Inspection robot control methods, devices, computer equipment, readable storage media and program products

Abstract

This application relates to a control method, device, computer equipment, computer-readable storage medium, and computer program product for an inspection robot. It includes: controlling the inspection robot to collect cabinet door image data of a target substation equipment cabinet; extracting a region image of the handle area on the target substation equipment cabinet door based on the cabinet door image data; inputting the region image of the handle area into a preset handle classification model to obtain the handle type; extracting the handle's structural parameters from the handle area image; searching a preset operation parameter template library based on the handle type and structural parameters to obtain a robotic arm operation template matching the handle; performing a reliability check on the robotic arm operation template; if the reliability check result is passed, extracting a set of robotic arm operation parameters from the robotic arm operation template; and controlling the robotic arm of the inspection robot based on the set of robotic arm operation parameters. This method can adapt to multiple specifications of non-standard handles.

Description

Technical Field

[0001] This application relates to the field of intelligent power inspection technology, and in particular to an inspection robot control method, device, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] With the development of intelligent power operation and maintenance, mobile inspection robots and machine vision technology, substation autonomous inspection robots equipped with robotic arms are gradually being put into use on site. These robots can move autonomously to equipment cabinets such as terminal boxes, control cabinets, and mechanism boxes, and rely on vision to perceive the cabinet door handles and drive the robotic arms to complete automatic cabinet opening inspections, which greatly reduces the workload of manual on-site operations, thus leading to the development of automatic cabinet opening control solutions based on vision recognition.

[0003] However, existing traditional processing methods require modifying the cabinet or manually calibrating each unit individually, and have a weak ability to adapt to multiple specifications of non-standard handles. Summary of the Invention

[0004] Therefore, it is necessary to provide a control method, device, computer equipment, computer-readable storage medium, and computer program product for inspection robots that can adapt to multiple specifications of non-standard handles, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a method for controlling an inspection robot, including:

[0006] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0007] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0008] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0009] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0010] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0011] In one embodiment, based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template matching the handle, including:

[0012] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0013] In one embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; the method further includes:

[0014] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0015] In one embodiment, the credibility verification of the robotic arm operation template includes:

[0016] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0017] In one embodiment, the robotic arm of the inspection robot is controlled based on a set of robotic arm operation parameters, including:

[0018] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0019] In one embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end approach direction, gripping offset, pressing stroke, and target pressing force;

[0020] According to the operating parameters, control the robotic arm of the inspection robot to perform corresponding actions, including: controlling the robotic arm to move along the end approach direction to the preset operating point of the handle, adjusting the contact position of the robotic arm according to the grasping offset; controlling the robotic arm to complete the pressing and unlocking action according to the pressing stroke and target pressing force.

[0021] Secondly, this application also provides a control device for an inspection robot, comprising:

[0022] The data acquisition module is used to control the inspection robot to acquire cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, it extracts the area image of the handle area on the target substation equipment cabinet door;

[0023] The extraction module is used to input the region image of the handle area into a preset handle classification model to obtain the handle type; and to extract the structural parameters of the handle from the region image of the handle area.

[0024] The retrieval module is used to search the preset operation parameter template library based on the handle type and structural parameters of the handle, and obtain the robotic arm operation template that matches the handle;

[0025] The verification module is used to verify the credibility of the robotic arm operation template. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0026] The control module is used to control the robotic arm of the inspection robot based on the set of robotic arm operation parameters.

[0027] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0028] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0029] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0030] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0031] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0032] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0033] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0034] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0035] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0036] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0037] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0038] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0039] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0040] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0041] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0042] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0043] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0044] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0045] The aforementioned inspection robot control method, device, computer equipment, computer-readable storage medium, and computer program product control the inspection robot to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, extract the area image of the handle region on the target substation equipment cabinet door; input the area image of the handle region into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the area image of the handle region; based on the handle type and structural parameters, search a preset operation parameter template library to obtain a robotic arm operation template matching the handle; perform a credibility verification on the robotic arm operation template; if the credibility verification result is passed, extract the set of robotic arm operation parameters from the robotic arm operation template; and control the robotic arm of the inspection robot based on the set of robotic arm operation parameters. This application, by extracting the complete geometric structural parameters of the cabinet door handle, combining the handle type and structural parameter dual-index matching of the robotic arm operation template, and adding a multi-level credibility verification mechanism, can adaptively adapt to non-standard cabinet door handles of different sizes and installation forms without modifying the cabinet or manually calibrating each unit individually. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart illustrating the control method for an inspection robot in one embodiment;

[0048] Figure 2 This is a flowchart illustrating the process of retrieving a preset operation parameter template library and obtaining a robotic arm operation template that matches the handle in one embodiment.

