Cable tunnel inspection method and system based on inspection robot

Abstract

The invention relates to the technical field of power system robots, in particular to a cable tunnel inspection method and system based on an inspection robot, an inspection task is divided into a plurality of inspection road sections, in each inspection road section, the inspection robot runs in the tunnel direction, and a detection area advances; when the inspection robot reaches a set position in front of the obstacle, the inspection robot stops advancing and enters a scanning preparation state; the inspection robot uses the mechanical arm to drive the camera assembly to move to the direction facing the inspection target, and the cable layer conditions of the upper boundary and the lower boundary of the target and the obstacle conditions in the range of the left boundary and the right boundary of the inspection target are determined; and driving the camera assembly to extend into the space between the cable layers and approach the inspection target until the distance between the camera assembly and the inspection target meets a set requirement, and taking image information acquired by the camera assembly as inspection data during the period. On the basis of not changing the existing obstacle avoidance control, the obstacle avoidance of cable layer inspection is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power system robots, and in particular to a cable tunnel inspection method and system based on an inspection robot. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In power cable tunnels, due to the large number of cables, they are usually laid out in layers, that is, the cables are distributed on both sides of the tunnel, leaving a channel in the middle, and the cables are laid out in layers on each side. For example, they are divided into 3 to 4 layers, and the height of the brackets of each layer is about 50 cm apart. After the cables are laid, the distance between the cables and the bottom surface of the upper bracket is about 30 cm.

[0004] In the power tunnels where inspection robots are used, the inspection robots with fixed cameras can only operate in the middle of the tunnel, which limits the scanning range of the cameras and makes it impossible to inspect the area between two adjacent layers of cables close to the tunnel wall, and it is impossible to fully cover the cable layer. If there are abnormal conditions in this area, such as cable joint damage, local cable ablation, and abnormal local cable temperature, it is difficult to observe.

[0005] To address the above problems, some inspection robots can use robotic arms to carry cameras as monitoring devices at the end. Although the cameras can penetrate into the cable layer for inspection and detect abnormal conditions deep in the cable layer through cameras or infrared lenses, in addition to cables, there are various acquisition devices and control cabinets distributed in the tunnel. Due to size limitations, these devices are often arranged in the middle channel. For the cameras at the end of the robotic arms, these devices will become obstacles in the movement of the robots during inspections, and are prone to motion interference with the robot's robotic arm and the cameras at the end. This requires adjustments to the relevant obstacle avoidance control algorithms, which are too complex and costly to adjust. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a cable tunnel inspection method and system based on a patrol robot, which controls the patrol robot to run in segments. In each patrol section, the robot stops on the track and scans the space between cable layers, and returns to the scanning preparation state along the original route after the scanning, thereby realizing the inspection between the cable layers in the power tunnel.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A first aspect of the present invention provides a cable tunnel inspection method based on an inspection robot, comprising the following steps:

[0009] The inspection task is divided into multiple inspection sections. In each inspection section, the inspection robot runs along the cable tunnel direction, with the detection area facing the forward direction.

[0010] When the inspection robot reaches the set position in front of the obstacle, it stops moving forward and enters the scanning preparation state;

[0011] In the scanning preparation state, the inspection robot uses the mechanical arm to drive the camera assembly to move in the direction of the inspection target, determine the cable layer conditions at the upper and lower boundaries of the inspection target and the obstacles in the left and right boundaries of the inspection target, and confirm that there are no obstacles that cause motion interference when approaching the inspection target position;

[0012] According to the position of the inspection target, the robotic arm drives the camera component into the cable layer and approaches the inspection target until the distance between the camera component and the inspection target meets the set requirements. During this period, the image information obtained by the camera component is used as the inspection data.

[0013] Furthermore, the inspection task is divided into multiple inspection sections. Specifically, according to the location and shape of obstacles in the pre-established tunnel environment model, combined with the coverage of the inspection robot and the inspection task requirements, the inspection task is divided into multiple inspection sections.

[0014] Furthermore, the inspection robot moves along the direction of the cable tunnel, and the state in which the detection area faces the forward direction is the inspection state. In the inspection state, the detection area of ​​the inspection robot covers the width direction of the cable tunnel.

