Tunnel inspection robot and inspection method

Abstract

The present invention relates to a tunnel inspection robot and an inspection method. It includes an autonomous mobile robot, a frame is provided on the autonomous mobile robot, a plurality of laser radar sensors are provided on the frame, and a data processing module is also provided. The data processing module is used to receive the detection data of each laser radar sensor, and generate a standard tunnel model and an updated tunnel model according to the detection data, and obtain a matching error by matching the standard tunnel model and the updated tunnel model. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformation position, and the deformation size at the deformation position is further determined according to the detection data. The present application can automatically obtain the deformation position in the tunnel and the deformation size at the deformation position.

Description

Technical Field

[0001] The present invention relates to the field of tunnel detection, and in particular to a tunnel inspection robot and an inspection method. Background Art

[0002] Many concrete tunnels have been built on railways, highways and roads over the past 30 years, and this construction seems to be continuing. However, recently many accidents have occurred, which were caused by parts of the inner wall of concrete tunnels falling down. In the following, concrete tunnels are simply referred to as "tunnels". Tunnel accidents have caused serious losses to national property and have caused great concern in the whole society. These accidents may become more frequent from now on, because the construction of some tunnels has exceeded 30 years. In order to avoid the occurrence of these accidents, it is necessary to conduct regular inspections of tunnels and detect specific parts of the inner wall of the tunnel that may collapse in the near future. For example, the deformation of the inner wall of the tunnel includes bulges and depressions. Bulges mean that the corresponding part of the inner wall of the tunnel is likely to fall down in the near future, and depressions mean that the corresponding part has already fallen down.

[0003] In the prior art, the tunnel can be inspected by ultrasonic wave, but the location of severe deformation in the tunnel cannot be automatically found by a robot, nor can the size of the deformation at that location be determined by a robot.

[0004] For example, Chinese patent application CN202210052361.5 discloses an ultrasonic detection system for tunnels, which cannot solve the above problems. Summary of the invention

[0005] Based on this, a tunnel inspection robot is provided, which can automatically obtain the location of severe deformation in the tunnel according to the detection data obtained by the laser radar sensor.

[0006] A tunnel inspection robot includes a mobile robot, characterized in that a frame is provided on the mobile robot, a plurality of laser radar sensors are provided on the frame, and a data processing module is also included, wherein the data processing module is used to receive detection data of each laser radar sensor, and generate a standard tunnel model and an updated tunnel model according to the detection data, and obtain a matching error by matching the standard tunnel model with the updated tunnel model. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformed position.

[0007] This application obtains detection data by walking a mobile robot in a tunnel, and generates a standard tunnel model and an updated tunnel model based on the detection data. The standard tunnel model is the tunnel model obtained by the mobile robot when the tunnel is just built and the inner wall of the tunnel has not yet deformed. The updated tunnel model is the tunnel model obtained by the mobile robot after a certain period of time after the tunnel begins to be used. Each time the mobile robot detects the tunnel in the future, an updated tunnel model can be obtained.

[0008] The standard tunnel model and the updated tunnel model are matched to obtain a matching error. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is the deformation position.

[0009] In one implementation, a mobile robot coordinate system ∑ R , the forward direction of the mobile robot is the Y axis, denoted as Y R , establish a Z axis pointing upward and an X axis perpendicular to the Z axis and pointing to the left of the Z axis on a plane perpendicular to the Y axis. The Z axis is denoted as Z R , the X-axis is denoted as X R , the Z axis, Y axis and X axis intersect at the origin O R , several laser radar sensors are located at the origin O R The X-axis on the left side is distributed along 180 degrees to the origin O R The right side of the X-axis.

[0010] In one implementation, the number of lidar sensors is 24.

[0011] In one implementation, the standard tunnel model and the updated tunnel model are both constructed by a mobile robot moving in a tunnel, specifically including the following method:

[0012] Establish world coordinate system∑ W , Xw is the X axis of the world coordinate system, Yw is the Y axis of the world coordinate system, Zw is the Z axis of the world coordinate system, S j represents the LiDAR sensor, is W Based on the time ti of the laser radar sensor S j The coordinates of a point are obtained by reading Yes R Based on the lidar sensor S at time ti j The coordinates of the point obtained by the readings are Obtained according to the following formula:

[0013]

[0014]

[0015]

[0016] Among them, D ij It is a LiDAR sensor S j The reading at time ti, R j It is a LiDAR sensor S j Distance from coordinate origin O R The distance to point to Z R X on the left side of the axis R Axis is the reference axis, Φ j It is a LiDAR sensor S j The angle from the reference axis,

[0017] Obtained by the following formula,

[0018] θ iH It is time Y R The angle between X and Yw is also R The angle between Xw and is the laser radar sensor Sj in the mobile robot coordinate system ∑ R X R The coordinates of the axes,

[0019] YesS j The laser radar sensor is located in the mobile robot coordinate system R Y R The coordinates of the axes,

[0020]

[0021] The tunnel model is constructed by point coordinates in the world coordinate system.