[0049] Figure 3 This is a structural block diagram of the inspection robot control device in one embodiment;

[0050] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] In one embodiment, such as Figure 1As shown, a control method for an inspection robot is provided. This embodiment illustrates the method applied to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0053] Step 102: Control the inspection robot to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, extract the area image of the handle area of ​​the handle on the target substation equipment cabinet.

[0054] After the robot reaches the target cabinet door, it uses an end-effector camera or an onboard camera to capture images of the cabinet door area and obtains the camera's intrinsic parameters, the extrinsic parameters between the camera and the robotic arm's end effector, and the current pose of the robotic arm. After obtaining the cabinet door image data, the cabinet door image area is preprocessed, including distortion correction, brightness equalization, reflection suppression, and edge enhancement.

[0055] Optionally, the region image of the handle area on the door of the target substation equipment cabinet can be extracted using a target detection model.

[0056] Step 104: Input the region image of the handle area into the preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area.

[0057] Optionally, the handle classification model can be any image classification model. The structural parameters of the handle include: Handle main axis direction: the inclination angle vector of the longest extension axis of the handle's main frame relative to the horizontal plane of the cabinet door, distinguishing the main extension orientation of handles and knobs; Circumscribed rectangle aspect ratio: the ratio of the long side to the short side of the smallest enclosing rectangle of the handle outline, used to quickly distinguish between slender handles, round knobs, and short push-button handles; Estimated convex height: the vertical distance from the outer surface of the handle to the reference plane of the cabinet door, used to determine whether the gripper can encircle and grasp; Griptable edge length: the physical length of a continuous and effective line segment on the unobstructed surface of the handle that can be stably gripped by the gripper; Candidate rotation center point: the three-dimensional coordinates of the axis of rotation for unlocking and rotating knobs; Pressing surface normal area: the normal vector of the force-bearing plane of the press-type handle and the complete pressing boundary range; Embedded groove opening boundary: the outline dimensions and boundary coordinates of the recessed groove opening of the embedded hidden handle; Handle posture relative to the cabinet door plane: horizontal, vertical, and inclined installation postures based on the main axis inclination angle.

[0058] The extraction methods include: Handle main axis direction: Perform least-squares straight-line fitting on all line segments of the handle frame, select the longest fitted line as the handle main axis, and calculate the tilt vector angle of this main axis relative to the horizontal axis of the cabinet door plane to obtain the handle main axis direction; for knob-type handles, use the perpendicular line from the center of the circle perpendicular to the cabinet door plane as the main axis direction. Circumscribed rectangle aspect ratio:

[0059] The minimum bounding rectangle of the handle outline is determined by obtaining the lengths of the long and short sides of the rectangle and calculating the ratio of the long to short sides, which is the aspect ratio of the bounding rectangle. The estimated convex height is obtained by subtracting the average depth of the cabinet door reference plane from the average depth of all pixels within the handle outline to get the vertical distance from the handle to the cabinet door plane. When depth data is unavailable, a monocular geometric dimension conversion estimation is used, and the height estimation confidence level is output simultaneously. The graspable edge length is determined by filtering continuous, effective gripping line segments along the handle's main axis without obstructions, grooves, or chamfers, removing truncated segments obstructed by the cabinet structure, and calculating the physical length of the effective line segments to obtain the graspable edge length. For knob-type handles, the 3D coordinates of the center of the circular outline are extracted using Hough circle detection; for rotary handles, the 3D coordinates of the rotation axis are obtained by fitting the rotation trajectory contours at both ends, serving as candidate rotation center points. Pressing surface normal region: For press-type handles, the flat pressing surface region is segmented, and plane fitting is performed on the depth pixels of the pressing surface to solve for the plane normal vector. The complete boundary range of the pressing surface is marked to form the pressing surface normal region parameters. Embedded groove opening boundary: For embedded handles, the inner and outer boundaries of the groove are identified based on grayscale and depth abrupt changes. The coordinates of the four boundaries of the groove opening and the length and width dimensions of the opening are extracted to obtain the embedded groove opening boundary parameters. Handle attitude relationship relative to the cabinet door plane: Using the cabinet door plane as a reference, the angle between the handle's principal axis vector and the horizontal axis of the cabinet door is compared to classify three installation attitudes: horizontal, vertical, and tilted. The tilt angle is output as the attitude relationship parameter of the handle relative to the cabinet door plane.

