Tracking control method of inspection robot, inspection robot and storage medium

Abstract

The invention discloses an inspection robot tracking control method, an inspection robot and a storage medium. The method comprises the following steps: acquiring path line image data and tracking identification information acquired by an inspection robot; based on the path line image data, generating a path line track for controlling the inspection robot to track; and obstacle detection is carried out between the tracking identification information and the path line track, and the inspection robot is controlled to execute avoidance under the condition that an obstacle exists. The technical problem that the tracking efficiency of the inspection robot is not ideal is solved.

Description

Technical Field

[0001] The present invention relates to the field of inspection robots, and in particular to a tracking control method for an inspection robot, an inspection robot and a storage medium. Background Art

[0002] The tracking sign navigation system is an automatic navigation system that uses special patterns, colors or markings to set signs on the predetermined route, uses sensors to identify the signs, and controls the direction and speed of the mobile device based on the identification results. However, the existing inspection sign system focuses on the rationality of data collection and task allocation, and there is still a lack of tracking sign systems designed for intelligent navigation.

[0003] In the inspection of cable tunnels, the existing cable tunnel inspection robots are mainly track-based, wheeled and quadrupedal, but each movement mode faces its inherent challenges. Specifically, track-based robots can only be used in open areas; wheeled robots have difficulty crossing complex terrains such as steps and steep slopes; quadrupedal robots are prone to losing balance and are difficult to design. In view of the above technical bottlenecks, the level of automation of robots is still insufficient when performing complex road inspection tasks, and it is difficult to achieve the ideal efficiency and stability.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a patrol robot tracking control method, a patrol robot and a storage medium, so as to at least solve the technical problem that the patrol robot has unsatisfactory tracking efficiency.

[0006] According to one aspect of an embodiment of the present invention, there is provided a method for controlling tracking of a patrol robot, comprising: acquiring path line image data and tracking identification information collected by the patrol robot; generating a path line trajectory for controlling the tracking of the patrol robot based on the path line image data; performing obstacle detection between the tracking identification information and the path line trajectory, and controlling the patrol robot to avoid obstacles when there are any.

[0007] Optionally, generating a path line trajectory for controlling the tracking of the inspection robot based on the path line image data includes: performing edge detection based on the path lines included in the path line image data to determine a path line contour, wherein the path line is a stripe marking used to indicate a direction of movement of the inspection robot; performing parameter mapping on the path line contour to determine a position and direction of the path line; and determining the path line trajectory based on the position and the direction.

[0008] Optionally, performing parameter mapping on the path line profile to determine the position and direction of the path line includes: performing parameter mapping on the path line profile to determine the vertical distance from the coordinate point included in the path line profile to a predetermined origin in the following manner;

[0009] rho=xcos(theta)+ysin(theta)

[0010] Among them, x is the horizontal coordinate of the coordinates of the point on the path line profile, y is the vertical coordinate of the coordinates of the point on the path line profile, theta is the direction angle of the path line profile, and rho is the vertical distance; based on the vertical distance from the coordinate point included in the path line profile to the predetermined origin, determine the peak point in the peak of the path line; based on the peak point, determine the position and the direction.

[0011] Optionally, controlling the inspection robot to perform avoidance includes: controlling the inspection robot to pause movement; determining the forward distance and forward turning angle that have been executed before the inspection robot pauses movement; reversely processing the forward distance to determine the backward distance; reversely processing the forward turning angle to determine the reverse turning angle; and controlling the inspection robot to perform avoidance by using the backward distance and the reverse turning angle.

[0012] Optionally, the method also includes: determining the pixel size of the tracking identification information collected by the inspection robot in the absence of obstacles; determining the interval distance between the inspection robot and the tracking identification information based on the focal length of the camera used by the inspection robot for image acquisition, the actual size of the tracking identification information, and the pixel size; in the case where there are multiple tracking identification information, determining the target identification information that matches the predetermined distance among the interval distances corresponding to the multiple tracking identification information; determining the operating status information of the inspection robot; and controlling the inspection robot to perform inspection movement based on the target identification information and the operating status information.

[0013] Optionally, determining the running status information of the inspection robot includes: obtaining the positioning information, angular velocity information, and acceleration information corresponding to each of the predetermined multiple axes of the inspection robot; determining the posture angle, current speed, and current position of the inspection robot based on the positioning information, angular velocity information, and acceleration information corresponding to each of the predetermined multiple axes; and correcting a predetermined motion model based on the posture angle, the current speed, and the current position to determine the running status information.

[0014] Optionally, generating a path line trajectory for controlling the tracking of the inspection robot based on the path line image data includes: determining multiple path points along the path of the inspection robot based on the path line image data; processing two adjacent points included in the multiple path points to determine an equidistant parameter, an arc length parameter, and a time parameter between the two adjacent points, wherein the equidistant parameter represents the distance between the two adjacent points, the arc length parameter represents the arc length of a sub-trajectory, and the time parameter represents the time interval between the two adjacent points; determining the sub-trajectory between the two adjacent points based on the equidistant parameter, arc length parameter, and time parameter between the two adjacent points; generating the path line trajectory based on the equidistant parameter, arc length parameter, and time parameter between the two adjacent points.

[0015] According to another aspect of an embodiment of the present invention, a patrol robot is provided, comprising: a head module, a plurality of joints, and at least one body module, and a tail module, the plurality of joints being respectively used to connect the head module, the at least one body module, and the tail module, the head module, the at least one body module, and the tail module being respectively provided with a longitudinal motor, the longitudinal motor being used to provide a driving force for the longitudinal movement of the patrol robot, the at least one body module, and the tail module being respectively provided with a transverse motor, the transverse motor being used to provide a driving force for the transverse movement of the patrol robot.

[0016] Optionally, the system also includes: the head module is provided with an image acquisition device, a laser positioning sensor, and a piezoelectric sensor, the laser positioning sensor is used to determine the environmental distance, and the piezoelectric sensor is used to sense the electrical signal generated by the deformation of the inspection robot; and / or, the first body module included in the at least one body module is provided with an electrochemical sensor, and the electrochemical sensor is used for gas detection; the second body module included in the at least one body module is provided with a power supply device; the third body module included in the at least one body module is provided with at least any one of the following: a temperature sensor, a humidity sensor and a sound sensor; and / or, the tail module is provided with a wireless transceiver, and the wireless transceiver is used for the inspection robot to interact with the cloud inspection background.

[0017] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is provided, wherein the non-volatile storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing any one of the inspection robot tracking control methods.