[0015] Furthermore, in the inspection state, when the distance between the inspection robot and the obstacle is not greater than the limit value, the forward movement is stopped and the scanning preparation state is entered; in the scanning preparation state, the robotic arm drives the camera assembly to move, respectively determine the upper and lower boundary ranges of the obstacle, and use the width of the middle channel as the left and right boundary ranges of the obstacle, and determine the corresponding obstacle avoidance action according to the obtained boundary range.

[0016] Furthermore, after the spatial inspection between the cable layers is completed, the robot arm drives the camera assembly to move along the original path and returns to the scanning ready state.

[0017] Furthermore, after the inspection operation of the current inspection section is completed, the inspection robot switches to the inspection state and continues to run along the track to the next inspection section to perform the corresponding cable layer inspection.

[0018] A second aspect of the present invention provides a cable tunnel inspection system based on an inspection robot, comprising:

[0019] The inspection robot includes a camera assembly connected by a mechanical arm, one end of the mechanical arm is movably connected to a guide rail, and the other end is connected to the camera assembly, the guide rail is located in the top space of the cable tunnel, and the camera assembly has a laser radar, an inspection lens and a distance sensor;

[0020] The central control unit uses the inspection lens to obtain image information, uses the distance sensor to obtain the distance information between the inspection robot and the obstacle, uses the lidar to obtain point cloud data and combines it with the pre-established tunnel environment model to determine the obstacle avoidance action of the robotic arm.

[0021] Furthermore, the camera assembly includes at least two groups of laser radars arranged in parallel on the front panel, and the area between the two groups of laser radars is provided with a patrol lens, which includes an infrared lens and a visible light lens arranged in parallel, and a distance sensor is also provided on the front panel.

[0022] Furthermore, the central control unit determines the motion angles of the joints of the robotic arm corresponding to the motion instructions according to the speed and position of the inspection robot running along the guide rail, and performs obstacle avoidance.

[0023] Furthermore, the union of the areas formed by the two groups of laser radars is the detection area, and the detection area covers the range of the inspection lens.

[0024] Compared with the prior art, one or more of the above technical solutions have the following beneficial effects:

[0025] 1. Considering that obstacles in power tunnels are generally located in the middle channel and rarely appear between cable layers, after determining the obstacle position according to the tunnel environment model, the robot is stopped at the set position in front of the obstacle to inspect the cable layer. Before the robot arm extends into the cable layer to start inspection, the cable layer conditions and possible obstacles within the boundary of the target position are further determined to ensure that there are no obstacles that cause motion interference when the robot extends into the cable layer. This allows obstacle avoidance to be achieved during inspections between cable layers in a low-cost manner while minimizing changes to the existing obstacle avoidance control algorithm.

[0026] 2. After completing the inspection of a certain cable layer, the robotic arm returns to the scanning preparation state according to the original motion path, which can maximize the use of the planned motion route and reduce the dependence on the existing obstacle avoidance control algorithm, further ensuring that the robot can achieve obstacle avoidance in a relatively low-cost manner and ensure that the process of executing the cable layer inspection will not interfere with obstacles in the channel.

[0027] 3. Divide all inspection tasks into sections, perform cable layer inspections in each inspection section, and aggregate data from multiple inspection sections to form the final inspection results, which is beneficial for subsequent section fault analysis of the cable tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0029] Figure 1 is a schematic diagram of the structure of a camera carried in a cable inspection robot provided by one or more embodiments of the present invention;

[0030] Figure 2 is a schematic diagram of the detection area distribution of the cable inspection robot provided by one or more embodiments of the present invention;

[0031] Figure 3 is a schematic diagram of the structure of a cable inspection robot in an inspection state provided by one or more embodiments of the present invention;

[0032] Figure 4 is a schematic diagram of the side view structure of the cable inspection robot in the inspection state provided by one or more embodiments of the present invention;

[0033] Figure 5-Figure 18 It is a schematic diagram of the inspection state A-detection state N of the cable inspection robot provided by one or more embodiments of the present invention.