[0022] In one implementation, the deformation position is determined by the following method:

[0023] (1)T U To update the tunnel model, T S is the standard tunnel model, T U Translate so that T U Starting point and T S The starting point coincides with

[0024] (2) For T U Rotate so that T U The end point and T S The end points coincide with each other.

[0025] (3) Calculate T S and T UThe matching error between the corresponding points in the S and T U The Euclidean distance between corresponding points in is taken as the matching error.

[0026] In one implementation, the method further includes: determining a deformation size at the deformation position according to the detection data.

[0027] In one implementation, the deformation position is an area C k , the deformation size is area C k The size of the area C k The width W,

[0028] The width W is obtained by the following formula:

[0029]

[0030]

[0031]

[0032] Where Sm refers to the number of readings located in area C k Each lidar sensor, r m YesS m The readings and S m Distance from coordinate origin O R The sum of the distances, β m Represents S m and S m+1 The angle between the two lidar sensors, D k YesS k The reading of the lidar sensor, S k The LiDAR sensor is the one located at the edge that can detect area C k The lidar sensor, α is S k The angle between the laser radar's transmitted signal and recovered signal.

[0033] In one implementation, the deformation size is area C k The size of the area C k The length L is obtained by the following formula:

[0034] L=y vR -y uR -2d F ,

[0035]

[0036] Among them, D um One of the S mThe lidar sensor is at time t u When in area C k The front end reading, α is S m The angle between the laser radar's transmitted signal and the received signal, y uR is the mobile robot at time t u When in area C k The world coordinate system of the front end ∑ W The Yw axis coordinate, y vR is the mobile robot at time t v When in area C k The world coordinate system of the backend ∑ W The Yw axis coordinates below.

[0037] In one implementation, for each lidar sensor S j , before emitting the laser, delay a certain time T wait , and T wait Alternates between two values ​​ai and bi.

[0038] A tunnel inspection method is provided, wherein a frame is provided on an autonomous mobile robot, and a plurality of laser radar sensors are provided on the frame. The autonomous mobile robot is moved along the tunnel, and detection data of each laser radar sensor is received. A standard tunnel model and an updated tunnel model are generated according to the detection data. The standard tunnel model and the updated tunnel model are matched to obtain a matching error. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformed position. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of a tunnel inspection robot according to an embodiment of the present application.

[0040] Figure 2 A schematic diagram of the arrangement of various laser radars implemented in this application.

[0041] Figure 3 Schematic diagram of interference or crosstalk generated by various laser radars in an embodiment of the present application.

[0042] Figure (a) is the direct critical path, and Figure (b) is the indirect critical path.

[0043] Figure 4 Schematic diagram of the world coordinate system and the mobile robot coordinate system of this application.

[0044] Figure 5 Schematic diagram of the calculation method of the width and length of the region Ck. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0046] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be an intermediate element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be an intermediate element at the same time.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0048] like Figure 1 As shown, an embodiment of the present application provides a tunnel inspection robot, which includes a mobile robot. The mobile robot can be an autonomous mobile robot, that is, the robot can move autonomously according to a set route or determine the moving route by itself. Such a robot can be implemented using an existing corresponding robot. It can be understood that the above-mentioned mobile robot can also be other types of movable mechanisms.

[0049] The mobile robot is provided with a frame, on which are provided a plurality of laser radar sensors, and also includes a data processing module, the data processing module being used to receive detection data from each laser radar sensor, and to generate a standard tunnel model and an updated tunnel model according to the detection data, and to obtain a matching error by matching the standard tunnel model with the updated tunnel model, if the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformation position, and the deformation size at the deformation position is further determined according to the detection data.