[0018] In the embodiment of the present invention, by acquiring the path line image data and tracking identification information collected by the inspection robot; generating the path line trajectory for controlling the inspection robot to track based on the path line image data; performing obstacle detection between the tracking identification information and the path line trajectory, and controlling the inspection robot to avoid obstacles when obstacles exist. The purpose of controlling the inspection robot to avoid obstacles during the tracking process is achieved, and the technical effect of improving the tracking efficiency of the inspection robot is achieved, thereby solving the technical problem of the unsatisfactory tracking efficiency of the inspection robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 is a flow chart of an optional inspection robot tracking control method provided according to an embodiment of the present invention;

[0021] Figure 2 is a schematic flow chart of an optional inspection robot tracking control method provided according to an embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of an optional inspection robot provided according to an embodiment of the present invention;

[0023] Figure 4 It is a structural block diagram of an optional inspection robot provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to an embodiment of the present invention, a method embodiment of tracking control of a patrol robot is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] Figure 1 is a flow chart of an optional inspection robot tracking control method according to an embodiment of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0028] Step S102, obtaining path line image data and tracking identification information collected by the inspection robot;

[0029] It can be understood that the inspection robot collects path line image data and tracking identification information. The path line is a stripe identification used to indicate the movement direction of the inspection robot. The tracking identification information mainly refers to QR code image data. The QR code is an image identification used to instruct the inspection robot to perform the inspection task.

[0030] Optionally, the two-dimensional code image data is a graphic composed of a series of specific black and white pixels arranged according to certain coding rules. The black and white pixels represent binary data, usually "dots" represent binary "1", and blanks represent binary "0". The two-dimensional code can store various types of data, such as numbers, letters, Chinese characters and other symbols, and has a certain degree of fault tolerance, and can be correctly read even if it is partially damaged.

[0031] Step S104, generating a path line trajectory for controlling the inspection robot to follow based on the path line image data;

[0032] It can be understood that path line extraction is used to provide data support for the inspection robot's tracking trajectory. An image processing-based algorithm can be used to identify the center of the path line, assist in determining which path the inspection robot follows, and make corresponding steering controls. Based on the extracted path line information, a control strategy can be designed to guide the inspection robot to travel along a predetermined path.

[0033] Optionally, before generating the path line trajectory followed by the inspection robot, the path line image data needs to be preprocessed first, which may include image binarization, edge detection, etc., to extract a clear path line contour.

[0034] In an optional embodiment, a path line trajectory for controlling the tracking of a patrol robot is generated based on the path line image data, including: performing edge detection based on the path lines included in the path line image data to determine the path line contour, wherein the path line is a stripe mark used to indicate the direction of movement of the patrol robot; performing parameter mapping on the path line contour to determine the position and direction of the path line; and determining the path line trajectory based on the position and direction.

[0035] It can be understood that the path line trajectory followed by the inspection robot is composed of the path line contour, position and direction. The path line contour provides the predetermined trajectory of the robot's movement, the position indicates the robot's current point on the path, and the direction ensures that the robot moves along the correct path. Through the above processing, the basic framework of the inspection robot's tracking route trajectory is constructed, ensuring that the robot can safely and accurately perform inspections on the cable track.

[0036] Optionally, the above-mentioned path line image data acquisition method can be multiple, for example, using the Canny edge detection algorithm to find the edge of the path line and extract the path line contour. The Canny edge detection algorithm is a classic image processing technology that aims to automatically detect edges in an image and achieve low error rate, high positioning accuracy and minimum response, that is, it can mark as many actual edges as possible while minimizing false detection and missed detection, and ensuring that each edge point corresponds to an actual edge point.

[0037] Optionally, the above-mentioned path line trajectory formation can be multiple, for example, using the Hough transform algorithm, mapping the path line contour to the parameter space, determining the position and direction of the path line, and forming the path line trajectory. Hough transform is an image processing technology used to identify objects of specific shapes in digital images, such as straight lines, circles, ellipses, etc. The shape of the object is detected by performing a voting mechanism in the parameter space, which is achieved by accumulating the local maximum in the space. The basic principle is to map the pixel points representing the geometric shape in the image to the parameter space, and determine the parameters of the geometric shape by finding the peak point in the parameter space.

[0038] In an optional embodiment, performing parameter mapping on the path line profile to determine the position and direction of the path line includes:

[0039] By performing parameter mapping on the path line profile, a vertical distance from a coordinate point included in the path line profile to a predetermined origin is determined;

[0040] rho=xcos(theta)+ysin(theta)

[0041] Among them, x is the horizontal coordinate of the coordinates of the point on the path line contour, y is the vertical coordinate of the coordinates of the point on the path line contour, theta is the direction angle of the path line contour, and rho is the vertical distance; based on the vertical distance from the coordinate point included in the path line contour to the predetermined origin, determine the peak point in the peak of the path line; based on the peak point, determine the position and direction.

[0042] It can be understood that by projecting the components of the x-direction and y-direction in the polar coordinates, the coordinates on the two-dimensional plane are transformed to obtain the vertical distance from the coordinate point in the path line contour to the predetermined point, and the peak point is determined based on the vertical distance, and then the position and direction of the inspection robot are determined. Through the relationship between the vertical distance and the peak point, the position of the inspection robot in the environment can be determined more accurately, reducing the positioning error caused by environmental changes.

[0043] Step S106, performing obstacle detection between the tracking identification information and the path line trajectory, and controlling the inspection robot to avoid obstacles if there are obstacles.

[0044] It can be understood that the inspection robot performs obstacle detection based on the tracking identification information and the path line trajectory. When there is an obstacle between the two, it is considered that avoidance processing needs to be performed.

[0045] Optionally, obstacle detection is performed by the obstacle unit. When the obstacle unit finds an obstacle in the path line trajectory and the two-dimensional code image data (i.e., tracking identification information), the movement of the inspection robot is immediately stopped and the control instruction issued by the instruction planning unit is followed.

[0046] The obstacle unit is used to analyze the obstacles that appear on the path line and the QR code, determine the type of obstacle, and call the control instructions issued by the instruction planning unit, calculate and issue control instructions to the inspection robot for returning to the original path. The instruction planning unit calls the calculation results of the path line recognition unit and the positioning unit, analyzes the motion trajectory of the inspection robot, determines the motion mode of the inspection robot based on the motion model, and issues control instructions to the inspection robot for movement. This method can identify obstacles in advance and predict the motion state of the obstacles, reserving sufficient time for subsequent identification and obstacle avoidance. It has high pertinence and success rate, and updates the robot's motion trajectory in real time to cope with the dynamic changes of obstacles and ensure the safe passage of the robot.

