Cable tunnel robot grounding current visual automatic guiding detection system and method

Abstract

The invention relates to the technical field of grounding current detection, in particular to a cable tunnel robot grounding current visual automatic guiding detection system and method, and the method comprises the steps: obtaining a routing inspection route of a cable tunnel robot; acquiring grounding network data of the cable tunnel; a robot inspection feature set is constructed according to the inspection route information, a robot inspection rule is constructed according to the robot inspection feature set, motion data information of the cable tunnel robot is acquired, and whether the motion state of the cable tunnel robot meets the robot inspection rule or not is judged; obtaining mechanical arm motion video data of the cable tunnel robot, and predicting a motion track of a mechanical arm; acquiring multispectral image data of the detection point, and establishing association between the multispectral image data and the position information of the tunnel robot; identifying the grounding electrode by analyzing the multispectral image data, and judging whether the grounding electrode is a detection point or not; and performing grounding current detection on the detection point, and transmitting the obtained detection data to an upper computer.

Description

Technical Field

[0001] The present invention relates to the technical field of ground current detection, and in particular to a visual automatic guidance detection system and method for a cable tunnel robot ground current. Background Art

[0002] In the operation and maintenance of power systems, cable tunnels are an important part of the power system. They are responsible for the centralized layout and protection of cables to ensure the stable transmission of electricity. With the increase in power demand, the scale of cable tunnels is also expanding. Therefore, the inspection and maintenance of cable tunnels are particularly important. In recent years, the rapid development of robotics has provided new solutions for the inspection of cable tunnels. In particular, tunnel robots (quadruped robots, etc.) have become an ideal choice for cable tunnel inspection due to their excellent terrain adaptability and stability. These tunnel robots are equipped with various sensors and detection equipment, and can perform automatic inspection tasks in cable tunnels.

[0003] Cables in cable tunnels usually need to be tested for ground current to ensure the safe operation of the power system. Abnormal ground current may cause equipment failure or even cause safety accidents. Therefore, accurate detection of the ground current status of the cable is an important part of ensuring the safety of the power system. At present, cable ground current detection in cable tunnels mainly relies on manual or fixed detection equipment. These methods have many limitations. Manual detection is difficult to achieve comprehensive coverage of the complex environment inside the tunnel, and it is inefficient and easily affected by human factors. Although fixed detection equipment can be used for long-term monitoring, it is complex to install, costly, and cannot adapt to the changing environment in the cable tunnel. Summary of the invention

[0004] The present invention monitors the measurement process of the ground current and the operating state of the tunnel robot by means of visual recognition and visual guidance.

[0005] The technical solution proposed by the present invention is: a visual automatic guidance detection method for ground current of a cable tunnel robot, the method comprising: Obtaining inspection route information of the cable tunnel robot, wherein the inspection route information includes the coordinates of each inspection point, the total inspection time, the number of inspection points, and the length of the inspection route; Acquire grounding network data of the cable tunnel, wherein the grounding network data includes but is not limited to the coordinates of each grounding electrode, ID information of the grounding electrode, quantity information of the grounding electrode, and grounding resistance of the grounding electrode; Construct a robot inspection feature set based on the inspection route information, construct a robot inspection rule based on the robot inspection feature set, obtain the motion data information of the cable tunnel robot, and determine whether the motion state of the cable tunnel robot meets the robot inspection rule; Obtain the motion video data of the robot arm of the cable tunnel robot, and predict the motion trajectory of the robot arm by analyzing the motion video data of the robot arm; Acquire multispectral image data of the detection point, and associate the multispectral image data with the position information of the tunnel robot; Identify the ground electrode by analyzing the multispectral image data, and determine whether the ground electrode is a detection point. If yes, proceed to the next step, otherwise return to the previous step; Conduct ground current detection at the detection point and transmit the obtained detection data to the host computer.