[0034] In the figure: 1 tunnel, 2 cable layer, 3 guide rail, 4 robot body, 5 robotic arm, 6 camera assembly, 7 lidar, 8 detection area, 9 obstacle, 10 visible light lens, 11 data cable, 12 radar cover, 13 infrared lens, 14 laser ranging sensor. DETAILED DESCRIPTION

[0035] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0036] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0037] It should be noted that the terms herein are only intended to describe specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0038] As introduced in the background technology, if a structural type of using a robotic arm to carry an end camera is chosen to deal with the problem of cable layer inspection, although it is possible to reach into the space between the cable layers to carry out inspections, there will still be problems with obstacle interference, and the control algorithm required to achieve obstacle avoidance is relatively complex, and the cost of adjusting the algorithm is huge.

[0039] Therefore, the following embodiments provide a cable tunnel inspection method and system based on a patrol robot, taking into account that the remaining equipment in the power tunnel except the cables will be arranged in the middle channel rather than in the space between the cable layers. This means that the obstacles of the patrol robot only need to be considered during the operation along the track. If the patrol robot can patrol between cable layers at a certain specific position, it only needs to consider that the range of motion of the robotic arm does not interfere with the obstacle. Therefore, there is no need to make major changes to the existing obstacle avoidance control algorithm. It is only necessary to determine the location of the obstacle in the middle channel and control the patrol robot to run in segments. After reaching each point, it scans the space between the cable layers and returns to the original route after the scan, thereby realizing the patrol between the cable layers of the power tunnel in a relatively low-cost manner.

[0040] Embodiment 1:

[0041] The cable tunnel inspection system based on the inspection robot includes:

[0042] The inspection robot runs on the guide rail, the inspection robot is connected to the camera assembly 6 via the mechanical arm 5, and the guide rail is located in the top space of the power tunnel;

[0043] The central control unit determines the obstacle avoidance action of the robot arm 5 according to the image information acquired by the camera assembly 6 and the distance information to the obstacle, combined with the pre-established tunnel environment model.

[0044] like Figure 1-Figure 3 As shown, the inspection robot includes a robot body 4 running on a guide rail 3 , and the robot body 4 is connected to a camera assembly 6 via a mechanical arm 5 .

[0045] In this embodiment, the camera assembly 6 may be a camera assembly 6 that can rotate within a certain angle range to obtain multi-directional images, commonly known as a "dome camera".

[0046] like Figure 1 As shown, the camera assembly 6 includes at least two groups of laser radars 7 arranged in parallel on the front panel, and the area between the two groups of laser radars 7 is provided with an infrared lens 13, a visible light lens 10 and a laser ranging sensor 14. Each group of laser radars 7 has a corresponding radar shield 12. The data acquired by each laser radar, sensor and lens is transmitted to the robot body 4 through a data line 11.

[0047] In this embodiment, the camera assembly 6 is installed at the end of the robot arm 5 , is driven by the robot arm 5 , and moves along the direction of the guide rail 3 following the robot body 4 .

[0048] Each laser radar 7 has its own independent detection area 8. There is an intersection between the two detection areas. The overlapping area can enhance the sensitivity of radar detection. The union of the two radar detection areas can cover the passing area in front of the inspection robot, which is a safety guarantee for the inspection robot to avoid obstacles in front. Figure 2 As shown, the detection area of ​​the laser radar 7 is a four-sided pyramid projected forward, and can ensure sufficient detection accuracy within 50 cm in front. A single radar has an effective detection width of 50 cm in front, and the union of the effective detection range widths of the two radars is 60 cm within 50 cm in front of the radar. This range width is sufficient to cover the activity space in front of the robot 5, and can provide safety protection for the robot 5 to drive the camera assembly 6 to avoid obstacles.

[0049] When the camera component 6 senses the presence of an obstacle 9 in the front radar detection area 8, it will report the obstacle 9 signal to the central controller of the inspection robot through the data cable 11. The central controller will issue instructions to operate the robot or mechanical arm 5 to avoid the obstacle 9.