[0047] Optionally, the above-mentioned method of identifying obstacles can be multiple, for example, using a color segmentation algorithm to identify and segment areas of different colors by analyzing the color information in the image. According to the color of the obstacle, the type of obstacle is determined, including cables, fixed parts, and building materials. First, it is necessary to use a camera acquisition device to capture an image containing the obstacle, and convert the acquired image from the RGB color space to a color space more suitable for color segmentation, such as HSV (hue, saturation, brightness) or LAB (brightness, red, green, yellow, and blue); secondly, a color threshold range is set for each obstacle type, and the image is segmented using the set color threshold, which involves checking each pixel in the image to check whether the color falls within the color threshold range of a specific obstacle type. If so, the pixel is marked as part of the target obstacle. According to the connected domain (i.e., continuous area of ​​the same color) in the segmented image, the area, shape and other features of each area are calculated, and combined with the color information, a machine learning algorithm or a simple decision tree is used to determine which type of obstacle each area belongs to.

[0048] In an optional embodiment, controlling the inspection robot to perform avoidance includes: controlling the inspection robot to pause movement; determining the forward distance and forward turning angle that the inspection robot has executed before pausing movement; performing reverse processing on the forward distance to determine the backward distance; performing reverse processing on the reverse turning angle to determine the reverse turning angle; and controlling the inspection robot to perform avoidance by means of the backward distance and the reverse turning angle.

[0049] It can be understood that when the inspection robot detects an obstacle, it immediately stops moving, calculates the forward distance and turning angle of the inspection robot, and then reversely calculates the backward distance and reverse turning angle. Through precise calculation and adjustment, the robot can replan the path without contacting obstacles. The above processing helps to reduce the occurrence of abnormal operation of the inspection robot.

[0050] Optionally, the method for calculating the forward distance and the turning angle of the inspection robot at least includes: calculating the forward distance and the turning angle of the inspection robot, and then reversely calculating to obtain the backward distance and the reverse turning angle, wherein the formula for calculating the forward distance and the turning angle of the inspection robot is:

[0051] x'=x'0+d*cos(theta′0+phi / 2)

[0052] y'=y'0+d*sin(theta′0+phi / 2)

[0053] theta′=ttheta′0+phi

[0054] Among them, x'0, y'0, theta′0 represent the initial position and angle of the inspection robot respectively, d represents the forward distance of the inspection robot, phi represents the turning angle of the inspection robot, x, y represent the final position of the inspection robot, theta′ represents the final angle of the robot, and the obstacle unit replaces the forward distance d with the backward distance -d, and replaces the turning angle phi with the reverse turning angle -phi, and then calculates to obtain the backward distance and reverse turning angle.

[0055] In an optional embodiment, the method further includes: determining the pixel size of the tracking identification information collected by the patrol robot in the absence of obstacles; determining the interval distance between the patrol robot and the tracking identification information based on the focal length of the camera used by the patrol robot for image acquisition, the actual size of the tracking identification information, and the pixel size; in the case of multiple tracking identification information, determining the target identification information matching the predetermined distance among the interval distances corresponding to the multiple tracking identification information; determining the operating status information of the patrol robot; and controlling the patrol robot to perform patrol movements based on the target identification information and the operating status information.

[0056] It can be understood that the inspection robot has a QR code recognition unit, which identifies the QR code area based on the QR code image data received by the data receiving unit, estimates the distance between the inspection robot and the QR code based on the preset QR code size and the focal length of the inspection robot's camera, screens out the QR codes within the preset distance, decodes the QR codes, converts the information in the QR codes into control instructions, and controls the inspection robot to perform inspection movements. Through QR code technology, the inspection robot can automatically identify inspection points without human intervention, thereby speeding up the inspection speed and response time and reducing errors caused by improper human operation.

[0057] Optionally, the above-mentioned method for identifying the QR code area can be multiple, for example, a target recognition algorithm based on the OpenCV library, which provides a variety of traditional target recognition algorithms, including a method based on the Histogram of Oriented Gradients (HOG) and a classifier based on Haar features. HOG is a feature descriptor that captures shape information by calculating the distribution of gradient directions in a local area in an image. The Haar cascade classifier is a machine learning method for quickly identifying simple objects in an image. By constructing multiple layers of rectangular features (Haar features) and learning which feature combinations are most suitable for distinguishing between targets and backgrounds through a training process. With the development of deep learning technology, OpenCV has also begun to support the integration of deep learning-based target detection algorithms. For example, YOLOv3 (You Only Look Once version 3) can predict multiple bounding boxes and category probabilities through a single neural network in a single forward propagation. Traditional methods based on manual features (such as HOG and Haar cascade classifiers) are generally computationally efficient, but may not perform as well as deep learning methods in complex scenes. Deep learning methods such as YOLOv3 provide higher accuracy and are more efficient when dealing with diverse and challenging data sets.

[0058] Optionally, the estimating the distance between the inspection robot and the QR code at least includes: the QR code recognition unit estimates the distance according to the focal length of the camera of the inspection robot and based on a preset QR code size, and the estimation formula is as follows:

[0059] distance≡(QR_size*focal_length) / pixel_size

[0060] Among them, QR_size is the preset QR code size, focal_length is the focal length of the inspection robot's camera, and pixel_size is the pixel size of the QR code in the QR code image.

[0061] In an optional embodiment, the operating status information of the inspection robot is determined, including: obtaining the positioning information, angular velocity information, and acceleration information corresponding to a plurality of predetermined axes of the inspection robot; determining the posture angle, current speed, and current position of the inspection robot based on the positioning information, angular velocity information, and acceleration information corresponding to a plurality of predetermined axes; and correcting a predetermined motion model based on the posture angle, current speed, and current position to determine the operating status information.

[0062] It can be understood that to determine the running status information of the inspection robot, the state vector of the inspection robot is a multidimensional vector including the position, posture, speed, angular velocity and other information of the inspection robot in space. The positioning unit is required to obtain the longitude and latitude coordinate information of the inspection robot from the GPS (Global Positioning System) sensor data, and record the timestamp to obtain the position information of the inspection robot; obtain the angular velocity information of the inspection robot from the gyroscope data, measure the rotation speed and posture angle, and obtain the rotation speed of the inspection robot on each axis; obtain the acceleration information and tilt angle of the inspection robot on each axis from the accelerometer data, and obtain the linear acceleration of the inspection robot in each direction; and then use the covariance matrix of the inspection robot to describe the correlation and uncertainty between each element in the state vector. Each element in the covariance matrix represents the covariance between the corresponding state quantities, that is, the variance and correlation coefficient of the error, and updates the state according to the sensor measurement value, thereby improving the accuracy and reliability of the inspection robot.