[0060] Step 106: Based on the handle type and structural parameters, retrieve the preset operation parameter template library to obtain the robotic arm operation template that matches the handle.

[0061] The preset operation parameter template library is a database pre-built and stored within the edge computing unit of the inspection robot. The library stores multiple operation templates, each corresponding to a handle type and a suitable range of structural parameters. Each operation template stores the following information: handle type identifier, allowable ranges for each structural parameter, robotic arm motion parameters, force control safety parameters, and abnormal stop conditions. The robotic arm motion parameters include end-effector approach direction, gripping offset, pressing stroke, rotation axis direction, and rotation angle. The force control safety parameters include target pressing force, maximum allowable torque, and door opening force threshold. The abnormal stop conditions are used to determine the rules for triggering a reset during robotic arm operation.

[0062] Step 108: Perform a credibility verification on the robotic arm operation template. If the credibility verification result is passed, extract the set of robotic arm operation parameters from the robotic arm operation template.

[0063] If the credibility verification result is unsuccessful, the robotic arm of the inspection robot is prohibited from performing high-risk cabinet opening actions such as grasping, pressing, and rotating. The current cabinet door image, extracted handle structure parameters, and real-time pose data of the inspection robot are collected and uploaded to the remote operation and maintenance terminal, triggering the remote operation demonstration collection process. Based on the action data obtained from manual remote operation, a new robotic arm operation template is generated and updated to the operation parameter template library.

[0064] Step 110: Control the robotic arm of the inspection robot based on the set of robotic arm operation parameters.

[0065] The cabinet door handles can be either push-button or rotary. In the push-button type, the handle is pressed by a push-button contact; in the rotary type, it is rotated by a gripper. Other structures may also be included, allowing for adaptive door opening via replaceable operating terminals.

[0066] The aforementioned inspection robot control method involves controlling the inspection robot to collect cabinet door image data of the target substation equipment cabinet; extracting the area image of the handle region on the target substation equipment cabinet door based on the cabinet door image data; inputting the area image of the handle region into a preset handle classification model to obtain the handle type; extracting the structural parameters of the handle from the area image of the handle region; searching a preset operation parameter template library based on the handle type and structural parameters to obtain a robotic arm operation template matching the handle; performing a credibility verification on the robotic arm operation template; if the credibility verification result is passed, extracting the robotic arm operation parameter set from the robotic arm operation template; and controlling the robotic arm of the inspection robot based on the robotic arm operation parameter set. This application, by extracting the complete geometric structural parameters of the cabinet door handle, combining the handle type and structural parameter dual-index matching of the robotic arm operation template, and adding a multi-level credibility verification mechanism, can adaptively adapt to non-standard cabinet door handles of different sizes and installation forms without modifying the cabinet or manually calibrating each unit individually.