[0050] In this embodiment, the operating environment structure of the inspection robot is as follows: Figure 4 As shown, the tunnel 1 is arranged with cables on both sides, the middle part is a channel, the upper space of the channel is provided with guide rails 3 for the robot body 4 to move, and the cables are laid in layers on both sides of the tunnel 1. This embodiment has three groups of cable layers 2 arranged in parallel, and the robot arm 5 drives the camera assembly 6 to extend from the two adjacent groups of cable layers 2 to scan and inspect each layer of cable layer 2. At the same time, the other equipment arranged in the channel, such as cabinets, electrical boxes, etc., will interfere with the movement of the robot arm 5 along the track direction, so the lenses and sensors on the camera assembly 6 and the laser radar will be used to achieve obstacle avoidance.

[0051] Embodiment 2:

[0052] The cable tunnel inspection method based on the inspection robot is characterized by comprising the following steps:

[0053] According to the pre-established tunnel environment model, the shape and location of obstacles are determined, and the inspection task is divided into multiple inspection sections based on the inspection coverage and inspection task requirements;

[0054] In each inspection section, the inspection robot runs along the direction of the cable tunnel, with the detection area facing the forward direction, and enters the inspection state;

[0055] According to the location information of the obstacle in the tunnel environment model and the distance information obtained by the laser ranging sensor, after reaching the set position in front of the obstacle, the inspection robot stops moving and drives the camera assembly into a scanning preparation state through the mechanical arm;

[0056] In the scanning preparation state, the robot arm drives the camera assembly to move, respectively determine the upper and lower boundary ranges of the obstacle, and use the width of the middle channel as the left and right boundary ranges of the obstacle, and determine the corresponding obstacle avoidance action based on the obtained boundary ranges;

[0057] In the scanning preparation state, the robot arm drives the camera assembly to move toward the direction of the inspection target according to the inspection task, determines the cable layer conditions at the upper and lower boundaries of the target and the obstacles in the left and right boundaries of the target, and confirms the location of the inspection target;

[0058] According to the inspection target position, the robot arm drives the camera assembly into the cable layer and approaches the inspection target until the distance between the camera assembly and the inspection target meets the set requirements;

[0059] Using the image information obtained during the movement of the camera assembly as inspection data, the robotic arm drives the camera assembly along the previous step motion path in sequence and returns to the scanning preparation state;

[0060] The inspection robot switches from the scanning preparation state to the inspection state, runs along the track to the next inspection section, and performs corresponding cable layer inspection or obstacle avoidance.

[0061] The inspection task is divided into multiple inspection sections. Specifically, after the tunnel environment model is established, the location of the obstacle can be preliminarily determined. Since the range that can be covered by the camera assembly carried by the robotic arm is limited, the entire power tunnel is divided into multiple inspection sections based on the inspection task requirements and the location of the obstacle. In each inspection section, in addition to the normal inspection along the tunnel direction, it is also necessary to use the robotic arm to drive the camera assembly into the space between adjacent cable layers according to the requirements of the inspection task to scan the relevant inspection targets. For example, if there is a joint in a certain layer of cable, the cable joint is used as the inspection target to scan whether the joint is ablated, etc.

[0062] Obstacles in power tunnels are usually facilities such as instruments, equipment and cabinets required for operation, which are generally arranged in the middle channel and rarely in the space between cable layers. Therefore, interference will only occur when the robot runs along the track. It is only necessary to ensure that the range of motion of the robot arm does not interfere with the obstacle. After determining the position of the obstacle according to the tunnel environment model, this embodiment stops the robot near the obstacle to inspect the cable layer, and adds further boundary range scanning during the inspection of the cable layer, which will not change the existing obstacle avoidance control algorithm. At the same time, considering that the coverage of the robot for cable layer inspection is limited, the entire power tunnel is divided into multiple inspection sections. After the robot runs to a set position (which can be near the obstacle, or a position set according to the inspection requirements), it will perform the cable layer inspection. This can ensure that the robot's cable layer inspection process will not interfere with obstacles in the channel without changing the existing obstacle avoidance control algorithm.