[0063] Optionally, the state vector information of the inspection robot at least includes the posture angle, speed and position of the robot calculated using Euler integral. The positioning unit calculates the posture angle of the robot using Euler integral, and the formula is as follows:

[0064] θ(t)=θ(t-1)+ω(t)*Δt

[0065] Among them, θ(t) represents the current posture angle, θ(t-1) represents the posture angle at the previous moment, ω(t) represents the angular velocity at the current moment, Δt represents the time interval, and t is the mark of the moment;

[0066] The positioning unit uses Euler integration to convert acceleration into velocity, the formula is as follows:

[0067] v(t)=v(t-1)+(a(t)-g(t))*Δt

[0068] Where v(t) represents the linear velocity at the current moment, v(t-1) represents the linear velocity at the previous moment, a(t) represents the acceleration at the current moment, g(t) represents the gravitational acceleration at the current moment, and Δt represents the time interval;

[0069] The positioning unit uses Euler integration to estimate the position, the formula is as follows:

[0070] p(t)=p(t-1)+v(t)*Δt

[0071] Among them, p(t) represents the current position, p(t-1) represents the position at the previous moment, v(t) represents the speed at the current moment, and Δt represents the time interval.

[0072] Optionally, the positioning unit constructs a motion model of the inspection robot, combining the position, speed and angle of the inspection robot, and at least calculates the state vector and covariance matrix of the inspection robot. The formula for calculating the covariance is as follows:

[0073] P(t)=A(t-1)*P(t-1)*A(t-1)^T+Q(t-1)

[0074] Among them, P(t) represents the covariance matrix, A(t-1) represents the Jacobian matrix, and Q(t-1) represents the covariance matrix of the noise;

[0075] The positioning unit corrects the state vector and covariance matrix of the inspection robot based on the GPS sensor data. The formula is as follows:

[0076] K(t)=P(t)*H(t)^T*(H(t)*P(t)*H(t)^T+R(t))^-1

[0077] x(t)=x_hat(t)+K(t)*(z(t)-h(x_hat(t)))

[0078] P(t)=(IK(t)*H(t))*P(t)

[0079] Among them, K(t) represents the Kalman gain, H(t) represents the Jacobian matrix, R(t) represents the covariance matrix of the noise, z(t) represents the GPS data value, and h(x_hat(t)) represents the motion model.

[0080] In an optional embodiment, a path line trajectory for controlling the inspection robot to follow is generated based on the path line image data, including: determining multiple path points along the inspection machine's path based on the path line image data; processing two adjacent points included in the multiple path points to determine an equidistant parameter, an arc length parameter, and a time parameter between the two adjacent points, wherein the equidistant parameter represents the distance between the two adjacent points, the arc length parameter represents the arc length of a sub-trajectory, and the time parameter represents the time interval between the two adjacent points; determining a sub-trajectory between the two adjacent points based on the equidistant parameter, arc length parameter, and time parameter between the two adjacent points; generating a path line trajectory based on the equidistant parameter, arc length parameter, and time parameter between the two adjacent points.

[0081] It can be understood that by locating the path points passed by the inspection robot, the motion trajectories between all the path points are spliced ​​together according to the equidistant parameters, arc length parameters and time parameters to obtain the complete motion trajectory of the inspection robot. The equidistant and arc length parameters can better handle curves and turns, allowing the robot to adapt to changing environments and improve inspection efficiency and accuracy.

[0082] Optionally, the instruction planning unit can be used to construct a three-dimensional map of the cable tunnel and retrieve the trajectory of the path line in the path line recognition unit.

[0083] Optionally, the above method for constructing the motion trajectory of the inspection robot may be multiple, for example, defining the path points that the inspection robot passes through, which are P 1 , P 2 ,…,P n , calculate two adjacent path point pairs P i and P i+1 The parameter m between i , calculate the interpolation coefficient a i , b i , c i , d i , calculate P i and P i+1 In the parameter interval [m i , m i+1 ] represents the motion trajectory of the inspection robot in this interval. The motion trajectories between all path points are spliced ​​together to obtain the complete motion trajectory of the inspection robot.

[0084] Among them, two adjacent path points P i and P i+1 The parameter m between i Including equidistant parameters, arc length parameters and time parameters, the calculation formula of equidistant parameters is:

[0085] m i =i*(L / (n-1))

[0086] Where n represents the total number of path points, L represents the path length, i = 0, 1, 2, ..., n-1;

[0087] The calculation formula of arc length parameter is:

[0088] m i =si*(L / s n-1 )

[0089] Wherein, si, sn-1 represent the cumulative path length of the path points, L represents the path length, i = 0, 1, 2, ..., n-1;

[0090] The calculation formula of the time parameter is:

[0091] m i =m 0 +(m 1 -m 0 )*(s i / s n-1 )

[0092] Among them, m0 、m 1 、m i Indicates the time interval between path points, s i 、s n-1 represents the cumulative path length of the path points, i = 0, 1, 2, ..., n-1;

[0093] Among them, for each adjacent path point pair P i and P i+1 , we get the following equation:

[0094] S(m i )=a i +b i (m i -m i )+c i (m i -m i )^2+d i (m i -m i )^3=P i

[0095] S(m i+1 )=a i +b i (m i+1 -m i )+c i (m i+1 -m i )^2+d i (m i+1 -m i )^3=P i+1

[0096] S′(m i+1 )=b i +2c i (m i+1 -m i )+3d i (m i+1 -m i )^2=P i+1 '

[0097] S″(m i+1 )=2c i +6d i (m i+1 -m i )=P i+1 ″

[0098] The instruction planning unit solves the above equations to obtain the interpolation coefficient a in each interpolation interval. i , b i , ci , d i The value of

[0099] Among them, for each parameter interval [m i , m i+1 ], the interpolation function expression is:

[0100] S(m i )=a i +b i (m i -m i+1 )+c i (m i -m i+1 )^2+d i (m i -m i+1 )^3

[0101] Among them, a i , b i , c i , d i is the interpolation coefficient, m i is a parameter within this interval.

[0102] Optionally, the three-dimensional map of the cable tunnel can be constructed in multiple ways. For example, the starting point of the path line is selected as the origin to define the coordinate system of the three-dimensional map; the instruction planning unit defines the coordinate system of the inspection robot with the center point of the inspection robot as the origin; the instruction planning unit uses the rotation matrix to calculate the rotation relationship between the inspection robot coordinate system and the three-dimensional map coordinate system, and expresses the rotation relationship in the form of quaternion. Quaternion is a mathematical tool used to describe rotation in three-dimensional space, which is expressed as:

[0103] q=[w, x″i, y″j, z″k]

[0104] Among them, w is the scalar part, (x″, y″, z″) is the vector part, i, j, k are imaginary units, and the geometric meaning of the quaternion is to represent the rotation of w angles around the vector (x″, y″, z″);

[0105] The instruction planning unit uses quaternions to convert the inspection robot coordinate system into a three-dimensional map coordinate system. The formula is as follows:

[0106] P_map=q*P_local*q^-1

[0107] Among them, P_local represents a vector in the coordinate system of the inspection robot, P_map represents the position of the vector in the three-dimensional map coordinate system, q^-1 represents the inverse of the quaternion q, and * represents quaternion multiplication.