[0006] Preferably, the step of constructing a robot inspection feature set according to the inspection route information comprises the following steps: Construct inspection route parameter set based on inspection route information ;in, Respectively represent the detection points To the testing point Coordinate set, detection point To the testing point The inspection duration, total number of inspection points and inspection points To the testing point Length of inspection route; Extract multiple key features from the inspection parameter set to form an inspection key feature set ,in, They represent the tunnel robot from the detection point To the testing point The average movement speed of the tunnel inspection robot is The dwell time of each detection point, the tunnel robot from the detection point To the testing point Movement time and correction time parameters, .

[0007] Preferably, a grounding network parameter set is constructed based on the obtained grounding network data of the cable tunnel. ,in, Respectively indicate the grounding point To the grounding point The coordinate set of the grounding points, the total number of grounding points To the grounding point The grounding network length, grounding point To the grounding point The grounding resistance set.

[0008] Preferably, the step of constructing a robot inspection rule according to the robot inspection feature set, acquiring motion data information of the cable tunnel robot, and determining whether the motion state of the cable tunnel robot satisfies the robot inspection rule comprises the following steps: Set speed matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The running speed is , then when When the speed is matched to the value Assign 1 and output the speed matching value; otherwise, return abnormal alarm information 1; Set time matching value The initial value of is 0. Suppose the tunnel robot is in the The dwell time at the landing point , then when When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, return abnormal alarm information 2; Set time matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The residence time of When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, judge that the tunnel returns abnormal alarm information three.

[0009] Preferably, the acquiring of the motion video data of the mechanical arm of the cable tunnel robot and predicting the motion trajectory of the mechanical arm by analyzing the motion video data of the mechanical arm comprises the following steps: Preprocess the acquired robotic arm motion video; Divide the preprocessed robotic arm motion video into multiple continuous frame images; The key points and descriptors in the frame images are extracted by the feature extraction algorithm ORB, and the key points are matched between consecutive frame images by using a matcher; The correspondence matrix between the matched key points is estimated by the random sampling consensus RANSAC algorithm to calculate the posture change of the robot arm; Identify key nodes of the robot arm according to the posture changes, wherein the key nodes include the robot arm joints and the end effector; The optical flow method is used to calculate the motion trajectory of key nodes in continuous frame images, and the motion trajectory of key nodes is predicted by the pre-trained convolutional neural network ResNet18 model.

[0010] Preferably, the step of acquiring the multispectral image data of the detection point and associating the multispectral image data with the position information of the tunnel robot comprises the following steps: Preprocessing the acquired multi-spectral image data of the grounding electrode of the grounding point; Obtain the location information of the corresponding grounding electrode and convert it into an image coordinate system; Coordinate conversion through Geographic Information System (GIS) tools; Iterate over each pixel in the image, associate it with its corresponding geographic location, and generate a map containing the pixel and its geographic location.

[0011] Preferably, the step of identifying the ground electrode by analyzing the multispectral image data and determining whether the ground electrode is a detection point comprises the following steps: Obtain multispectral images of the grounding electrodes at the detection points and perform preprocessing; The pre-processed multi-spectral image of the ground electrode is compared with the pre-stored image of the ground electrode by a matcher to determine whether the ground electrode in the current multi-spectral image is a target to be detected; According to the matching results, it is judged whether the grounding electrode in the current multispectral image is structurally complete; Obtaining coordinate data of the location of the tunnel robot, coordinate data of the target detection point, and geographic location information of the multispectral image data; If the error between the coordinate data of the location of the tunnel robot, the coordinate data of the target detection point and the geographical location information of the multispectral image data is less than a preset position threshold P, it is determined that the multispectral image data matches the target detection point.

[0012] Preferably, the method of performing ground current detection at the detection point and transmitting the obtained detection data to the host computer includes the following steps: If it is judged that the multispectral image data matches the target detection point, the host computer issues a work instruction to the tunnel robot; After receiving the work instruction, the inspection robot controls the mechanical arm to drive the ground current measuring device to measure the ground current of the grounding point and start timing; the measured ground current is returned to the host computer, and the ground current measurement time is returned.