[0050] The tunnel inspection robot of the present application includes two steps when used. The first step is to build a standard tunnel model. The standard tunnel model refers to a tunnel model without deformation. The standard tunnel model is built in advance by running the mobile robot throughout the tunnel. The second step is automatic tunnel detection, which is performed regularly, for example, once a month, once every two months, etc. The mobile robot continuously performs automatic tunnel inspection while moving, and the detection system detects the shape of the inner wall of the tunnel and builds an updated tunnel model. Then, it performs a partial match between the standard tunnel model and the updated tunnel model. If the matching error exceeds a predetermined threshold, the system will make a mark, such as a red dot, where the deformation exists. In this way, the deformed area in the tunnel can be determined.

[0051] Specifically, the mobile robot itself can carry the above data processing module. The data of the data processing module can be transmitted to an external terminal.

[0052] Specifically, Figure 2 and Figure 4 As shown, in one embodiment, the present application establishes a mobile robot coordinate system ∑ R , the forward direction of the mobile robot is the Y axis, denoted as Y R , establish a Z axis pointing upward and an X axis perpendicular to the Z axis and pointing to the left of the Z axis on a plane perpendicular to the Y axis. The Z axis is denoted as Z R , the X-axis is denoted as X R , the Z axis, Y axis and X axis intersect at the origin O R , several laser radar sensors are located at the origin O R The X-axis on the left side is distributed along 180 degrees to the origin O R The right side of the X-axis.

[0053] Specifically, each laser radar sensor is located at a distance from the origin O R The distances can be the same or different.

[0054] Specifically, the number of lidar sensors can be set according to actual conditions, for example, it can be 10, 20, 24, etc.

[0055] Figure 2 In, S j represents the jth lidar sensor, R j Indicates that from S j To the origin O R To point to Z R X on the left side of the axis R axis is the reference axis, Φj is the laser radar sensor S j The angle from the reference axis.

[0056] On the basis of the above, if Figure 4As shown, the standard tunnel model and the updated tunnel model are both constructed by moving a mobile robot in a tunnel, specifically including the following methods:

[0057] Establish world coordinate system∑ W , Xw is the X axis of the world coordinate system, Yw is the Y axis of the world coordinate system, Zw is the Z axis of the world coordinate system, S j represents the LiDAR sensor, is W Based on the time ti of the laser radar sensor S j The initial moving point of the mobile robot is ∑ W The origin of Yes R Based on the lidar sensor S at time ti j The coordinates of the point obtained by reading, Obtained according to the following formula:

[0058]

[0059]

[0060]

[0061] Among them, D ij It is a LiDAR sensor S j The reading at time ti, R j It is a LiDAR sensor S j Distance from coordinate origin O R The distance to point to Z R X on the left side of the axis R Axis is the reference axis, Φ j is the angle between the lidar sensor Sj and the reference axis.

[0062] Obtained by the following formula.

[0063] θ iH It is time Y R The angle between X and Yw is also R The angle between Xw and YesS j The laser radar sensor is located in the mobile robot coordinate system R X R The coordinates of the axis.

[0064] YesS j The sensor is in the mobile robot coordinate system∑ R YR The coordinates of the axis.

[0065]

[0066] The above method can be used to obtain the point coordinates in the world coordinate system, and the tunnel model is constructed by the point coordinates in the world coordinate system. That is, the standard tunnel model and the updated tunnel model are both constructed by the above method.

[0067] The mobile robot of this application moves throughout the tunnel. All readings from the LiDAR sensor are converted to world coordinates. With these coordinate data, an updated tunnel model can be constructed. Let T S and T U represent the standard tunnel model and the updated tunnel model respectively, and the deformation position is determined by the following method:

[0068] (1)T U To update the tunnel model, T S is the standard tunnel model, T U Translate so that T U Starting point and T S The starting point coincides with

[0069] (2) For T U Rotate so that T U The end point and T S The end points coincide with each other.

[0070] (3) Calculate T S and T U The matching error between the corresponding points in the S and T U The Euclidean distance between corresponding points in is taken as the matching error.

[0071] Here, T S and T U The Euclidean distance between corresponding points in is taken as the matching error, that is:

[0072]

[0073] in, It is T S The point in It is T U The point in and i represents the time, j represents the number of lidar sensors, and 24 are given in the above example, so there are 24 data here.

[0074] Furthermore, after finding the deformation area, the present application can also determine the deformation size at the deformation position according to the detection data. The deformation size may include length and width, etc.