[0108] Through the above step S102, the path line image data and tracking identification information collected by the inspection robot are obtained; in step S104, based on the path line image data, a path line trajectory for controlling the inspection robot to track is generated; in step S106, obstacle detection is performed between the tracking identification information and the path line trajectory, and in the presence of obstacles, the inspection robot is controlled to avoid them.

[0109] The purpose of controlling the inspection robot to avoid obstacles during the tracking process can be achieved, and the technical effect of improving the tracking efficiency of the inspection robot is achieved, thereby solving the technical problem of the unsatisfactory tracking efficiency of the inspection robot.

[0110] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation mode, in step S1, the inspection robot tracking identification step includes a path line and a QR code. The path line is a stripe mark used to indicate the movement direction of the inspection robot, and the QR code is an image mark used to indicate that the inspection robot performs a detection task.

[0111] Step S2, the inspection robot collects cable tunnel image data, path line image data, QR code image data, GPS sensor data, gyroscope data and accelerometer data, and sends the data to the inspection background.

[0112] Step S3, the inspection background includes a data receiving unit, a path line recognition unit, a two-dimensional code recognition unit, a positioning unit, an instruction planning unit and an obstacle unit.

[0113] Step S4: data receiving unit, the data receiving unit is used to receive the data sent by the inspection robot and restore the data.

[0114] Step S5, the path line recognition unit uses an edge detection algorithm to find the edge of the path line based on the path line image data received by the data receiving unit, extracts the path line contour, uses a transformation algorithm to map the path line contour to the parameter space, determines the position and direction of the path line, and forms a trajectory of the path line.

[0115] Step S6, the QR code recognition unit identifies the QR code area based on the QR code image data received by the data receiving unit, estimates the distance between the QR code and the inspection robot, screens out the QR codes within a preset distance, decodes the QR codes, and converts the information in the QR code into control instructions.

[0116] Optionally, the estimating the distance between the inspection robot and the QR code at least includes: the QR code recognition unit estimates the distance according to the focal length of the camera of the inspection robot and based on a preset QR code size, and the estimation formula is as follows:

[0117] distance≡(QR_size*focal_length) / pixel_size

[0118] Among them, QR_size is the preset QR code size, focal_length is the focal length of the inspection robot's camera, and pixel_size is the pixel size of the QR code in the QR code image.

[0119] Step S7: The positioning unit constructs a motion model of the inspection robot, and calculates the current position, speed and angle of the inspection robot based on the GPS sensor data, gyroscope data and accelerometer data received by the data receiving unit.

[0120] Step S8, the obstacle unit is used to analyze the obstacles appearing on the path line and the QR code, determine the type of the obstacle, and retrieve the control instructions issued by the instruction planning unit, calculate and issue a control instruction for the inspection robot to return to the original path.

[0121] Step S9, the instruction planning unit retrieves the calculation results of the path line recognition unit and the positioning unit, analyzes the motion trajectory of the inspection robot, determines the motion mode of the inspection robot based on the motion model, and issues a control instruction for the inspection robot to move.

[0122] The optional implementations described above achieve at least the following effects: using identification technology to achieve high-precision navigation, provide positioning with lower errors, adapt to variable route shapes, reduce navigation costs, and have lower requirements for identification in tunnels, saving maintenance and management costs, and enabling faster automation of inspection tasks. It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowcharts, in some cases the steps shown or described can be executed in an order different from that shown here.

[0123] In another optional embodiment, taking the example of an inspection robot inspecting a normally operating cable tunnel, black and white striped path lines and a black and white QR code containing control instructions are preset in the cable tunnel. The instruction planning unit pre-completes the construction of a three-dimensional map of the cable tunnel, and defines the coordinate system of the three-dimensional map and the coordinate system of the inspection robot, and converts the inspection robot coordinate system into the three-dimensional map coordinate system so that the inspection robot can be accurately displayed in the three-dimensional map.

[0124] Start the inspection robot. Figure 2 is a schematic flow chart of an optional inspection robot tracking control method provided according to an embodiment of the present invention, such as Figure 2As shown, GPS signals are easily blocked and reflected by cable tunnels, resulting in positioning errors and signal loss. Therefore, relying solely on GPS sensors is not sufficient to provide high-precision positioning.

[0125] After the inspection robot is started, it first uses the visible light camera to collect cable tunnel image data, path line image data, QR code image data, GPS sensor data, gyroscope data and accelerometer data, and sends the data to the inspection background. The positioning unit then integrates the GPS sensor data with the estimated position, speed, and angle to locate and calibrate the position of the inspection robot to improve the problem of low positioning accuracy of the inspection robot.

[0126] Subsequently, the path line recognition unit identifies the trajectory of the path line based on the path line image data; the instruction planning unit defines the path points passed by the inspection robot based on the trajectory of the path line and obtains the motion trajectory of the inspection robot. The instruction planning unit then issues control instructions for the inspection robot to move based on the motion trajectory, so that the inspection robot moves along the path line.

[0127] During the movement, the QR code recognition unit estimates the distance between the inspection robot and the QR code, identifies the control instructions contained in the close-range QR code, and performs relevant detection operations according to the control instructions, including rotating the camera, getting close, etc., thereby eliminating the need for repeated high-precision calculations of the coordinates of the inspection robot inside the cable tunnel, saving computing resources in the inspection background.

[0128] In another optional embodiment, an inspection robot is used as an example to inspect a cable tunnel where a fault occurs.

[0129] In this embodiment, since various fault scenarios may occur in the cable tunnel, the inspection robot performs the following processing for various scenarios:

[0130] For this embodiment, obstacles on the path line and the QR code are the core problem that hinders the movement of the inspection robot. Since the inspection robot always uses the visible light camera to shoot and identify the path line and the QR code during operation, when obstacles appear on the path line and the QR code, the obstacle unit immediately stops the movement of the inspection robot and uses the formula according to the control command issued by the command planning unit:

[0131] x'=x'0+d*cos(theta′0+phi / 2)

[0132] y'=y'0+d*sin(theta′0+phi / 2)

[0133] theta′=ttheta′0+phi

[0134] Calculate the forward distance and turning angle of the inspection robot, then reversely calculate to obtain the backward distance and reverse turning angle, and issue a control command to the inspection robot to return to the original route, so that the inspection robot returns to the inspection background along the original route to avoid damage to the inspection robot caused by continuing to move forward.

[0135] In addition, the obstacle unit uses a color segmentation algorithm based on the path line image data and the QR code image data to determine the type of obstacle according to the color of the obstacle, including cables, fixed parts, building materials, etc.

[0136] A patrol robot is also provided in an embodiment of the present invention. The patrol robot provided in the embodiment of the present invention is introduced below.