[0063] The set position in front of the obstacle is specifically: the tunnel environment model has the theoretical position of the obstacle, and the distance sensor can obtain the distance of the obstacle in the direction of the robot's advance, and then determine the actual position of the obstacle. The two parts of position information are combined with the coverage of the robot when the robot arm is extended to determine the position where the camera component will not interfere with the obstacle when it is driven by the robot arm to rotate in the horizontal plane (especially when the robot arm rotates in the direction close to the obstacle). Since the surface of the obstacle is irregular, the use of the tunnel environment model and the distance sensor can ensure that the robot will not interfere with the obstacle when operating in the middle channel, but it is difficult to ensure that the robot arm will interfere with the obstacle when driving the camera component to extend into or exit the space between the cable layers. Therefore, the robot is stopped at the set position in front of the obstacle and enters the scanning preparation state, so that the robot arm can confirm whether there are any obstacles that have not been monitored by inspecting the boundary conditions of the target position before extending into the space between the cable layers.

[0064] In this embodiment, when the robot and the mechanical arm 5 carry the camera assembly 6 for inspection in the tunnel, the data obtained by the various sensors carried by the camera assembly 6 can be used to ensure the safety of the movement of the mechanical arm 5.

[0065] The laser radar 7 is used to obtain the point cloud data of the tunnel environment, and the location and shape of obstacles in the environment model are determined by establishing a tunnel environment model. The point cloud data can be obtained through each inspection, and the tunnel environment model is continuously improved through the data accumulated through multiple inspections. The process of modeling and updating the model is a prior art, and the detailed process is not repeated in this embodiment.

[0066] The image information of the tunnel environment is obtained by using the infrared lens 13 and the visible light lens 10 as the image required for the inspection. During the inspection, the infrared lens or the visible light lens is selected according to the lighting conditions.

[0067] The laser ranging sensor 14 is used to determine the distance between the camera assembly and the obstacle, as well as the boundary information of the obstacle, and to determine the action instructions required for the robot arm 5 to drive the camera assembly 6 to achieve obstacle avoidance. The obstacle position information and shape information determined in the tunnel environment model are preliminary information. In order to ensure that no interference occurs, the distance information obtained by the laser ranging sensor 14 is used as the basis for further confirmation and robot control instructions.

[0068] It is a complex process for laser radar to obtain obstacle information. This embodiment only introduces the basic process, including:

[0069] emitting laser pulses;

[0070] Measuring reflection time: laser pulses will hit surrounding objects and reflect back, measuring the time difference between each pulse being emitted and received;

[0071] Calculate the distance, using the speed of light and time difference to calculate the distance between each reflection point and the emission point;

[0072] Construct obstacle 3D coordinate information based on distance;

[0073] The shape, size and position of obstacles are identified through their three-dimensional coordinate information.

[0074] According to the running speed and position of the robot body along the guide rail, the movement angles of the joints of the robot arm 5 corresponding to the obstacle avoidance action instruction are determined to execute obstacle avoidance. The specific process is also the existing technology and will not be described in detail in this embodiment.

[0075] like Figure 3 As shown in the robot inspection state diagram A, when the robot is performing on-hook inspection along the guide rail, the detection area 8 of the laser radar 7 is facing the forward direction of the inspection. When the existence of an obstacle 9 is sensed, the obstacle information will be uploaded to the central control unit. Figure 3 There is a relatively narrow space between the cable layers 2. When it is necessary to enter this space for detection, the cables will also appear as obstacles to the detection.

[0076] like Figure 4As shown, it is the inspection state diagram B, which is the rear view of the robot in the inspection state, showing that the space between adjacent cable layers 2 needs to be detected. It should be noted that the cable layer inspection in this embodiment is worthy of using a visible light lens or an infrared lens to reach into the space above or below the cable layer 2 (in most cases, the space above) and obtain the corresponding image information. The "space between the cable layers 2" here is only an example, and it can also be the space above the first-layer cables. Since the space between the first-layer cables and the tunnel ceiling is higher, there is generally less chance of obstacles, so the space between the cable layers 2 is used as an example here.

[0077] The central control unit determines the obstacle avoidance action of the robot 5 based on the image information obtained by the camera assembly 6 and the distance information to the obstacle, combined with the pre-established tunnel environment model. Combined with the global path planning and local path planning algorithms, a safe and efficient operation path is planned for the robot 5. This ensures the movement safety of the camera ball 6. While the planned action is running, the detection area 8 is monitored in real time, and the movement strategy of the robot 5 is adjusted according to the sensor data to avoid collision.