[0067] In one exemplary embodiment, such as Figure 2 As shown, based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle, including:

[0068] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0069] For example, firstly, a candidate template set is generated based on the handle type. Each robotic arm operation template in the preset operation parameter template library is pre-bound with a unique standard handle type (press-type, knob-type), and the allowable range of structural parameters and robotic arm operation parameters for that handle type are stored accordingly. The target handle type identified in real time is compared one by one with the bound handle types of all operation templates in the template library. All operation templates that are completely consistent with the target handle type are selected, and irrelevant templates with mismatched types are eliminated, forming the initial candidate template set. Secondly, the comprehensive matching degree of structural parameters is calculated for each candidate template. For each candidate template in the candidate template set, eight handle structural parameters extracted in real time (handle main axis direction, aspect ratio of the circumscribed rectangle, estimated convex height, grippable edge length, candidate rotation center point, normal area of ​​the pressing surface, boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane) are compared with the corresponding preset allowable range of parameters for that candidate template. An interval-normalized deviation-weighted algorithm is used to calculate the overall matching degree of a single template. The matching degree ranges from 0 to 1, with a higher value indicating a higher compatibility between the current handle and the template. Specifically, the matching degree is calculated as follows: First, the matching score for each parameter is calculated. If the real-time extracted structural parameter value falls within the allowable range of the corresponding parameter of the template, the matching score for that parameter is 1. If it exceeds the range, the score decreases linearly based on the deviation distance. The single-item score = 1 - (actual deviation value of the parameter / maximum allowable deviation value of the parameter range), with a minimum single-item score of 0. Subsequently, fixed weighted weights were assigned to eight structural parameters (with core safety parameters having higher weights). The core weighted parameters were the handle's spindle direction, estimated outward convex height, graspable edge length, and candidate rotation center point. Auxiliary weighted parameters included the normal area of ​​the pressing surface, the boundary of the embedded slot opening, the aspect ratio of the circumscribed rectangle, and the handle's posture relative to the cabinet door plane. Finally, the individual matching scores of all parameters were multiplied by their corresponding weights and summed to obtain the comprehensive weighted matching degree of the candidate template. Then, candidate templates that passed the safety parameter boundary check were identified from the candidate template set. The candidate template with the highest matching degree was selected from these passed parameter boundary checks to serve as the robotic arm operation template for matching the handle.

[0070] In this embodiment, the computational workload is reduced by coarse screening based on handle type, and the compatibility between handle and template is quantified by weighted matching. At the same time, a core parameter out-of-bounds check is added to remove templates with safety risks. This allows for the quick and safe selection of operation templates that are compatible with the current cabinet door handle, improving the adaptability and operational safety of the robotic arm's autonomous cabinet opening.

[0071] In an exemplary embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; the method further includes:

[0072] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0073] Among the above structural parameters, the handle spindle direction, the estimated outward convex height, the grippable edge length, and the candidate point of the rotation center are core safety parameters, while the rest are non-core safety parameters.

[0074] For example, the safety parameter out-of-bounds verification only performs mandatory verification on four core structural parameters that affect the safety of the robotic arm's gripping, rotation, and alignment. During template matching, the handle spindle direction, estimated outward convex height, graspable edge length, and candidate rotation center point extracted from the current handle are compared with the corresponding parameter allowable ranges preset by the candidate template. If any core parameter exceeds the allowable range, it indicates that the current handle size, position, or shape is poorly adapted to the template. Continuing to use the template to perform actions is likely to cause safety problems such as alignment deviation, gripping failure, and torque exceeding limits. Therefore, the candidate template is directly judged to have failed the verification and is eliminated. If all four core parameters fall within the preset parameter range of the template, it proves that the template meets the safe operation conditions, and the safety parameter out-of-bounds verification is judged to have passed.

[0075] In this embodiment, only the four core structural parameters that affect the gripping and rotation safety of the robotic arm are subject to mandatory boundary checks. The remaining structural parameters are only used for matching degree ranking. This ensures the safety of the robotic arm operation and avoids alignment errors, gripping failures or collision risks caused by abnormal key parameters. It also preserves the adaptation and fault tolerance space for non-core parameters, thereby improving the rationality of template matching and the ability to adapt to different scenarios.

