A method and system for inspection using a quadruped robot

Abstract

The application discloses a kind of application four-legged robot's inspection method and system, its method when receiving current time's to be analyzed inspection data, obtain inspection database and the equipment parameter of four-legged robot, to be analyzed with the standard inspection data of inspection database preset one by one matching inspection data, if not matched standard inspection data, then the inspection image in to-be-analyzed inspection data is input into the geometric analytic model that is well trained in advance, generate obstacle data, judge whether the passing height value in obstacle data is greater than the height threshold value of preset, if passing height value is greater than height threshold value, then according to to-be-analyzed inspection data and equipment parameter construction passing analysis function, generate passing analysis value, according to the comparison result of passing analysis value and preset passing threshold, adjust the inspection path of four-legged robot. Solve the existing in the process of inspection, due to the influence of maintenance and construction and other human activities, leading to the technical problem of low accuracy of four-legged robot to obtain inspection information.

Description

Technical Field

[0001] This invention relates to the field of robot inspection technology, and in particular to an inspection method and system using a quadruped robot. Background Technology

[0002] With the continuous deepening of condition-based maintenance of power equipment, and the continuous optimization of strategies such as equipment inspection intervals, maintenance timing, and maintenance methods, especially in power distribution systems where the types of equipment are complex and the number is huge, the impact of power distribution automation data on the power supply reliability of the power distribution system cannot be ignored. However, traditional manual inspection methods have problems such as abnormal detection methods, missed detections, repeated detections, low efficiency, and personnel safety hazards during the inspection process due to insufficient experience and varying skill levels of operators.

[0003] In existing technologies, quadruped robots are mainly used to replace traditional manual inspection methods. However, during the inspection process, human activities such as maintenance and construction affect the completeness of the inspection information obtained, resulting in a low accuracy rate for quadruped robots to obtain inspection information. Summary of the Invention

[0004] This invention provides an inspection method and system using a quadruped robot, which solves the technical problem that the accuracy of inspection information obtained by quadruped robots is low because human activities such as maintenance and construction affect the integrity of the inspection information obtained during the inspection process.

[0005] The first aspect of this invention provides an inspection method using a quadruped robot, comprising:

[0006] When the inspection data to be analyzed at the current moment is received, the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed are obtained.

[0007] The inspection data to be analyzed is matched one by one with the standard inspection data preset in the inspection database;

[0008] If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image.

[0009] Determine whether the clearance height value in the obstacle data is greater than a preset height threshold;

[0010] If the clearance height value is greater than the height threshold, then a clearance analysis function is constructed based on the obstacle data and the equipment parameters to generate a clearance analysis value corresponding to the inspection data to be analyzed;

[0011] The inspection path of the quadruped robot is adjusted based on the comparison between the passability analysis value and the preset passability threshold.

[0012] Optionally, the step of inputting the inspection image in the inspection data to be analyzed into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image if the inspection data to be analyzed does not match any of the standard inspection data includes:

[0013] If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analysis model to generate the pass width value, pass height value, and adjacent segment angle value corresponding to the inspection image; wherein, the geometric analysis model is used to analyze the input inspection image and output the corresponding pass width value, pass height value, and adjacent segment angle value;

[0014] The pass width value, the pass height value, and the adjacent segment angle value are used as obstacle data.

[0015] Optionally, the passability analysis function is specifically:

[0016]

[0017] Among them, T is the passability analysis value, l2 is the width value, l1 is the length value, α is the angle value of adjacent segments, L is the passability width value, and ε is the angle compensation coefficient of adjacent segments.

[0018] Optionally, the step of adjusting the inspection path of the quadruped robot based on the comparison result of the passability analysis value and the preset passability threshold includes:

[0019] Determine whether the passability analysis value is greater than a preset pass threshold;

[0020] If the passability analysis value is greater than the passability threshold, then the quadruped robot is determined to be passable;

[0021] If the passability analysis value is less than or equal to the passability threshold, then the target path adjacent to the current inspection path is selected as the new inspection path.

