target boxes respectively corresponding to the multiple inspection objects in a three-dimensional space corresponding to the target inspection area, including: In response to the inspection request, retrieving three-dimensional point cloud data corresponding to the target inspection area; Determining a three-dimensional space corresponding to the target inspection area according to the three-dimensional point cloud data; Marking target boxes respectively corresponding to the multiple inspection objects in the three-dimensional space corresponding to the target inspection area.

8. An apparatus for determining an inspection route, characterized in that, Including: A receiving module, configured to receive an inspection request for a target inspection area, where the inspection request is a request for inspecting multiple inspection objects; A response module, configured to, in response to the inspection request, mark target boxes respectively corresponding to the multiple inspection objects in a three-dimensional space corresponding to the target inspection area; A first determination module, configured to determine a normal vector of the target box respectively corresponding to each of the multiple inspection objects, where the normal vector is perpendicular to the operation plane of the corresponding inspection object; A second determination module, configured to determine projection coordinates respectively corresponding to the multiple normal vectors in the three-dimensional space; A third determination module, configured to determine docking parameters respectively corresponding to the multiple inspection objects according to the multiple projection coordinates; A fourth determination module, configured to determine a target inspection route according to the docking parameters respectively corresponding to the multiple inspection objects.

9. An electronic device, characterized in that, Including: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the inspection route determination method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the inspection route determination method according to any one of claims 1 to 7.

Images