[0013] The present invention also provides a cable tunnel robot ground current visual automatic guidance detection system, comprising a tunnel robot, a navigation and positioning module, a visual guidance module, a motion control module, a current detection device, a communication module, and a data processing module arranged inside the tunnel robot; The navigation and positioning module is used to locate the detection point and the position of the tunnel robot; the visual guidance module is used to obtain the video information of the robotic arm and the image information of the grounding electrode at the detection point; the data processing module is used to process the robotic arm movement video, the grounding electrode image and the current data; the data processing module sends the processing results of the robotic arm video and the grounding electrode image information to the motion control module; the motion control module controls the robotic arm to drive the current detection device to move according to the processing results, performs live grounding current detection at the detection point, and transmits the detected current data to the data processing module to obtain the final result; the communication module is used to connect the tunnel robot and the host computer.

[0014] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the cable tunnel robot ground current visual automatic guidance detection method.

[0015] Beneficial effects of the present invention: 1. The present invention can autonomously perform live detection of ground current in a cable tunnel without manual intervention, thereby reducing the workload and time of manual inspections, avoiding the danger of direct manual contact with live parts, reducing personal safety risks, and reducing measurement accuracy problems caused by inaccurate manual operations.

[0016] 2. The present invention uses a visual guidance method, the acquired grounding electrode image of the detection point and the motion video information of the robot arm, and a machine learning algorithm to compare the image of the detection point with the pre-stored standard image of the grounding electrode to determine whether the grounding electrode at the detection point is abnormal (structure missing, grounding material specifications not meeting, etc.); by analyzing the motion video of the robot arm, predicting the motion trajectory of the robot arm, and adjusting the motion trajectory of the robot arm in real time, so that the current detection device can accurately contact the grounding electrode. In this way, the level of power cable detection is improved and the detection is made more intelligent. When the tunnel robot moves, the distance to the obstacle is monitored by the monocular ranging camera in the visual guidance module; when measuring the current, the monocular ranging camera can be used to monitor the distance between the current detection device and the grounding electrode at the detection point, so that it can adapt well to the complex power cable tunnel environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a flow chart of the cable tunnel robot ground current automatic guidance detection method. DETAILED DESCRIPTION

[0018] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0019] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0020] refer to Figure 1The technical solution provided by the present invention is: a cable tunnel robot ground current visual automatic guidance detection method, comprising the following steps: 1. Obtaining the inspection route information of the cable tunnel robot, wherein the inspection route information includes the coordinates of each inspection point, the total inspection time, the number of inspection points, and the length of the inspection route; 2. Obtaining the grounding network data of the cable tunnel, wherein the grounding network data includes but is not limited to the coordinates of each grounding electrode, the ID information of the grounding electrode, the number information of the grounding electrode, and the grounding resistance of the grounding electrode; 3. Construct a robot inspection feature set based on the inspection route information, construct a robot inspection rule based on the robot inspection feature set, obtain the motion data information of the cable tunnel robot, and determine whether the motion state of the cable tunnel robot meets the robot inspection rule. Specifically, the following steps are included:

[0021] Construct inspection route parameter set based on inspection route information ;in, Respectively represent the detection points To the testing point Coordinate set, detection point To the testing point The inspection duration, total number of inspection points and inspection points To the testing point Length of inspection route; Extract multiple key features from the inspection parameter set to form an inspection key feature set ,in, They represent the tunnel robot from the detection point To the testing point The average movement speed of the tunnel inspection robot is The dwell time of each detection point, the tunnel robot from the detection point To the testing point Movement time and correction time parameters, ; Based on the obtained grounding network data of the cable tunnel, a grounding network parameter set is constructed. ,in, Respectively indicate the grounding point To the grounding point The coordinate set of the grounding points, the total number of grounding points To the grounding point The grounding network length, grounding point To the grounding point The ground resistance set; Set speed matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The running speed is , then when When the speed is matched to the value Assign 1 and output the speed matching value; otherwise, return abnormal alarm information 1; Set time matching value The initial value of is 0. Suppose the tunnel robot is in the The dwell time at the landing point , then when When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, return abnormal alarm information 2; Set time matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The residence time of When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, judge that the tunnel returns abnormal alarm information three.