[0076] In one exemplary embodiment, the credibility verification of the robotic arm operation template includes:

[0077] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0078] For example, firstly, three key data points are acquired: 1) the type confidence score corresponding to the handle type identified by the model, representing the accuracy of the handle classification result; 2) the parameter estimation confidence score corresponding to all structural parameter extraction processes, representing whether the current image is unobstructed, glare-free, and the parameter extraction result is accurate; and 3) the overall matching degree between the current handle structural parameters and the optimal template parameter range, representing the template fit. The system then compares these three data points with pre-set thresholds. Only when the handle classification is sufficiently accurate, the structural parameter extraction is sufficiently reliable, and the template matching degree is sufficiently high—when all three conditions are met simultaneously—is the template matching result considered reliable, the reliability check passed, and subsequent calls to the template parameters to control the robotic arm's movements allowed. Conversely, if any of the three conditions—inaccurate type identification, large parameter extraction deviation, or low template fit—occur, it indicates a defect in the current visual recognition or template matching result, posing a safety hazard to direct operation. Therefore, the reliability check is deemed unsuccessful, and the robotic arm's cabinet-opening action is not executed.

[0079] In this embodiment, a three-layer joint verification is used to comprehensively screen for problems such as identification errors, parameter distortion, and template mismatch, avoiding blind operation of the robot based on unreliable data and greatly improving the stability and safety of the substation robot's autonomous cabinet opening.

[0080] In one exemplary embodiment, the robotic arm of the inspection robot is controlled based on a set of robotic arm operation parameters, including:

[0081] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0082] For example, based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action to unlock the cabinet door. After completing the cabinet door unlocking action, the robotic arm outputs a smooth pulling force to open the cabinet door, and the pulling force is strictly controlled throughout the process to not exceed the preset door opening force threshold, so as to avoid excessive pulling force causing cabinet door deformation, handle breakage or cabinet damage, and insufficient pulling force causing door opening failure or machine jamming.

[0083] In this embodiment, by using a tension threshold limit protection, the cabinet equipment and mechanical structure are effectively protected while ensuring smooth door opening, thereby improving the stability and safety of the robot's autonomous cabinet opening operation.

[0084] In an exemplary embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end approach direction, gripping offset, pressing stroke, and target pressing force;

[0085] According to the operating parameters, control the robotic arm of the inspection robot to perform corresponding actions, including: controlling the robotic arm to move along the end approach direction to the preset operating point of the handle, adjusting the contact position of the robotic arm according to the grasping offset; controlling the robotic arm to complete the pressing and unlocking action according to the pressing stroke and target pressing force.

[0086] For example, when the handle is identified as a push-button type, the robotic arm first moves to a preset operating point in front of the handle according to the set end-approach direction. Then, it fine-tunes the contact position of the front-end push-button contact by relying on the gripping offset to ensure that the contact is aligned with the pressing area. After the position is adjusted, the robotic arm advances forward by the set pressing stroke while maintaining a force not less than the target pressing force to complete the press-to-unlock mechanism. This ensures that the latch can be successfully released without failing to unlock due to insufficient force. When the handle is identified as a knob type, the corresponding operating parameters include: end-approach direction, gripping offset, rotation axis direction, rotation angle, and maximum allowable torque. At this time, the robotic arm moves to the knob operating point along the corresponding end-approach direction and adjusts the gripper position according to the gripping offset to ensure that the gripper firmly wraps around the knob. Then, the gripper rotates according to the preset rotation axis direction and the limited rotation angle, and the torque during operation does not exceed the maximum allowable torque set by the template. This completes the knob rotation unlock and prevents the handle from being broken or the cabinet lock cylinder from being damaged due to excessive torque.

[0087] In this embodiment, dedicated operating parameters are configured for both push-button and knob-type handles to finely constrain the robotic arm's travel path, contact position, range of motion, and force, enabling standardized automatic unlocking of different types of handles. At the same time, force control parameters such as target pressing force and maximum allowable torque limit the output force, ensuring smooth completion of the unlocking action while avoiding damage to the handle, lock cylinder, and cabinet due to excessive force, thus improving the adaptability and protection effect of the robot's autonomous cabinet opening operation.