[0137] Figure 3 is a structural block diagram of an inspection robot provided according to an embodiment of the present invention, such as Figure 3 As shown, the inspection robot tracking control method provided in any one of the above embodiments is applied to perform movement, and the system includes:

[0138] A head module 1, multiple joints 4, at least one body module 2, and a tail module 3. The multiple joints 4 are respectively used to connect the head module 1, at least one body module 2, and the tail module 3. The head module 1, at least one body module 2, and the tail module 3 are respectively provided with a longitudinal motor 41. The longitudinal motor 41 is used to provide a driving force for the longitudinal movement of the inspection robot. At least one body module 2 and the tail module 3 are respectively provided with a longitudinal motor 41. The longitudinal motor 41 is used to provide a driving force for the lateral movement of the inspection robot.

[0139] It can be understood that the inspection robot is connected by multiple joints 4 to the head module 1, the body module 2 and the tail module 3, and the driving force is provided by the longitudinal motor 41 and the transverse motor 42.

[0140] Alternatively, if Figure 3 As shown, the rear of the snake head module (i.e., the head module 1) is rotatably connected to a joint 4, the rear of the joint 4 is rotatably connected to a plurality of snake body modules (i.e., the body modules 2) which are interconnected through a plurality of joints 4, and the end of the joint 4 is connected to a snake tail module (i.e., the tail module 3).

[0141] Optionally, both sides of the snake head module, the snake body module and the snake tail module are rotatably connected with the driving wheels 5, the tails of the snake head module and the snake body module are fixedly connected with the longitudinal motor 41, the output end of the longitudinal motor 41 is fixedly connected with the longitudinal rotation axis of the joint 4, the front ends of the snake body module and the snake tail module are fixedly connected with the longitudinal motor 41, and the output end of the longitudinal motor 41 is fixedly connected with the transverse rotation axis of the joint 4. The inspection robot uses the joint 4 and the longitudinal motor 41 to realize the rotation on the plane, and the inspection robot uses the joint 4 and the transverse motor 42 to realize the rotation on the vertical plane, and then under the mutual cooperation of the longitudinal motor 41 and the transverse motor 42, the inspection robot has the function of crossing some obstacles.

[0142] As an optional embodiment, the head module 1 is provided with an image acquisition device, a laser positioning sensor, and a piezoelectric sensor. The laser positioning sensor is used to determine the environmental distance, and the piezoelectric sensor is used to sense the electrical signal generated by the deformation of the inspection robot; and / or, the first body module 2 included in at least one body module 2 is provided with an electrochemical sensor, and the electrochemical sensor is used for gas detection; the second body module 2 included in at least one body module 2 is provided with a power supply device; the third body module 2 included in at least one body module 2 is provided with at least any one of the following: a temperature sensor, a humidity sensor and a sound sensor; and / or, the tail module 3 is provided with a wireless transceiver, and the wireless transceiver is used for the inspection robot to interact with the cloud inspection background.

[0143] It can be understood that the snake head module is equipped with image acquisition equipment, laser positioning sensors, and piezoelectric sensors, the snake body module is equipped with electrochemical sensors, power supply equipment, temperature sensors, humidity sensors and sound sensors, and the tail module 3 is equipped with wireless transceiver equipment. The above technical configuration enables the inspection robot to efficiently and accurately complete remotely controlled inspection tasks.

[0144] Optionally, a visible light camera, an infrared camera and an ultraviolet camera are arranged at the top of the snake head module, and the bottoms of the visible light camera, the infrared camera and the ultraviolet camera are fixedly connected with a pan-tilt, and the bottom of the pan-tilt is rotatably connected to the snake head module. The sides of the visible light camera and the ultraviolet camera are provided with lighting lamps, and the bottoms of the lighting lamps are fixedly connected to the upper surface of the pan-tilt, and the front end of the snake head module is provided with a laser transmitter, a laser receiver and a piezoelectric sensor. The visible light camera, the infrared camera and the ultraviolet camera are used to collect video signals and transmit the video signals to the inspection background in real time; the pan-tilt includes bearings, gears, a drive motor and a microprocessor, and the microprocessor parses the control signal and controls the rotation of the pan-tilt, so that the visible light camera, the infrared camera and the ultraviolet camera can rotate horizontally from -170° to 170° and vertically from -30° to 90°. The lighting lamp is used to automatically turn on in dim light in the tunnel to provide sufficient light to ensure the normal operation of the visible light camera and the ultraviolet camera. The laser transmitter is used to emit a narrow and focused laser beam, and the laser receiver is used to receive the signal reflected by the laser beam. The laser transmitter and laser receiver are used to monitor the distance between the snake head module and the surrounding objects in real time and upload the data to the inspection background. The piezoelectric sensor is composed of piezoelectric material, motor and lead wire.

[0145] Optionally, a gas measuring device is provided at the top of the snake body module, and the gas measuring device includes several electrochemical sensors. A battery is provided inside the snake body module, and a coulomb meter is provided on the output end of the battery. A temperature sensor, a humidity sensor and a sound sensor are provided on the side of the snake body module. The number of snake body modules can be increased or decreased according to demand, so the number of batteries loaded by the inspection robot can be increased according to the length of the path line in the cable tunnel, and since the driving device is added at the same time, the moving speed of the inspection robot will not be greatly affected. The temperature sensor is used to detect the temperature of the inspection robot body and the tunnel environment in real time, and upload the data to the inspection background. The humidity sensor is used to detect the humidity of the tunnel environment in real time, and upload the data to the inspection background. The electrochemical sensor consists of a working electrode, a reference electrode, a counting electrode and an electrolyte, and is used to detect the concentration of various flammable gases and harmful gases in the tunnel in real time, such as gas, CO, etc. The sound sensor is used to detect the noise situation in the tunnel in real time, and upload the data to the inspection background. The coulomb meter is a programmable digital meter in the prior art, which is used to output the current battery power.

[0146] Optionally, the tail of the snake tail module is provided with a wireless transmitter, a wireless receiver and an antenna, and the inside of the snake tail module is provided with a core processor and a hard disk, the wireless transmitter includes a data input interface, a signal modulator and a radio frequency transmitter, and the wireless receiver includes a radio frequency receiver, a signal demodulator and a command output interface. The wireless transmitter includes a data input interface, a signal modulator and a radio frequency transmitter, the data input interface is used to receive the data collected by the snake head module and the snake body module, the signal modulator is used to convert the received data into a modulated signal suitable for wireless transmission, the radio frequency transmitter is used to convert the modulated signal into a radio frequency signal and amplify it, the wireless receiver includes a radio frequency receiver, a signal demodulator and a command output interface, the radio frequency receiver is used to receive the radio frequency signal from the inspection background, and convert the radio frequency signal into a modulated signal, the signal demodulator is used to convert the modulated signal into a control instruction that can be recognized by each module, and the command output interface is used to transmit the control instruction to each module, so as to realize the remote control of the inspection robot by the inspection background. The core processor is responsible for executing various algorithms and control tasks, and the hard disk is used to store various data in the operation of the inspection robot. Installing the core processor and the hard disk in the snake tail module can also start a certain protection effect.