[0022] When the host computer receives the alarm information, it determines that the running speed of the tunnel robot does not meet the preset requirements, obtains the surrounding images of the tunnel robot through the multi-spectral camera module, and determines whether the movement of the tunnel robot is affected by external reasons. After eliminating the slow operation of the tunnel robot due to external reasons, the speed control command is sent to the tunnel robot to increase the running speed of the tunnel robot. If the speed requirement is still not met, it is determined that there is a fault in the power device of the tunnel robot.

[0023] When the host computer receives the second alarm message, it determines that the tunnel robot stays at the detection point for too long, obtains the grounding current measurement time of the tunnel robot, and determines the reason why the tunnel robot stays for too long. If the grounding current measurement time exceeds the preset standard measurement time (T1) by 2 seconds, it is determined that the alarm is caused by the long grounding current measurement time. The robot arm operation video information is analyzed again, the operation speed of each key node of the robot arm is obtained, and corresponding measures are taken to improve the operation speed of the robot arm.

[0024] , obtaining the motion video data of the mechanical arm of the cable tunnel robot, and predicting the motion trajectory of the mechanical arm by analyzing the motion video data of the mechanical arm; this is specifically achieved by the following steps: Preprocess the acquired robotic arm motion video; Divide the preprocessed robotic arm motion video into multiple continuous frame images; The key points and descriptors in the frame images are extracted by the feature extraction algorithm ORB, and the key points are matched between consecutive frame images by using a matcher; The correspondence matrix between the matched key points is estimated through the RANSAC algorithm to calculate the posture change of the robot arm; Identify key nodes of the robot arm according to the posture changes, wherein the key nodes include the robot arm joints and the end effector; The optical flow method is used to calculate the motion trajectory of key nodes in continuous frame images, and the motion trajectory of key nodes is predicted by the pre-trained ResNet18 model.

[0025] , obtaining multispectral image data of the detection point, and associating the multispectral image data with the position information of the tunnel robot; this is specifically achieved through the following steps: Preprocess the multispectral image data of the grounding electrode at the grounding point; obtain the location information of the corresponding grounding electrode and convert it into an image coordinate system; perform coordinate conversion through a geographic information system (GIS) tool; Iterate over each pixel in the image, associate it with its corresponding geographic location, and generate a map containing the pixel and its geographic location.

[0026] , identifying the ground electrode by analyzing the multispectral image data, and determining whether the ground electrode is a detection point, if yes, proceeding to the next step, otherwise returning to the previous step; specifically comprising the following steps: Obtain multispectral images of the grounding electrodes at the detection points and perform preprocessing; The pre-processed multispectral image of the ground electrode is compared with the pre-stored image of the ground electrode through a matcher (FLANN) to determine whether the ground electrode in the current multispectral image is a target that needs to be detected; Whether the grounding electrode in the current multispectral image is structurally intact is determined based on the matching results; whether the structure is intact here refers to whether the horizontal grounding flat steel structure of the grounding electrode is intact (for example, whether it contains gaps, fractures, etc.), and whether a vertical electrode is connected.

[0027] Obtaining coordinate data of the location of the tunnel robot, coordinate data of the target detection point, and geographic location information of the multispectral image data; If the coordinate data of the tunnel robot's location, the coordinate data of the target detection point and the multispectral image If the error between the geographical location information of the data is less than the preset location threshold P, it is determined that the multispectral image data matches the target detection point.

[0028] , perform ground current detection at the detection point and transmit the obtained detection data to the host computer.

[0029] Specifically, if it is determined that the multispectral image data matches the target detection point, the host computer sends a work instruction to the tunnel robot; after receiving the work instruction, the inspection robot controls the robotic arm to drive the grounding current measuring device to measure the grounding current of the grounding point, and starts timing; the measured grounding current is returned to the host computer, and the grounding current measurement time is returned.

[0030] The present invention also provides a cable tunnel robot ground current visual automatic guidance detection system, including a tunnel robot, a navigation and positioning module, a visual guidance module, a motion control module, a current detection device, a communication module, and a data processing module arranged inside the tunnel robot.