[0022] Jump to execute the step of obtaining the inspection database and the quadruped robot's equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

[0023] Optionally, before the step of obtaining the inspection database and the quadruped robot's equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received, the method further includes:

[0024] Several collection points are set on a pre-established inspection map, along with the execution instructions for the quadruped robot corresponding to each collection point;

[0025] The collection points are connected sequentially according to the preset inspection path to generate an inspection road network;

[0026] The shortest path formed by connecting all the collection points in the inspection road network is selected as the target path;

[0027] The quadruped robot is controlled to execute the execution instructions on the target path and to acquire the inspection data to be analyzed in real time.

[0028] A second aspect of the present invention provides an inspection system using a quadruped robot, comprising:

[0029] The inspection information acquisition module is used to acquire the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

[0030] The inspection information comparison module is used to match the inspection data to be analyzed with the standard inspection data preset in the inspection database one by one.

[0031] The fault data acquisition module is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image if the inspection data to be analyzed does not match any of the standard inspection data.

[0032] The judgment and analysis module is used to determine whether the clearance height value in the obstacle data is greater than a preset height threshold;

[0033] The passability analysis function construction module is used to construct a passability analysis function based on the obstacle data and the equipment parameters if the passability height value is greater than the height threshold, and generate a passability analysis value corresponding to the inspection data to be analyzed.

[0034] The inspection path adjustment module is used to adjust the inspection path of the quadruped robot based on the comparison result between the passability analysis value and the preset passability threshold.

[0035] Optionally, the inspection information comparison module includes:

[0036] The geometric analysis submodule is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analysis model if the inspection data to be analyzed does not match any of the standard inspection data, thereby generating the pass width value, pass height value, and adjacent segment angle value corresponding to the inspection image; wherein, the geometric analysis model is used to analyze the input inspection image and output the corresponding pass width value, pass height value, and adjacent segment angle value;

[0037] The fault data acquisition submodule is used to use the pass width value, the pass height value, and the adjacent segment angle value as obstacle data.

[0038] Optionally, the passability analysis function is specifically:

[0039]

[0040] Among them, T is the passability analysis value, l2 is the width value, l1 is the length value, α is the angle value of adjacent segments, L is the passability width value, and ε is the angle compensation coefficient of adjacent segments.

[0041] Optionally, the inspection path adjustment module includes:

[0042] The second judgment and analysis submodule determines whether the passability analysis value is greater than a preset passability threshold;

[0043] If the passability analysis value is greater than the passability threshold, then the quadruped robot is determined to be passable;

[0044] If the passability analysis value is less than or equal to the passability threshold, then the target path adjacent to the current inspection path is selected as the new inspection path.

[0045] The jump execution submodule is used to jump to execute the step of obtaining the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

[0046] Optional, also includes:

[0047] The inspection path planning module is used to set up several collection points on a pre-established inspection map, as well as the execution instructions of the quadruped robot corresponding to the collection points.

[0048] The collection points are connected sequentially according to the preset inspection path to generate an inspection road network;

[0049] The shortest path formed by connecting all the collection points in the inspection road network is selected as the target path;

[0050] The quadruped robot is controlled to execute the execution instructions on the target path and to acquire the inspection data to be analyzed in real time.

[0051] As can be seen from the above technical solutions, the present invention has the following advantages:

[0052] This invention addresses the problem of low accuracy in quadruped robot inspections caused by incomplete inspection information acquired during construction and maintenance due to the presence of incoming inspection data. The process involves matching the data against preset standard inspection data in the database. If no match is found, the inspection image is input into a pre-trained geometric model to generate obstacle data. The system then checks the obstacle height to determine if it exceeds a preset threshold. If so, a passability analysis function is constructed based on the data and equipment parameters, generating a passability analysis value. Finally, the quadruped robot's inspection path is adjusted based on the comparison between the passability analysis value and the preset passability threshold. This solution resolves the technical issue of incomplete inspection information acquired by quadruped robots during inspections, often due to construction and maintenance disruptions. This invention acquires and analyzes image and sound information in real time during the inspection process of a quadruped robot, enabling the quadruped robot to avoid routes involving construction and maintenance in a timely manner. This eliminates the impact of construction and maintenance on the acquisition of inspection information and improves the accuracy of the quadruped robot in acquiring inspection information. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating the steps of an inspection method using a quadruped robot, as provided in Embodiment 1 of the present invention;

[0055] Figure 2 This is a flowchart illustrating the steps of an inspection method using a quadruped robot, as provided in Embodiment 2 of the present invention.

[0056] Figure 3 This is a structural block diagram of an inspection system using a quadruped robot, provided in Embodiment 3 of the present invention. Detailed Implementation

[0057] This invention provides an inspection method and system using a quadruped robot, addressing the issue that in existing inspection processes, human activities such as maintenance and construction affect the integrity of the acquired inspection information, resulting in low accuracy of the quadruped robot in obtaining inspection information.

[0058] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0059] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of an inspection method using a quadruped robot, as provided in Embodiment 1 of the present invention.

[0060] This invention provides an inspection method using a quadruped robot, comprising:

[0061] Step 101: When the inspection data to be analyzed at the current moment is received, obtain the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed.

[0062] The inspection data to be analyzed refers to the inspection images and audio data acquired by the quadruped robot at the current moment.

[0063] The inspection database refers to an information database that contains all standard audio data and inspection images marked with obstacles related to construction and maintenance.

[0064] Equipment parameters refer to the length, width, height, adjacent segment angle compensation coefficient, passability coefficient, and motion amplitude coefficient of the quadruped robot.

[0065] In this embodiment of the invention, when the inspection image and audio data at the current moment are received, the inspection database corresponding to the inspection image and audio data and the length value, width value, height value, adjacent segment angle compensation coefficient, passability coefficient and motion amplitude coefficient of the quadruped robot are obtained.

[0066] Step 102: Match the inspection data to be analyzed with the standard inspection data preset in the inspection database one by one.

[0067] Standard inspection data refers to standard inspection images and standard audio data.

[0068] In this embodiment of the invention, the inspection images and audio data are matched one by one with the standard inspection images and standard audio data preset in the inspection database.

[0069] Step 103: If the inspection data to be analyzed does not match any of the standard inspection data, input the inspection image in the inspection data to be analyzed into the pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image.

[0070] Fault data refers to the width, height, and angle values ​​of obstacles in the inspection images.

[0071] In this embodiment of the invention, if the inspection image and the standard inspection image, and the audio data and the standard audio data do not match, the inspection image is input into a pre-trained geometric analysis model to generate the pass width value, pass height value and adjacent segment angle value corresponding to the inspection image.

[0072] It should be noted that the inspection image is generally a planar channel image of the inspection path. The geometric analytical model denotes the line connecting the inflection points of the planar channel in the planar channel image as the dividing line. When the boundary of the planar channel is a curve, according to the differential principle, a very small segment of the curve can be regarded as a straight line for calculation. The planar channel is divided into several segments with the dividing line as the boundary. Take the middle point on the intersection line of a segment and the obstacle area, and denote the line segment with the other intersection point of the segment as the starting point and perpendicular to the intersection line as the ending point. The line segment is the judgment line segment. The intersection of the region passing through the judgment line segment and perpendicular to the segment with the three-dimensional channel is the judgment section. The passing width value, passing height value and adjacent segment angle value of the inspection image are obtained from the judgment section.

[0073] Step 104: Determine whether the clearance height value in the obstacle data is greater than the preset height threshold.

[0074] The height threshold refers to the multiplication factor between the height value of a quadruped robot and the motion amplitude coefficient.

[0075] In this embodiment of the invention, it is determined whether the height value is greater than the product of the height value of the quadruped robot and the motion amplitude coefficient.