[0088] In an exemplary embodiment, a method for controlling an inspection robot includes: controlling the inspection robot to collect cabinet door image data of a target substation equipment cabinet; extracting a region image of the handle area on the target substation equipment cabinet based on the cabinet door image data; inputting the region image of the handle area into a preset handle classification model to obtain the handle type; extracting structural parameters of the handle from the region image of the handle area; the structural parameters include: handle spindle direction, aspect ratio of the circumscribed rectangle, estimated convex height, graspable edge length, candidate rotation center point, normal area of ​​the pressing surface, boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; filtering operation templates in the operation parameter template library based on the handle type and the handle type matched by each operation template in the operation parameter template library to obtain a candidate template set; calculating the matching degree of the candidate template for each candidate template in the candidate template set according to the structural parameters and the preset parameter allowable range in the candidate template; identifying candidate templates that pass the safety parameter boundary check from the candidate template set, and selecting the candidate template with the highest matching degree from the identified candidate templates as the robotic arm operation template matched with the handle. The system obtains the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template. It then compares the type confidence score, parameter estimation confidence score, and matching degree with their respective preset verification thresholds. If the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, the credibility verification is considered successful. If any one of these three scores is less than the corresponding preset verification threshold, the credibility verification is considered unsuccessful. If the credibility verification result is successful, the system extracts the robotic arm operation parameter set from the robotic arm operation template. Based on the handle type, it reads the corresponding operation parameters and door opening force threshold from the robotic arm operation parameter set. According to the operation parameters, it controls the robotic arm of the inspection robot to perform the corresponding action, applying a force not exceeding the door opening force threshold to open the cabinet door. When the handle type is a push-button type, the corresponding operating parameters include: end-effector approach direction, gripping offset, pressing stroke, and target pressing force. According to the operating parameters, control the robotic arm of the inspection robot to perform the corresponding actions, including: controlling the robotic arm to move along the end-effector approach direction to the preset operating point of the handle, adjusting the contact position of the robotic arm according to the gripping offset, and controlling the robotic arm to complete the pressing and unlocking action according to the pressing stroke and target pressing force.The parameter out-of-bounds verification process includes: for any candidate template, determining whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtaining the judgment result; if the judgment result shows that any parameter among the handle spindle direction, estimated outward convex height, grippable edge length, or rotation center candidate point exceeds the preset allowable range, then the candidate template is determined to have failed the safety parameter out-of-bounds verification; if the judgment result shows that the handle spindle direction, estimated outward convex height, grippable edge length, and rotation center candidate point all do not exceed the preset allowable range, then the candidate template is determined to have passed the safety parameter out-of-bounds verification.

[0089] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0090] In one exemplary embodiment, such as Figure 3 As shown, a control device for an inspection robot is provided, including: a data acquisition module 301, an extraction module 302, a retrieval module 303, a verification module 304, and a control module 305, wherein:

[0091] The data acquisition module is used to control the inspection robot to acquire cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, it extracts the area image of the handle area on the target substation equipment cabinet door;

[0092] The extraction module is used to input the region image of the handle area into a preset handle classification model to obtain the handle type; and to extract the structural parameters of the handle from the region image of the handle area.

[0093] The retrieval module is used to search the preset operation parameter template library based on the handle type and structural parameters of the handle, and obtain the robotic arm operation template that matches the handle;

[0094] The verification module is used to verify the credibility of the robotic arm operation template. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0095] The control module is used to control the robotic arm of the inspection robot based on the set of robotic arm operation parameters.

[0096] In one exemplary embodiment, the retrieval module is further configured to:

[0097] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0098] In an exemplary embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; the retrieval module is further used for:

[0099] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0100] In one exemplary embodiment, the verification module is further configured to:

[0101] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0102] In one exemplary embodiment, the control module is further configured to:

[0103] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0104] In an exemplary embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end-approach direction, gripping offset, pressing stroke, and target pressing force; the control module is further configured to:

[0105] Control the robotic arm to move along the end-effector approach direction to the preset operating point of the handle, and adjust the contact position of the robotic arm according to the grasping offset; control the robotic arm to complete the press-to-unlock action according to the pressing stroke and target pressing force.

[0106] Each module in the aforementioned inspection robot control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0107] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores control data for the inspection robot. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an inspection robot control method.