[0078] The principles of path optimization are as follows:

[0079] 1. Give priority to the safest straight-in and straight-out path to reduce the hidden dangers of obstacles caused by lateral blind spots.

[0080] 2. Avoid crossing between trajectory lines to reduce lateral displacement and perform lateral displacement within the fire safety range.

[0081] 3. For fixed obstacles in the tunnel environment, try to first model the points of the obstacles, which can be used to assist path planning.

[0082] 4. Ensure smoothness of trajectory movement and avoid sharp turns and quick returns.

[0083] 5. When it is necessary to detect targets deep inside the cable, after observing the target, the camera should return to the original path, the same path, only the direction of travel is different.

[0084] like Figure 5As shown, it is the detection state A, which refers to the inspection state of the inspection robot along the direction of the cable tunnel. In this state, the inspection area 8 of the inspection robot faces the forward direction. The detection area 8 is a four-sided pyramid detection area 8 formed by two laser radars 7. The detection area 8 covers the detection area of ​​the lens (infrared lens 13 and visible light lens 10), ensuring that the inspection picture obtained by the infrared lens 13 and the visible light lens 10 is within the range of the tunnel environment model. In the inspection state, each robot arm 5 has a set angle, forming an on-hook inspection state to obtain a stable inspection picture. When the laser ranging sensor obtains the forward direction and encounters an obstacle 9 (the distance between the obstacle 9 and the laser ranging sensor is less than the limit value), it will be unlocked from the on-hook inspection state, restore the degree of freedom of the joints of the robot arm 5, and prepare to scan the obstacle within a certain range through the action of the robot arm 5 to confirm the size and outline of the obstacle.

[0085] like Figure 6 As shown, it is the detection state B. According to the distance information obtained by the laser ranging sensor, it is determined that the inspection robot has encountered an obstacle 9, the on-hook inspection state is released, and the mechanical arm 5 drives the camera assembly 6 to enter the scanning preparation state.

[0086] like Figure 7 As shown, it is detection state C, the robot arm 5 drives the camera assembly 6 to rotate upward first, and uses the detection area 8 of the laser radar to read the upper boundary of the obstacle, and the reading range does not exceed the movement space of the robot on the guide rail.

[0087] like Figure 8 As shown, it is the detection state D, the robot arm 5 drives the camera assembly 6 to rotate downward again, and uses the detection area 8 of the laser radar to read the lower boundary of the obstacle, and the reading range reaches the ground.

[0088] The upper and lower boundaries of the obstacle can be read using existing image recognition algorithms, or by combining point cloud data obtained by the lidar with a tunnel environment model. The specific process is existing technology and will not be described in detail here.

[0089] like Fig. 9As shown, it is the detection state E, the robot 5 drives the camera assembly 6 back to the scanning preparation state, and the detection in the up and down directions has been completed. Since the cable tunnel is a space with a much larger length direction than the width direction, the left and right detection range of the laser radar has covered the robot's wide movement space, and in this embodiment, the robot is running on a fixed track, and the robot has no additional movement stroke in the width direction. It only needs to ensure that the left and right range of the robot arm exceeds the width range of the robot body 4, so there is no need to perform left and right motion scanning. At this time, the data is uploaded to the central control unit, and according to the obstacle avoidance algorithm pre-stored in the central control unit, combined with the boundary size of the obstacle 9, the corresponding obstacle avoidance instruction is output, and the joints of the robot 5 are controlled to rotate to a set angle to form a posture to avoid the obstacle 9, and after passing the obstacle 9, it returns to the scanning preparation state, and at the same time saves the sequence number of the obstacle 9 and the corresponding obstacle avoidance action instruction. When returning, the corresponding obstacle avoidance action is executed according to the corresponding obstacle sequence number.

[0090] When detection is required inside the cable layer 2, due to the narrowness of the cable layer, it is even more necessary to determine the displacement path of the camera assembly and plan the reset path. In principle, the path planning is the same as the entry and the exit, which is explained in conjunction with the legend.