[0076] Step 105: If the height value is greater than the height threshold, then construct a passability analysis function based on the obstacle data and equipment parameters to generate the passability analysis value corresponding to the inspection data to be analyzed.

[0077] In this embodiment of the invention, if the height value is greater than the product of the quadruped robot's height value and the motion amplitude coefficient, a passability analysis function is constructed based on the obstacle data and the quadruped robot's equipment parameters, and the passability analysis value corresponding to the inspection data to be analyzed is calculated using the passability analysis function.

[0078] Step 106: Adjust the inspection path of the quadruped robot based on the comparison results between the passability analysis value and the preset passability threshold.

[0079] The pass threshold refers to the passability coefficient of a quadruped robot.

[0080] In this embodiment of the invention, it is determined whether the passability analysis value is greater than the passability coefficient of the quadruped robot. If the passability analysis value is greater than the passability coefficient of the quadruped robot, it is determined that the quadruped robot can pass through the current area. If the passability analysis value is less than or equal to the passability coefficient of the quadruped robot, it is determined that the quadruped robot cannot pass through the current area, and an adjacent path is selected from the preset inspection road network as a new inspection path.

[0081] In this embodiment of the invention, when the inspection data to be analyzed at the current moment is received, the inspection database corresponding to the inspection data to be analyzed and the equipment parameters of the quadruped robot are obtained. The inspection data to be analyzed is matched one by one with the preset standard inspection data in the inspection database. If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image. It is determined whether the passage height value in the obstacle data is greater than a preset height threshold. If the passage height value is greater than the height threshold, a passability analysis function is constructed based on the inspection data to be analyzed and the equipment parameters to generate a passability analysis value corresponding to the inspection data to be analyzed. Based on the comparison result of the passability analysis value and the preset passability threshold, the inspection path of the quadruped robot is adjusted. This solves the technical problem that in the existing quadruped robot inspection process, the inspection information obtained is incomplete due to the influence of construction and maintenance, resulting in low accuracy of the inspection information obtained by the quadruped robot. This invention acquires and analyzes image and sound information in real time during the inspection process of a quadruped robot, enabling the quadruped robot to avoid routes involving construction and maintenance in a timely manner. This eliminates the impact of construction and maintenance on the acquisition of inspection information and improves the accuracy of the quadruped robot in acquiring inspection information.

[0082] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps of an inspection method using a quadruped robot, as provided in Embodiment 2 of the present invention.

[0083] This invention provides an inspection method using a quadruped robot, comprising:

[0084] Step 201: When the inspection data to be analyzed at the current moment is received, obtain the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed.

[0085] In this embodiment of the invention, when the inspection image and audio data at the current moment are received, the preset inspection database and the length, width, height, adjacent segment angle compensation coefficient, passability coefficient and motion amplitude coefficient of the quadruped robot are used.

[0086] Before performing step 201, this method also includes the following steps:

[0087] S11. Set up several collection points on the pre-established inspection map, as well as the execution instructions of the quadruped robot corresponding to the collection points.

[0088] Execution instructions refer to the information acquired by the quadruped robot at the data collection point. For example, acquiring environmental information such as images and sounds at the data collection point, as well as patrol information.

[0089] In this embodiment of the invention, several collection points and corresponding operation instructions for the quadruped robot are set on a pre-established inspection map.

[0090] S12. Connect each collection point sequentially according to the preset inspection path to generate an inspection road network.

[0091] In this embodiment of the invention, each collection point is connected sequentially according to a preset inspection path to generate an inspection road network.

[0092] It should be noted that any two collection points may contain multiple different inspection paths.

[0093] S13. Select the shortest path formed by connecting all the collection points in the inspection road network as the target path.

[0094] In this embodiment of the invention, the shortest path formed by connecting various collection points in the inspection road network is selected as the target path.

[0095] S14. Control the quadruped robot to execute instructions on the target path and acquire the inspection data to be analyzed in real time.