[0147] Optionally, after completing the cable tunnel inspection, the inspection robot sends the data to the inspection background. The inspection background is provided with a data analysis unit, which performs intelligent analysis on the images and data sent back by the inspection robot and sends out an alarm signal when an abnormality occurs. The inspection background is also provided with a fault entry unit for recording faults that occur in the inspection background and the inspection robot.

[0148] Optionally, both sides of the snake head module, the snake body module and the snake tail module are provided with driving devices, the surfaces of the snake head module, the snake body module and the snake tail module are provided with shells, the insides of the snake head module, the snake body module and the snake tail module are provided with GPS sensors, gyroscopes and accelerometers, the driving devices include motors, reducers, drive wheels 5, brakes and uphill and downhill catchers, the shell is made of waterproof plastic, and the surface of the shell is filled with a protective coating to resist humid environments and chemical corrosion. In low-power mode, the inspection robot can be driven to move only by the driving devices of the snake head module and the snake tail module, making the inspection robot more suitable for completing long-distance cable tunnel inspection tasks. The GPS sensor is used to obtain position information, the gyroscope is used to measure the rotation speed and attitude angle, and the accelerometer is used to measure linear acceleration and tilt angle. The driving device includes an electric motor, a reducer, a driving wheel 5, a brake and an uphill and downhill catcher. The electric motor is used to generate power, the reducer is used to reduce the speed of the motor and increase the torque, the driving wheel 5 is used to realize the movement of the inspection robot, and the brake is used to realize the stopping and braking of the driving device, which is mainly achieved by electromagnetic tripping, friction, etc., to ensure that the inspection robot can stop safely or remain stable on the slope. The uphill and downhill catcher is used to detect and control the stability of the inspection robot moving on the slope.

[0149] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation manner. Figure 3 is a schematic diagram of an inspection robot provided according to an embodiment of the present invention, such as Figure 3 As shown, there is a head module 1, multiple joints 4, at least one body module 2, and a tail module 3, and the multiple joints 4 are used to connect the head module 1, at least one body module 2, and the tail module 3 respectively.

[0150] Figure 4 is a flowchart of the inspection robot provided according to an embodiment of the present invention, which is based on Figure 3 The inspection robot shown in the figure. Figure 4 The specific implementation steps are as follows:

[0151] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation mode. The inspection robot in this embodiment is a simple snake-like robot with a simple and stable structure and can be quickly assembled in a modular manner.

[0152] It should be noted that in the present embodiment, the number of snake body modules can be increased or decreased according to demand. Therefore, the number of batteries loaded on the inspection robot can be increased according to the length of the path line in the cable tunnel. Since the driving device is also added, the moving speed of the inspection robot will not be greatly affected. In addition, when the inspection robot is in low power consumption mode, it can drive the movement of the inspection robot only by relying on the driving devices of the snake head module and the snake tail module, making the inspection robot more suitable for completing long-distance cable tunnel inspection tasks.

[0153] This embodiment uses identification technology to achieve high-precision navigation, provide low-error positioning, adapt to changing route shapes, reduce navigation costs, and have low requirements for identification in tunnels, saving maintenance and management costs, and can more quickly realize the automation of inspection tasks.

[0154] In addition, in this embodiment, the snake head module is equipped with the following equipment:

[0155] The visible light camera, infrared camera and ultraviolet camera are used to collect video signals and transmit them to the inspection background in real time; the pan-tilt head includes bearings, gears, drive motors and microprocessors. The microprocessor parses the control signals and controls the rotation of the pan-tilt head, so that the visible light camera, infrared camera and ultraviolet camera can achieve horizontal rotation of -170° to 170° and vertical rotation of -30° to 90°.

[0156] The lighting is used to automatically turn on in dimly lit areas of the tunnel to provide sufficient light to ensure the normal operation of the visible light camera and ultraviolet camera.

[0157] The laser transmitter is used to emit a narrow and focused laser beam, and the laser receiver is used to receive the signal reflected by the laser beam. The laser transmitter and laser receiver are used to monitor the distance between the snake head module and surrounding objects in real time and upload the data to the inspection background.

[0158] Piezoelectric sensors consist of piezoelectric material, motor and leads.

[0159] The inspection robot uses joints 4 and longitudinal motors 41 to achieve rotation on the plane, and uses joints 4 and transverse motors 42 to achieve rotation on the vertical plane. Then, with the cooperation of longitudinal motors 41 and transverse motors 42, the inspection robot has the function of crossing some obstacles.

[0160] The snake module is equipped with the following equipment:

[0161] The temperature sensor is used to detect the temperature of the inspection robot body and the tunnel environment in real time, and upload the data to the inspection background.

[0162] The humidity sensor is used to detect the humidity of the tunnel environment in real time and upload the data to the inspection background.

[0163] The electrochemical sensor consists of a working electrode, a reference electrode, a counting electrode and an electrolyte, and is used to detect the concentration of various combustible gases and harmful gases in the tunnel in real time, such as gas, CO, etc.

[0164] When the harmful gas exceeds the safe concentration, the gas measuring device sends an alarm signal to the inspection background.

[0165] The sound sensor is used to detect the noise conditions in the tunnel in real time and upload the data to the inspection background.

[0166] The coulomb meter is a programmable digital meter in the prior art, which is used to output the current battery charge.

[0167] The tail module is equipped with the following equipment:

[0168] The wireless transmitter includes a data input interface, a signal modulator and a radio frequency transmitter. The data input interface is used to receive data collected by the snake head module and the snake body module. The signal modulator is used to convert the received data into a modulated signal suitable for wireless transmission. The radio frequency transmitter is used to convert the modulated signal into a radio frequency signal and amplify it. The wireless receiver includes a radio frequency receiver, a signal demodulator and a command output interface. The radio frequency receiver is used to receive the radio frequency signal from the inspection background and convert the radio frequency signal into a modulated signal. The signal demodulator is used to convert the modulated signal into a control command that can be recognized by each module. The command output interface is used to transmit the control command to each module to realize the remote control of the inspection robot by the inspection background.

[0169] The core processor is responsible for executing various algorithms and control tasks, and the hard disk is used to store various data during the operation of the inspection robot. Installing the core processor and hard disk in the tail module can also activate a certain protection effect.

[0170] The GPS sensor is used to obtain location information, the gyroscope is used to measure rotation speed and attitude angle, and the accelerometer is used to measure linear acceleration and tilt angle.

[0171] The driving device includes an electric motor, a reducer, a driving wheel 5, a brake and an uphill and downhill catcher. The electric motor is used to generate power, the reducer is used to reduce the speed of the motor and increase the torque, the driving wheel 5 is used to realize the movement of the inspection robot, and the brake is used to realize the stopping and braking of the driving device, which is mainly achieved by electromagnetic tripping, friction, etc., to ensure that the inspection robot can stop safely or remain stable on the slope. The uphill and downhill catcher is used to detect and control the stability of the inspection robot moving on the slope.