[0091] like Fig.10 As shown, it is the detection state F. Referring to the previous process, the height position and depth position of the cable layer are determined by the laser radar (it can also be based on the tunnel environment model or use the existing drawing data, construction data, etc. of the tunnel). The robot arm 5 first drives the camera assembly 6 to the corresponding layer of the cable layer 2 according to the height position, and then deflects (generally deflects 90°) toward the direction of the cable layer 2.

[0092] The cable layer 2 is generally located on the left and right sides of the cable tunnel. Therefore, this embodiment sets the scanning preparation state to position 0, that is, the camera assembly 6 is facing the axial direction of the cable tunnel and is located at the set height and width position. The set height and width position are as close as possible to the geometric center of the cable tunnel cross-section, or a position that makes the detection area 8 cover as large a range as possible.

[0093] like Fig.11 As shown, it is the detection state G. After the camera assembly 6 faces the direction of the cable layer 2, the upper and lower boundaries and the target position are confirmed through the up and down deflection action driven by the mechanical arm 5.

[0094] like Fig.12 As shown, it is the detection state H, the robot arm 5 drives the camera assembly 6 to deflect to the left to confirm the left boundary obstacle and the cable layer situation.

[0095] like Fig.13As shown, it is detection state I, the robot arm 5 drives the camera assembly 6 to deflect to the right to confirm the right boundary obstacle and the cable layer situation.

[0096] To confirm the obstacles and cable layer conditions in the upper, lower, left and right boundaries, existing image recognition algorithms can be used, or the image information can be observed by human eyes in a remote control manner. The specific process is not described in detail in this embodiment.

[0097] like Fig.14 As shown, it is the detection state J, the camera is facing and confirming the target position, preparing to approach the target at a close distance, and determining the distance to be moved by laser ranging.

[0098] Confirming the position of the observed target means obtaining images of obstacles and cable layers in the upper, lower, left and right boundaries, confirming that there are no obstacles within the boundary range that interfere with the movement of the robot 5 and the camera assembly 6, and after confirmation, the action of approaching the target can be performed. The obstacles here are generally two groups of cable layers 2 adjacent to each other, and components on the cable layers 2 (such as bridge support members, fasteners, etc.). If it is the cable layer of the first floor, the structural members on the top of the tunnel (such as the curved ceiling) may also be obstacles.

[0099] like Fig.15 As shown, it is the detection state K, the robot 5 drives the camera assembly 6 according to the depth position and enters the cable layer 2. The robot 5 approaches the target through the translational action and determines the distance for observing the target appropriately, until the visible light lens of the camera assembly 6 reaches the distance suitable for observing the target. At this time, the camera assembly 6 observes the target accurately, and if the lighting conditions in the cable layer are insufficient, it can be observed through the infrared lens.

[0100] like Fig.16 As shown, it is the detection state L, the camera assembly 6 has finished close-up observation of the target, and the robot arm 5 drives the camera assembly 6 to return from the original path to the detection state J, that is, the initial state of aiming at the target and preparing to penetrate and observe.

[0101] like Fig.17 As shown, it is the detection state M, the robot arm 5 drives the camera assembly 6 to deflect to the right, move out of the cable layer, until it faces the inspection direction.

[0102] like Fig.18 As shown, it is the detection state N. The cable layer area inspection is completed, and the robot arm 5 drives the camera assembly 6 to return to the detection state A, that is, to the on-hook inspection state.

[0103] Finally, the robot arm 5 returns to the on-hook inspection state and restarts the tunnel inspection process.

[0104] Under the control of the central control unit, the robot arm 5 will reasonably set the speed and acceleration limits of the robot arm 5 according to the environmental conditions and obstacle distribution in the tunnel in the above-mentioned work process to reduce the risk of collision.

[0105] Mainly, when approaching an obstacle, it automatically slows down and takes evasive measures. In more complicated situations, the robot arm 5 will stop moving and wait for rescue by staff to eliminate movement risks.

[0106] When the robot is in the inspection state or the robot arm 5 encounters an unexpected collision or an undiscovered obstacle, an emergency stop button or trigger condition is designed to immediately stop the movement of the robot arm 5 in case of an emergency.