[0096] In this embodiment of the invention, the quadruped robot is controlled to move along the target path and execute operation commands at the collection points, while simultaneously acquiring inspection images and audio data in real time during the movement.

[0097] Step 202: Match the inspection data to be analyzed with the standard inspection data preset in the inspection database one by one.

[0098] In this embodiment of the invention, the inspection images and audio data are matched one by one with the standard inspection images and standard audio data preset in the inspection database, and the inspection images and audio data are comprehensively analyzed.

[0099] It should be noted that if the inspection image or audio data matches any standard inspection data, it means that the current inspection path is under construction or maintenance, and the quadruped robot cannot pass through the current inspection path. It is necessary to select an adjacent inspection path as a new inspection path to continue to perform the inspection task.

[0100] Step 203: If the inspection data to be analyzed does not match any of the standard inspection data, input the inspection image in the inspection data to be analyzed into the pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image.

[0101] Furthermore, step 203 includes the following sub-steps:

[0102] S21. If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analysis model to generate the passing width value, passing height value, and adjacent segment angle value corresponding to the inspection image; wherein, the geometric analysis model is used to analyze the input inspection image and output the corresponding passing width value, passing height value, and adjacent segment angle value.

[0103] In this embodiment of the invention, when the inspection image and the standard inspection image, and the audio data and the standard audio data do not match, the inspection image is input into a pre-trained geometric analysis model to generate the pass width value, pass height value and adjacent segment angle value corresponding to the inspection image.

[0104] It should be noted that the geometric analytical model performs boundary calibration on the obstacle to obtain the coordinates of the obstacle's endpoints. By directly calculating the distance between the two endpoint coordinates and the difference between the ordinate distances of the two endpoints, the angle between the physical device measuring the length of the obstacle and the horizontal plane is obtained through the correction function in the geometric analytical model. The obtained distance difference is then corrected to obtain the relevant parameters of the obstacle, such as the width value, the height value, and the angle values ​​of adjacent segments.

[0105] It should be noted that the inspection images are captured by dual cameras on the quadruped robot. The object under test is parallel to the two imaging planes of the dual cameras. Based on the inspection images and the parameter information of the dual cameras, similarity calculations can be performed on the triangle formed by the direction and the lens optical center, as well as the triangle formed by the object and the lens optical center, to obtain the passing width value, passing height value, and angle value of adjacent segments, etc.

[0106] S22. Use the width value, height value, and angle value of adjacent segments as obstacle data.

[0107] In this embodiment of the invention, the width value, the height value, and the angle value of adjacent segments are selected as obstacle data.

[0108] Step 204: Determine whether the clearance height value in the obstacle data is greater than the preset height threshold.

[0109] In this embodiment of the invention, it is determined whether the clearance height value in the obstacle data is greater than the product between the height value of the quadruped robot and the motion amplitude coefficient.

[0110] Step 205: If the height value is greater than the height threshold, then construct a passability analysis function based on the obstacle data and equipment parameters to generate the passability analysis value corresponding to the inspection data to be analyzed.

[0111] The passivity analysis function is specifically:

[0112]

[0113] Among them, T is the passability analysis value, l2 is the width value, l1 is the length value, α is the angle value of adjacent segments, L is the passability width value, and ε is the angle compensation coefficient of adjacent segments.

[0114] In this embodiment of the invention, if the height value is greater than the product of the quadruped robot's height value and the motion amplitude coefficient, then a passability analysis function is constructed based on the l2 width value, l1 length value, α adjacent segment angle value, L passability width value, and ε adjacent segment angle compensation coefficient to generate the passability analysis value corresponding to the inspection data to be analyzed.

[0115] It should be noted that if the height value is less than or equal to the product of the quadruped robot's height value and the motion amplitude coefficient, it means that the quadruped robot cannot pass through the current inspection path and needs to select an adjacent inspection path as a new inspection path to continue the inspection task.