[0031] Among them, the visual guidance module is used to obtain the video information of the robotic arm and the image information of the grounding electrode at the detection point, including a monocular ranging camera and a multi-spectral camera module, with a positioning accuracy of 0.05mm.

[0032] The robotic arm is installed on the upper part of the tunnel robot, and its movement is controlled by the motion control module. The opening and closing degree of the robotic arm is ≥20cm, the arm span is ≥800mm, the degree of freedom is 6, and the load is ≥2kg.

[0033] Among them, the current detection device is integrated into the lightweight mechanical arm of the robot, which is used for current detection of the ground wire and the return line. The current detection bandwidth of the current detection device is 1kHz~10MHz, the power frequency current measurement range is 1A~500A, the current measurement accuracy is 0.5 level, and the resolution is 0.1A. The detected current data is transmitted to the data processing module to obtain the final result, and then transmitted to the host computer through the communication module.

[0034] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the cable tunnel robot ground current visual automatic guidance detection method.

[0035] In the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0036] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in a different order than the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0037] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be subject to any changes or modifications.

Claims

1. A cable tunnel robot ground current vision automatic guidance detection method, characterized in that: The method comprises: Obtaining inspection route information of the cable tunnel robot, wherein the inspection route information includes the coordinates of each inspection point, the total inspection time, the number of inspection points, and the length of the inspection route; Acquire grounding network data of the cable tunnel, wherein the grounding network data includes but is not limited to the coordinates of each grounding electrode, ID information of the grounding electrode, quantity information of the grounding electrode, and grounding resistance of the grounding electrode; Construct a robot inspection feature set based on the inspection route information, construct a robot inspection rule based on the robot inspection feature set, obtain the motion data information of the cable tunnel robot, and determine whether the motion state of the cable tunnel robot meets the robot inspection rule; Obtain the motion video data of the robot arm of the cable tunnel robot, and predict the motion trajectory of the robot arm by analyzing the motion video data of the robot arm; Acquire multispectral image data of the detection point, and associate the multispectral image data with the position information of the tunnel robot; Identify the ground electrode by analyzing the multispectral image data, and determine whether the ground electrode is a detection point, if yes, proceed to the next step, otherwise return to the previous step; Conduct ground current detection at the detection point and transmit the obtained detection data to the host computer.

2. The cable tunnel robot ground current visual automatic guidance detection method according to claim 1 is characterized in that: The method of constructing a robot inspection feature set according to the inspection route information includes the following steps: Construct inspection route parameter set based on inspection route information ;in, Respectively represent the detection points To the testing point Coordinate set, detection point To the testing point The inspection duration, total number of inspection points and inspection points To the testing point Length of inspection route; Extract multiple key features from the inspection parameter set to form an inspection key feature set ,in, They represent the tunnel robot from the detection point To the testing point The average movement speed of the tunnel inspection robot is The dwell time of each detection point, the tunnel robot from the detection point To the testing point Movement time and correction time parameters, .

3. The cable tunnel robot ground current visual automatic guidance detection method according to claim 2 is characterized in that: Based on the obtained grounding network data of the cable tunnel, a grounding network parameter set is constructed. ,in, Respectively indicate the grounding point To the grounding point The coordinate set of the grounding points, the total number of grounding points To the grounding point The grounding network length, grounding point To the grounding point The grounding resistance set.

4. The cable tunnel robot ground current visual automatic guidance detection method according to claim 3 is characterized in that: The method of constructing a robot inspection rule according to the robot inspection feature set, obtaining motion data information of the cable tunnel robot, and determining whether the motion state of the cable tunnel robot satisfies the robot inspection rule comprises the following steps: Set speed matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The running speed is , then when When the speed is matched to the value Assign 1 and output the speed matching value; otherwise, return abnormal alarm information 1; Set time matching value The initial value of is 0. Suppose the tunnel robot is in the The dwell time at the landing point , then when When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, return abnormal alarm information 2; Set time matching value The initial value of is 0. Suppose the tunnel robot starts from the ground point To the grounding point The residence time is When the speed is matched to the value Assign a value of 1 and output the time matching value; otherwise, judge that the tunnel returns abnormal alarm information three.