[0075] On the basis of the above, if Figure 5 As shown, the deformation position is an area C k , the deformation size is area C k The size of the area C k The width W,

[0076] The width W is obtained by the following formula:

[0077]

[0078]

[0079]

[0080] Among them, S m It means the reading is in area C k Each lidar sensor, r m YesS m The readings and S m Distance from coordinate origin O R The sum of the distances of S m and S m+1 The angle between the two lidar sensors, or S m With the coordinate origin O R The connection line with S m+1 With the coordinate origin O R The angle between the lines connecting the k YesS K The reading of the lidar sensor, S K The LiDAR sensor is the one located at the edge that can detect area C k LiDAR sensors, such as Figure 5 In, S K The LiDAR sensor is from Z R X on the left side of the axis R Axis starts, the last detectable area C in the clockwise direction k The lidar sensor. α is S K The angle between the laser radar's transmitted signal and recovered signal.

[0081] In one embodiment, the deformation size is area C k The size of the area C k The length L is obtained by the following formula:

[0082] L=yvR -y uR -2d F .

[0083]

[0084] Among them, D um One of the S m The lidar sensor is at time t u When in area C k The front end reading, that is, S m represents any lidar sensor. α is S m The angle between the laser radar sensor’s transmitted signal and the received signal, y uR is the mobile robot at time t u When in area C k The world coordinate system of the front end ∑ W The Yw axis coordinate, y vR is the mobile robot at time t v When in area C k The world coordinate system of the backend ∑ W The Yw axis coordinates below.

[0085] In one embodiment, crosstalk is an undesirable phenomenon in the detection process when multiple lidar sensors are used. Figure 3 Two scenarios of a mobile robot equipped with multiple lidar sensors in a tunnel environment are shown, which differ greatly in the crosstalk patterns. In the following discussion, the term "critical path" is used, which is defined as any path along which a laser signal emitted by one lidar sensor is received by one or more other lidar sensors. The emitting lidar sensor is labeled S m , each receiving lidar sensor is labeled S n . Figure 3 Figure (a) shows a direct critical path where the lidar sensor S n With LiDAR sensor S m On a critical path. If any lidar sensor S n In the LiDAR sensor S m It was launched shortly afterwards, so the LiDAR sensor S m When the echo reaches it, the lidar sensor S n will wait for its own signal to echo before taking action. Therefore, the signal from the lidar sensor S n The readings will result in some minor errors, which will depend on the lidar sensor S m and S nThe time difference between lag . Figure 3 Figure (b) shows the indirect critical path. Here, at a time point t0, the lidar sensor S m The signal is emitted and reflected by the three inner walls of the tunnel. Assuming the walls are fairly smooth, the reflected wave will reach the lidar sensor S after traveling a distance D = d1 + d2 + d3 + d4. n If at this time, the laser radar sensor S n It is waiting for its own echo, then it will receive the signal from the lidar sensor Sm and regard it as its own echo. These two critical paths are easily satisfied in a tunnel environment.

[0086] The above interference and crosstalk problems will have a certain impact on the detection accuracy. In order to solve the above problems, this application proposes a method. In order to overcome the system error caused by non-systematic external noise and crosstalk, as shown in Table 1, before each laser radar sensor is emitted, a delay of T wait , and then transmit. For each lidar sensor S i , T wait At two values ​​a i and b i Alternate between a i and b i It can be very small, in the order of milliseconds. The ai value is the T in Table 1 wait,a (ms), b i The value is T in Table 1 wait,b (ms). Any LiDAR sensor S i After receiving the echo, wait for a short delay T wait After each transmission, T wait In the corresponding T wait,a (ms) and T wait,b The modified signal transmission schedule is shown in Table 1. In Table 1, T lag It refers to the time when the sensor is scheduled to transmit after the start of a cycle; T wait It refers to the waiting time under the EERUF algorithm after the LiDAR sensor plans or prepares to launch. The EERUF (Error Eliminating Rapid Ultrasonic Firing) algorithm is an existing algorithm and will not be described here. fire It refers to the time from the start of a cycle to the actual emission of the LiDAR sensor.

[0087] Table 1 LiDAR sensor time cycle data table

[0088]

[0089]

[0090] Furthermore, an embodiment of the present application also provides a tunnel inspection method, which specifically includes: a frame is arranged on an autonomous mobile robot, and a plurality of laser radar sensors are arranged on the frame, the autonomous mobile robot is moved along the tunnel, detection data of each laser radar sensor is received, and a standard tunnel model and an updated tunnel model are generated according to the detection data, and a matching error is obtained by matching the standard tunnel model and the updated tunnel model. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is the deformation position.

[0091] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described 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 specification.