[0107] The operation status and environment of the robot arm 5 can be observed in real time through the remote monitoring system so as to find problems and take measures in time. When necessary, the operator is allowed to remotely intervene in the operation of the robot arm 5 to ensure safety.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cable tunnel inspection method based on an inspection robot, characterized in that: The following steps are involved: The inspection task is divided into multiple inspection sections. In each inspection section, the inspection robot runs along the cable tunnel direction, with the detection area facing the forward direction. When the inspection robot reaches the set position in front of the obstacle, it stops moving forward and enters the scanning preparation state; In the scanning preparation state, the inspection robot uses the mechanical arm to drive the camera assembly to move in the direction of the inspection target, determine the cable layer conditions at the upper and lower boundaries of the inspection target and the obstacles in the left and right boundaries of the inspection target, and confirm that there are no obstacles that cause motion interference when approaching the inspection target position; According to the position of the inspection target, the robotic arm drives the camera component into the cable layer and approaches the inspection target until the distance between the camera component and the inspection target meets the set requirements. During this period, the image information obtained by the camera component is used as the inspection data.

2. The cable tunnel inspection method based on the inspection robot according to claim 1, characterized in that: The inspection task is divided into multiple inspection sections. Specifically, according to the location and shape of obstacles in the pre-established tunnel environment model, combined with the coverage of the inspection robot and the inspection task requirements, the inspection task is divided into multiple inspection sections.

3. The cable tunnel inspection method based on the inspection robot according to claim 1, characterized in that: The inspection robot runs along the direction of the cable tunnel. The state in which the detection area faces the forward direction is the inspection state. In the inspection state, the inspection robot's detection area covers the width direction of the cable tunnel.

4. The cable tunnel inspection method based on the inspection robot according to claim 1, characterized in that: In the inspection state, when the distance between the inspection robot and the obstacle is not greater than the limit, the robot stops moving forward and enters the scanning preparation state; in the scanning preparation state, the robotic arm drives the camera assembly to move, respectively determine the upper and lower boundary ranges of the obstacle, and use the width of the middle channel as the left and right boundary ranges of the obstacle, and determine the corresponding obstacle avoidance action based on the obtained boundary ranges.

5. The cable tunnel inspection method based on the inspection robot according to claim 1, characterized in that: After the spatial inspection between the cable layers is completed, the robotic arm drives the camera assembly along the original path and returns to the scanning ready state.

6. The cable tunnel inspection method based on the inspection robot according to claim 1, characterized in that: After the inspection operation of the current inspection section is completed, the inspection robot enters the inspection state and continues to run along the track to the next inspection section to perform the corresponding cable layer inspection.

7. A system for implementing the cable tunnel inspection method based on an inspection robot according to any one of claims 1 to 6, comprising: The inspection robot includes a camera assembly connected by a mechanical arm, one end of the mechanical arm is movably connected to a guide rail, and the other end is connected to the camera assembly, the guide rail is located in the top space of the cable tunnel, and the camera assembly has a laser radar, an inspection lens and a distance sensor; The central control unit uses the inspection lens to obtain image information, uses the distance sensor to obtain the distance information between the inspection robot and the obstacle, uses the lidar to obtain point cloud data and combines it with the pre-established tunnel environment model to determine the obstacle avoidance action of the robotic arm.

8. The cable tunnel inspection system based on the inspection robot according to claim 7, characterized in that: The camera assembly includes at least two groups of laser radars arranged in parallel on the front panel. The area between the two groups of laser radars is provided with a patrol lens, which includes an infrared lens and a visible light lens arranged in parallel. A distance sensor is also provided on the front panel.

9. In the cable tunnel inspection system based on the inspection robot as described in claim 7, the central control unit determines the movement angles of each joint of the robotic arm corresponding to the action instruction according to the speed and position of the inspection robot along the guide rail, and performs obstacle avoidance.

10. As described in the cable tunnel inspection system based on the inspection robot in claim 8, the union of the areas formed by the two groups of laser radars is the detection area, and the detection area covers the range of the inspection lens.

Images