[0116] Step 206: Determine whether the passability analysis value is greater than the preset pass threshold.

[0117] In this embodiment of the invention, it is determined whether the passability analysis value is greater than the passability coefficient of the quadruped robot.

[0118] Step 207: If the passability analysis value is greater than the pass threshold, the quadruped robot is determined to pass.

[0119] In this embodiment of the invention, if the passability analysis value is greater than the passability coefficient of the quadruped robot, it means that the quadruped robot can cross the current obstacle and move along the current inspection path.

[0120] Step 208: If the passability analysis value is less than or equal to the passability threshold, then select the target path adjacent to the current inspection path as the new inspection path.

[0121] In this embodiment of the invention, if the passability analysis value is less than or equal to the passability coefficient of the quadruped robot, it means that the quadruped robot cannot cross the current obstacle and needs to select a target path adjacent to the current inspection path as a new inspection path to continue the inspection task.

[0122] It should be noted that the starting and ending points of the target path and the current inspection path are the same.

[0123] Step 209: Jump to execute the step of obtaining the inspection database and quadruped robot equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

[0124] In this embodiment of the invention, after a new inspection path is generated, the quadruped robot is controlled to move on the new inspection path, and then jumps to execute the steps of the preset inspection database and the quadruped robot's length, width, height, adjacent segment angle compensation coefficient, passability coefficient and motion amplitude coefficient when the inspection image and audio data at the current moment are received.

[0125] In this embodiment of the invention, when the inspection image and audio data at the current moment are received, the inspection database and the device parameters of the quadruped robot corresponding to the inspection image and audio data are obtained. The inspection image and audio data are matched one by one with the preset standard inspection image and standard audio data in the inspection database. If the inspection image and standard inspection image, and the audio data and standard audio data do not match, the inspection image is input into a pre-trained geometric analysis model to generate the passing width value, passing height value, and adjacent segment angle value corresponding to the inspection image. It is determined whether the passing height value is greater than the product between the quadruped robot's height value and the motion amplitude coefficient. If the passing height value is greater than the product, the inspection image is considered to be in good condition. The product of the quadruped robot's height and motion amplitude coefficient is used to construct a passability analysis function based on the width, passability height, adjacent segment angles, and the quadruped robot's equipment parameters. This function calculates the passability analysis value corresponding to the inspection data to be analyzed. It then determines whether the passability analysis value is greater than the quadruped robot's passability coefficient. If the value is greater, the quadruped robot is deemed to be able to pass the current area; otherwise, it is deemed unable to pass, and an adjacent path is selected from the preset inspection road network as a new inspection path. This solves the technical problem of incomplete inspection information obtained by existing quadruped robots during inspections due to construction and maintenance, resulting in low accuracy of the inspection information acquired. This invention acquires and analyzes image and sound information in real time during the inspection process of a quadruped robot, enabling the quadruped robot to avoid routes involving construction and maintenance in a timely manner. This eliminates the impact of construction and maintenance on the acquisition of inspection information and improves the accuracy of the quadruped robot in acquiring inspection information.

[0126] Please see Figure 3 , Figure 3 This is a structural block diagram of an inspection system using a quadruped robot, provided in Embodiment 3 of the present invention.

[0127] This invention provides an inspection system using a quadruped robot, comprising:

[0128] The inspection information acquisition module 301 is used to acquire the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed when it receives the inspection data to be analyzed at the current time.

[0129] The inspection information comparison module 302 is used to match the inspection data to be analyzed with the standard inspection data preset in the inspection database one by one.

[0130] The fault data acquisition module 303 is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image if the inspection data to be analyzed does not match any of the standard inspection data.

[0131] The judgment and analysis module 304 is used to determine whether the clearance height value in the obstacle data is greater than a preset height threshold.

[0132] The passability analysis function construction module 305 is used to construct a passability analysis function based on obstacle data and equipment parameters if the passability height value is greater than the height threshold, and generate the passability analysis value corresponding to the inspection data to be analyzed.