[0092] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A tunnel inspection robot, comprising a mobile robot, characterized in that: The mobile robot is provided with a frame, on which a plurality of laser radar sensors are provided, and further includes a data processing module, wherein the data processing module is used to receive detection data of each laser radar sensor, and generate a standard tunnel model and an updated tunnel model according to the detection data, and obtain a matching error by matching the standard tunnel model with the updated tunnel model. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformed position, and a mobile robot coordinate system ∑ R , the forward direction of the mobile robot is the Y axis, denoted as Y R , establish a Z axis pointing upward and an X axis perpendicular to the Z axis and pointing to the left of the Z axis on a plane perpendicular to the Y axis. The Z axis is denoted as Z R , the X-axis is denoted as X R , the Z axis, Y axis and X axis intersect at the origin O R , several laser radar sensors are located at the origin O R The X-axis on the left side is distributed along 180 degrees to the origin O R The right side of the X axis, The standard tunnel model and the updated tunnel model are both constructed by moving a mobile robot in a tunnel, specifically including the following methods: Establish world coordinate system∑ W , Xw is the X axis of the world coordinate system, Yw is the Y axis of the world coordinate system, Zw is the Z axis of the world coordinate system, S j represents the LiDAR sensor, is W Based on the time ti of the laser radar sensor S j The coordinates of a point are obtained by reading is Σ R Based on the lidar sensor S at time ti j The coordinates of the point obtained by the readings are Obtained according to the following formula: Among them, D ij It is a LiDAR sensor S j The reading at time ti, R j It is a LiDAR sensor S j Distance from coordinate origin O R The distance to point to Z R X on the left side of the axis R Axis is the reference axis, Φ j It is a LiDAR sensor S j The angle from the reference axis, Obtained by the following formula, θ iH It is time Y R The angle between Yw and θ iH Also X R The angle between Xw and YesS j The laser radar sensor is located in the mobile robot coordinate system Σ R X R The coordinates of the axes, YesS j The laser radar sensor is located in the mobile robot coordinate system Σ R Y R The coordinates of the axes, The tunnel model is constructed by the point coordinates of the world coordinate system. The method further comprises: determining the deformation size at the deformation position according to the detection data; The deformation position is an area C k , the deformation size is area C k The size of the area C k The width W, The width W is obtained by the following formula: Among them, S m It means the reading is in area C k Each lidar sensor, r m YesS m The readings and S m Distance from coordinate origin O R The sum of the distances, r m+1 It's S m+1 The readings and S m+1 Distance from coordinate origin O R The sum of the distances, β m Represents S m and S m+1 The angle between the two lidar sensors, D k YesS k The reading of the lidar sensor, S k The LiDAR sensor is the one located at the edge that can detect area C k The lidar sensor, α is S k The angle between the laser radar's transmitted signal and the recovered signal, the deformation size is area C k The size of the area C k The length L, The length L is obtained by the following formula: Ly vR -and uR -2d F , Among them, D um One of the S m The lidar sensor is at time t u When in area C k The front end reading, α is S m The angle between the laser radar's transmitted signal and the received signal, y uR is the mobile robot at time t u When in area C k The world coordinate system Σ of the front end W The Yw axis coordinate, y vR is the mobile robot at time t v When in area C k The world coordinate system Σ of the backend W The Yw axis coordinates below.

2. The tunnel inspection robot according to claim 1, characterized in that: The number of lidar sensors is 24.

3. The tunnel inspection robot according to claim 1, characterized in that: The deformation position is determined by the following method: (1)T U To update the tunnel model, T S is the standard tunnel model, T U Translate so that T U Starting point and T S The starting point coincides with (2) For T U Rotate so that T U The end point and T S The end points coincide with each other. (3) Calculate T S and T U The matching error between the corresponding points in the S and T U The Euclidean distance between corresponding points in is taken as the matching error.

4. The tunnel inspection robot according to claim 1, characterized in that: For each lidar sensor S j , before emitting the laser, delay a certain time T wait， And T wait It alternates between two values, ai and bi, where ai and bi are in the order of milliseconds, ai=25ms, and bi ranges from 0 to 46ms.

5. A tunnel inspection method, characterized in that: Using the tunnel inspection robot according to any one of claims 1 to 4, A frame is provided on the autonomous mobile robot, and a plurality of laser radar sensors are provided on the frame. The autonomous mobile robot is moved along the tunnel, and detection data of each laser radar sensor is received. A standard tunnel model and an updated tunnel model are generated according to the detection data. A matching error is obtained by matching the standard tunnel model with the updated tunnel model. If the matching error exceeds a predetermined threshold, the position corresponding to the matching error is a deformed position.

Images