[0108] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0109] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0110] The inspection robot is controlled to collect cabinet door image data of the equipment cabinet doors of the target substation; based on the cabinet door image data, the area image of the handle area on the handle of the equipment cabinet door of the target substation is extracted;

[0111] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0112] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0113] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0114] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0115] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0116] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0117] In one embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; when the processor executes the computer program, it also implements the following steps:

[0118] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0119] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0120] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0121] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0122] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0123] In one embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end-approach direction, gripping offset, pressing stroke, and target pressing force; the processor also implements the following steps when executing the computer program:

[0124] Control the robotic arm to move along the end approach direction to the preset operation point of the handle, and adjust the contact position of the robotic arm according to the grasping offset; control the robotic arm to complete the pressing and unlocking action according to the pressing stroke and target pressing force.

[0125] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0126] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0127] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0128] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0129] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0130] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0131] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0132] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0133] In one embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; when the computer program is executed by the processor, it also implements the following steps:

[0134] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0135] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0136] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0137] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0138] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0139] In one embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end-approach direction, gripping offset, pressing stroke, and target pressing force; when the computer program is executed by the processor, it also implements the following steps:

[0140] Control the robotic arm to move along the end-effector approach direction to the preset operating point of the handle, and adjust the contact position of the robotic arm according to the grasping offset; control the robotic arm to complete the press-to-unlock action according to the pressing stroke and target pressing force.

[0141] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0142] The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted;

[0143] Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area;

[0144] Based on the handle type and structural parameters, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle.

[0145] The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template.

[0146] The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

[0147] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0148] Based on the handle type and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated according to the structural parameters and the preset parameter allowable range in the candidate template. From the candidate template set, the candidate templates that pass the safety parameter out-of-bounds check are identified, and the candidate template with the highest matching degree is selected from the identified candidate templates as the robotic arm operation template that matches the handle.

[0149] In one embodiment, the structural parameters include: the handle's main axis direction, the aspect ratio of the circumscribed rectangle, the estimated convex height, the length of the grippable edge, the candidate point of the rotation center, the normal area of ​​the pressing surface, the boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; when the computer program is executed by the processor, it also implements the following steps:

[0150] For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range for each structural parameter in the candidate template, and obtain the judgment result; if the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, graspable edge length, or rotation center candidate point exceeds the preset allowable range, the candidate template is determined to have failed the safety parameter out-of-bounds check; if the judgment result shows that the handle spindle direction, estimated outward convex height, graspable edge length, and rotation center candidate point all do not exceed the preset allowable range, the candidate template is determined to have passed the safety parameter out-of-bounds check.

[0151] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0152] Obtain the type confidence score corresponding to the handle type, the parameter estimation confidence score corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; compare the type confidence score, parameter estimation confidence score, and matching degree with the corresponding preset verification thresholds; if the handle type confidence score is not less than the preset type verification threshold, the parameter estimation confidence score is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the credibility verification is deemed to have passed; if any one of the handle type confidence score, parameter estimation confidence score, and matching degree is less than the corresponding preset verification threshold, then the credibility verification is deemed to have failed.

[0153] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0154] Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; according to the operation parameters, the robotic arm of the inspection robot is controlled to perform the corresponding action, and the robotic arm is controlled to apply a pulling force not exceeding the door opening force threshold to open the cabinet door.

[0155] In one embodiment, when the handle type is a push-button type, the corresponding operating parameters include: end-approach direction, gripping offset, pressing stroke, and target pressing force; when the computer program is executed by the processor, it also implements the following steps:

[0156] Control the robotic arm to move along the end-effector approach direction to the preset operating point of the handle, and adjust the contact position of the robotic arm according to the grasping offset; control the robotic arm to complete the press-to-unlock action according to the pressing stroke and target pressing force.

[0157] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0158] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0159] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0160] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A control method for an inspection robot, characterized in that, The method includes: The inspection robot is controlled to collect cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, the area image of the handle area on the target substation equipment cabinet is extracted; Input the region image of the handle area into a preset handle classification model to obtain the handle type; extract the structural parameters of the handle from the region image of the handle area; Based on the handle type and structural parameters of the handle, a preset operation parameter template library is retrieved to obtain a robotic arm operation template that matches the handle; The credibility of the robotic arm operation template is verified. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template. The robotic arm of the inspection robot is controlled based on the set of robotic arm operation parameters.