5. The cable tunnel robot ground current visual automatic guidance detection method according to claim 4 is characterized in that: The method of obtaining the motion video data of the mechanical arm of the cable tunnel robot and predicting the motion trajectory of the mechanical arm by analyzing the motion video data of the mechanical arm comprises the following steps: Preprocess the acquired robotic arm motion video; Divide the preprocessed robotic arm motion video into multiple continuous frame images; The key points and descriptors in the frame images are extracted by the feature extraction algorithm ORB, and the key points are matched between consecutive frame images by using a matcher; The correspondence matrix between the matched key points is estimated by the random sampling consensus RANSAC algorithm to calculate the posture change of the robot arm; Identify key nodes of the robot arm according to the posture changes, wherein the key nodes include the robot arm joints and the end effector; The optical flow method is used to calculate the motion trajectory of key nodes in continuous frame images, and the motion trajectory of key nodes is predicted by the pre-trained convolutional neural network ResNet18 model.

6. The cable tunnel robot ground current visual automatic guidance detection method according to claim 5 is characterized in that: The method of acquiring multispectral image data of the detection point and associating the multispectral image data with the position information of the tunnel robot comprises the following steps: Preprocessing the acquired multi-spectral image data of the grounding electrode of the grounding point; Obtain the location information of the corresponding grounding electrode and convert it into an image coordinate system; Coordinate conversion through Geographic Information System (GIS) tools; Iterate over each pixel in the image, associate it with its corresponding geographic location, and generate a map containing the pixel and its geographic location.

7. The cable tunnel robot ground current visual automatic guidance detection method according to claim 1 is characterized in that: The method of identifying the ground electrode by analyzing the multispectral image data and determining whether the ground electrode is a detection point comprises the following steps: Obtain multispectral images of the grounding electrodes at the detection points and perform preprocessing; The pre-processed multi-spectral image of the ground electrode is compared with the pre-stored image of the ground electrode by a matcher to determine whether the ground electrode in the current multi-spectral image is a target to be detected; According to the matching results, it is judged whether the grounding electrode in the current multispectral image is structurally complete; Obtaining coordinate data of the location of the tunnel robot, coordinate data of the target detection point, and geographic location information of the multispectral image data; If the error between the coordinate data of the location of the tunnel robot, the coordinate data of the target detection point and the geographical location information of the multispectral image data is less than a preset position threshold P, it is determined that the multispectral image data matches the target detection point.

8. The cable tunnel robot ground current visual automatic guidance detection method according to claim 7 is characterized in that: The method of performing ground current detection at the detection point and transmitting the obtained detection data to the host computer includes the following steps: If it is judged that the multispectral image data matches the target detection point, the host computer issues a work instruction to the tunnel robot; After receiving the work instruction, the inspection robot controls the mechanical arm to drive the ground current measuring device to measure the ground current of the grounding point and starts timing; The measured ground current is returned to the host computer, and the ground current measurement time is returned.

9. Cable tunnel robot ground current visual automatic guidance detection system, characterized by: It includes a tunnel robot, a navigation and positioning module, a visual guidance module, a motion control module, a current detection device, a communication module, and a data processing module arranged inside the tunnel robot; The navigation and positioning module is used to locate the detection point and the position of the tunnel robot; the visual guidance module is used to obtain the video information of the robotic arm and the image information of the grounding electrode at the detection point; the data processing module is used to process the robotic arm movement video, the grounding electrode image and the current data; the data processing module sends the processing results of the robotic arm video and the grounding electrode image information to the motion control module; the motion control module controls the robotic arm to drive the current detection device to move according to the processing results, performs live grounding current detection at the detection point, and transmits the detected current data to the data processing module to obtain the final result; the communication module is used to connect the tunnel robot and the host computer.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the cable tunnel robot ground current visual automatic guidance detection method as described in any one of claims 1 to 8.

Images