[0133] The inspection path adjustment module 306 is used to adjust the inspection path of the quadruped robot based on the comparison result between the passability analysis value and the preset passability threshold.

[0134] Optionally, the inspection information comparison module 302 includes:

[0135] The geometric analysis submodule is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analysis model if the inspection data to be analyzed does not match any of the standard inspection data. This model generates the pass width value, pass height value, and adjacent segment angle value corresponding to the inspection image. The geometric analysis model is used to analyze the input inspection image and output the corresponding pass width value, pass height value, and adjacent segment angle value.

[0136] Optionally, the passability analysis function is specifically:

[0137]

[0138] Among them, T is the passability analysis value, l2 is the width value, l1 is the length value, α is the angle value of adjacent segments, L is the passability width value, and ε is the angle compensation coefficient of adjacent segments.

[0139] The fault data acquisition submodule 303 is used to use the pass width value, pass height value and adjacent segment angle value as obstacle data.

[0140] Optionally, the inspection path adjustment module 306 includes:

[0141] The second judgment and analysis submodule determines whether the passability analysis value is greater than the preset pass threshold;

[0142] If the passability analysis value is greater than the pass threshold, the quadruped robot is determined to pass.

[0143] If the passability analysis value is less than or equal to the pass threshold, then the target path adjacent to the current inspection path is selected as the new inspection path.

[0144] The jump execution submodule is used to jump to execute the steps of obtaining the inspection database and quadruped robot equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

[0145] Optionally, it also includes:

[0146] The inspection path planning module 307 is used to set several collection points on a pre-established inspection map and the execution instructions of the quadruped robot corresponding to the collection points.

[0147] Connect each data collection point sequentially according to the preset inspection path to generate an inspection road network;

[0148] The shortest path formed by connecting all the data collection points in the inspection road network is selected as the target path.

[0149] Control the quadruped robot to execute commands on the target path and acquire inspection data to be analyzed in real time.

[0150] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0151] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0152] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. The above descriptions and embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An inspection method using a quadruped robot, characterized in that, include: When the inspection data to be analyzed at the current moment is received, the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed are obtained. The inspection data to be analyzed is matched one by one with the standard inspection data preset in the inspection database; If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analysis model to generate obstacle data corresponding to the inspection image, wherein the obstacle data includes the pass width value, the pass height value and the angle value of adjacent segments; Determine whether the clearance height value in the obstacle data is greater than a preset height threshold; If the clearance height value is greater than the height threshold, then a clearance analysis function is constructed based on the obstacle data and the equipment parameters to generate a clearance analysis value corresponding to the inspection data to be analyzed; The inspection path of the quadruped robot is adjusted based on the comparison result between the passability analysis value and the preset passability threshold. The equipment parameters refer to the length, width, height, adjacent segment angle compensation coefficient, passability coefficient, and motion amplitude coefficient of the quadruped robot. The passability analysis function is specifically: T= ; Where T is the passability analysis value, This represents the width of the quadruped robot. This represents the length of the quadruped robot. These are the angle values ​​of adjacent segments. By width value, This is the angle compensation coefficient between adjacent segments of the quadruped robot.

2. The inspection method using a quadruped robot according to claim 1, characterized in that, The step of inputting the inspection image from the inspection data to be analyzed into a pre-trained geometric analytical model to generate obstacle data corresponding to the inspection image if the inspection data to be analyzed does not match any of the standard inspection data includes: If the inspection data to be analyzed does not match any of the standard inspection data, the inspection image in the inspection data to be analyzed is input into a pre-trained geometric analysis model to generate the pass width value, pass height value, and adjacent segment angle value corresponding to the inspection image; wherein, the geometric analysis model is used to analyze the input inspection image and output the corresponding pass width value, pass height value, and adjacent segment angle value.