2. The method according to claim 1, characterized in that, The step of retrieving a preset operation parameter template library based on the handle type and structural parameters of the handle to obtain a robotic arm operation template matching the handle includes: Based on the handle type of the handle and the handle type matched by each operation template in the operation parameter template library, the operation templates in the operation parameter template library are filtered to obtain a candidate template set. For each candidate template in the candidate template set, the matching degree of the candidate template is calculated based on the structural parameters and the preset parameter allowable range in the candidate template; From the candidate template set, identify the candidate templates that pass the safety parameter out-of-bounds check, and select the candidate template with the highest matching degree from the identified candidate templates as the robotic arm operation template that matches the handle.

3. The method according to claim 2, characterized in that, The structural parameters include: handle spindle direction, aspect ratio of the circumscribed rectangle, estimated convex height, grippable edge length, candidate rotation center point, normal area of ​​the pressing surface, boundary of the embedded slot opening, and the attitude relationship of the handle relative to the cabinet door plane; the method also includes: For any candidate template, determine whether each structural parameter of the handle exceeds the preset allowable range of each structural parameter in the candidate template, and obtain the determination result; If the judgment result shows that any parameter of the handle spindle direction, estimated outward convex height, grippable edge length, or candidate rotation center point exceeds the preset parameter allowable range, then the candidate template is determined to have failed the safety parameter out-of-bounds check. If the judgment result shows that the handle spindle direction, the estimated value of the outward convex height, the length of the graspable edge, and the candidate point of the rotation center are all within the preset parameter allowable range, then the candidate template is determined to have passed the safety parameter out-of-bounds check.

4. The method according to claim 1, characterized in that, The credibility verification of the robotic arm operation template includes: Obtain the type confidence level corresponding to the handle type, the parameter estimation confidence level corresponding to the structural parameters, and the matching degree between the structural parameters and parameter ranges in the robotic arm operation template; The type confidence, the parameter estimation confidence, and the matching degree are compared with the corresponding preset verification thresholds, respectively. If the confidence level of the handle type is not less than the preset type verification threshold, the confidence level of the parameter estimation is not less than the parameter verification threshold, and the matching degree is not less than the matching verification threshold, then the confidence verification is deemed to have passed. If any one of the handle type confidence, parameter estimation confidence, or matching degree is less than the corresponding preset verification threshold, the confidence verification is deemed to have failed.

5. The method according to claim 1, characterized in that, The control of the robotic arm of the inspection robot based on the set of robotic arm operation parameters includes: Based on the handle type, the corresponding operation parameters and door opening force threshold are read from the set of robotic arm operation parameters; According to the operating parameters, control the robotic arm of the inspection robot to perform the corresponding action, and control the robotic arm to apply a pulling force not exceeding the opening force threshold to open the cabinet door.

6. The method according to claim 5, characterized in that, When the handle type is a push-button type, the corresponding operating parameters include: end approach direction, gripping offset, pressing stroke, and target pressing force; The step of controlling the robotic arm of the inspection robot to perform corresponding actions according to the operating parameters includes: Control the robotic arm to move along the end approach direction to the preset operation point of the handle, and adjust the contact position of the robotic arm according to the grasping offset; control the robotic arm to complete the press unlocking action according to the pressing stroke and target pressing force.

7. A control device for an inspection robot, characterized in that, The device includes: The data acquisition module is used to control the inspection robot to acquire cabinet door image data of the target substation equipment cabinet; based on the cabinet door image data, it extracts the area image of the handle area of ​​the handle on the target substation equipment cabinet; The extraction module is used to input the region image of the handle area into a preset handle classification model to obtain the handle type; and to extract the structural parameters of the handle from the region image of the handle area. The retrieval module is used to retrieve a preset operation parameter template library based on the handle type and structural parameters of the handle, and obtain a robotic arm operation template that matches the handle; The verification module is used to verify the credibility of the robotic arm operation template. If the credibility verification result is passed, the set of robotic arm operation parameters is extracted from the robotic arm operation template. The control module is used to control the robotic arm of the inspection robot based on the set of robotic arm operation parameters.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images