3. The inspection method using a quadruped robot according to claim 1, characterized in that, The step of adjusting the inspection path of the quadruped robot based on the comparison result of the passability analysis value and the preset passability threshold includes: Determine whether the passability analysis value is greater than a preset pass threshold; If the passability analysis value is greater than the passability threshold, then the quadruped robot is determined to be passable; If the passability analysis value is less than or equal to the passability threshold, then the target path adjacent to the current inspection path is selected as the new inspection path. Jump to execute the step of obtaining the inspection database and the quadruped robot's equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

4. The inspection method using a quadruped robot according to claim 1, characterized in that, Before the step of obtaining the inspection database and the quadruped robot's equipment parameters corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current moment is received, the method further includes: Several collection points are set on a pre-established inspection map, along with the execution instructions for the quadruped robot corresponding to each collection point; The collection points are connected sequentially according to the preset inspection path to generate an inspection road network; The shortest path formed by connecting all the collection points in the inspection road network is selected as the target path; The quadruped robot is controlled to execute the execution instructions on the target path and to acquire the inspection data to be analyzed in real time.

5. An inspection system using a quadruped robot, characterized in that, include: The inspection information acquisition module is used to acquire the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received. The inspection information comparison module is used to match the inspection data to be analyzed with the standard inspection data preset in the inspection database one by one. The fault data acquisition module is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analysis model if the inspection data to be analyzed does not match any of the standard inspection data, and generate obstacle data corresponding to the inspection image, wherein the obstacle data includes the pass width value, the pass height value and the angle value of adjacent segments; The judgment and analysis module is used to determine whether the clearance height value in the obstacle data is greater than a preset height threshold; The passability analysis function construction module is used to construct a passability analysis function based on the obstacle data and the equipment parameters if the passability height value is greater than the height threshold, and generate a passability analysis value corresponding to the inspection data to be analyzed. The inspection path adjustment module is used to adjust the inspection path of the quadruped robot based on the comparison result between the passability analysis value and the preset passability threshold. The equipment parameters refer to the length, width, height, adjacent segment angle compensation coefficient, passability coefficient, and motion amplitude coefficient of the quadruped robot. The passability analysis function is specifically: T= ; Where T is the passability analysis value, This represents the width of the quadruped robot. This represents the length of the quadruped robot. These are the angle values ​​of adjacent segments. By width value, This is the angle compensation coefficient between adjacent segments of the quadruped robot.

6. The inspection system using a quadruped robot according to claim 5, characterized in that, The inspection information comparison module includes: The geometric analysis submodule is used to input the inspection image in the inspection data to be analyzed into a pre-trained geometric analysis model if the inspection data to be analyzed does not match any of the standard inspection data, and to generate the pass width value, pass height value and adjacent segment angle value corresponding to the inspection image; wherein, the geometric analysis model is used to analyze the input inspection image and output the corresponding pass width value, pass height value and adjacent segment angle value.

7. The inspection system using a quadruped robot according to claim 5, characterized in that, The inspection path adjustment module includes: The second judgment and analysis submodule determines whether the passability analysis value is greater than a preset passability threshold; If the passability analysis value is greater than the passability threshold, then the quadruped robot is determined to be passable; If the passability analysis value is less than or equal to the passability threshold, then the target path adjacent to the current inspection path is selected as the new inspection path. The jump execution submodule is used to jump to execute the step of obtaining the inspection database and the equipment parameters of the quadruped robot corresponding to the inspection data to be analyzed when the inspection data to be analyzed at the current time is received.

8. The inspection system using a quadruped robot according to claim 5, characterized in that, Also includes: The inspection path planning module is used to set up several collection points on a pre-established inspection map, as well as the execution instructions of the quadruped robot corresponding to the collection points. The collection points are connected sequentially according to the preset inspection path to generate an inspection road network; The shortest path formed by connecting all the collection points in the inspection road network is selected as the target path; The quadruped robot is controlled to execute the execution instructions on the target path and to acquire the inspection data to be analyzed in real time.

Images