[0172] After completing the cable tunnel inspection, the inspection robot sends the data to the inspection background. The inspection background is equipped with a data analysis unit to perform intelligent analysis on the images and data sent back by the inspection robot, and send out an alarm signal when an abnormality occurs. The inspection background is also equipped with a fault entry unit to record faults that occur in the inspection background and the inspection robot.

[0173] In another optional embodiment, when an obstacle suddenly falls in front of the inspection robot, the visible light camera of the inspection robot does not have time to shoot and upload data, and the snake head module first contacts the obstacle. At this time, the piezoelectric sensor senses the contact with the obstacle and quickly stops the movement of the inspection robot, effectively reducing the loss caused by the collision with the obstacle, and sending an alarm signal to the inspection background;

[0174] When the inspection robot loses contact with the inspection background, it saves the inspection data and stops moving immediately.

[0175] When the sound sensor detects that the tunnel environmental noise exceeds 80dB, it sends an alarm signal to the inspection background.

[0176] When the lighting equipment inside the tunnel is damaged, the snake head module automatically turns on the lighting to provide sufficient light to ensure the normal operation of the visible light camera and ultraviolet camera;

[0177] When the concentration of harmful gas exceeds the safe level, the gas measuring device sends an alarm signal to the inspection background;

[0178] When the temperature of the inspection robot or tunnel is higher than the safe value, the temperature sensor sends an alarm signal to the inspection background.

[0179] This embodiment can comprehensively monitor the entire tunnel environment by comprehensively analyzing the data from various sensors, thereby improving the comprehensive performance of the inspection. Through real-time data monitoring and alarm functions, it can more flexibly cope with complex inspection environments, improve the overall emergency response capabilities, and reduce the risk of collision between the inspection robot and obstacles when encountering obstacles, and return the inspection robot to the inspection background along the original route to avoid damage to the inspection robot as it continues to move forward.

[0180] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.

[0181] The inspection robot provided by the embodiment of the present invention comprises a head module 1, multiple joints 4, at least one body module 2, and a tail module 3. The multiple joints 4 are respectively used to connect the head module 1, at least one body module 2, and the tail module 3. The head module 1, at least one body module 2, and the tail module 3 are respectively provided with longitudinal motors 41, and the longitudinal motors 41 are used to provide driving force for the longitudinal movement of the inspection robot. At least one body module 2 and the tail module 3 are respectively provided with longitudinal motors 41, and the longitudinal motors 41 are used to provide driving force for the lateral movement of the inspection robot. The purpose of comprehensively detecting the overall tunnel environment is achieved, thereby achieving the technical effect of improving the overall response capability of the inspection robot, and thus solving the technical problem that the overall response capability of the inspection robot is not ideal.

[0182] It should be noted that the optional or preferred implementation of this embodiment can refer to the relevant description in the embodiment, which will not be repeated here.

[0183] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0184] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0185] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0186] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0187] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0188] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0189] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0190] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0191] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0192] The above are only embodiments of the present invention and are 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 substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A tracking control method for an inspection robot, characterized in that: include: Obtain path line image data and tracking identification information collected by the inspection robot; Based on the path line image data, generating a path line trajectory for controlling the inspection robot to follow the track; Obstacle detection is performed between the tracking identification information and the path line trajectory, and when an obstacle exists, the inspection robot is controlled to perform avoidance.

2. The method according to claim 1, characterized in that The generating of a path line trajectory for controlling the inspection robot to follow a track based on the path line image data comprises: Performing edge detection based on the path line included in the path line image data to determine the path line profile, wherein the path line is a stripe mark used to indicate the movement direction of the inspection robot; Performing parameter mapping on the path line profile to determine the position and direction of the path line; Based on the position and the direction, the path line trajectory is determined.

3. The method according to claim 2, characterized in that The performing parameter mapping on the path line profile to determine the position and direction of the path line includes: By performing parameter mapping on the path line profile, a vertical distance from a coordinate point included in the path line profile to a predetermined origin is determined; rho=xcos(theta)+ysin(theta) Wherein, x is the abscissa of the coordinates of the point on the path line profile, y is the ordinate of the coordinates of the point on the path line profile, theta is the direction angle of the path line profile, and rho is the vertical distance; Determining a peak point among the peaks of the path line based on the vertical distance from the coordinate point included in the path line profile to a predetermined origin; Based on the peak point, the position and the direction are determined.

4. The method according to claim 1, characterized in that The controlling the inspection robot to perform avoidance includes: Controlling the inspection robot to pause movement; Determine the forward distance and forward turning angle that the inspection robot has executed before pausing the movement; Reverse processing is performed on the forward distance to determine the backward distance; Reversely processing the steering angle to determine a reverse steering angle; The inspection robot is controlled to perform avoidance by using the retreat distance and the reverse steering angle.

5. The method according to claim 1, characterized in that The method further comprises: In the absence of obstacles, determine the pixel size of the tracking identification information collected by the inspection robot; based on the focal length of the camera used by the inspection robot for image acquisition, the actual size of the tracking identification information, and the pixel size, determine the interval distance between the inspection robot and the tracking identification information; In the case where there are multiple tracking identification information, determining the target identification information matching the predetermined distance among the interval distances respectively corresponding to the multiple tracking identification information; Determining the operating status information of the inspection robot; Based on the target identification information and the operating status information, the inspection robot is controlled to perform inspection movement.

6. The method according to claim 5, characterized in that The step of determining the operating status information of the inspection robot comprises: Obtaining positioning information, angular velocity information, and acceleration information corresponding to a plurality of predetermined axes of the inspection robot; Based on the positioning information, the angular velocity information, and the acceleration information corresponding to the predetermined multiple axes, determine the posture angle, current velocity, and current position of the inspection robot; Based on the posture angle, the current speed, and the current position, the predetermined motion model is corrected to determine the running state information.

7. The method according to any one of claims 1 to 6, characterized in that The generating of a path line trajectory for controlling the inspection robot to follow a track based on the path line image data comprises: Determining a plurality of path points along the path of the inspection machine based on the path line image data; Processing two adjacent points included in the multiple path points to determine an equidistance parameter, an arc length parameter, and a time parameter between the two adjacent points, wherein the equidistance parameter represents the distance between the two adjacent points, the arc length parameter represents the arc length of the sub-trajectory, and the time parameter represents the time interval between the two adjacent points; Determine a sub-trajectory between the two adjacent points based on an equidistance parameter, an arc length parameter, and a time parameter between the two adjacent points; The path line trajectory is generated based on the equidistant parameter, arc length parameter, and time parameter between the two adjacent points.

8. A patrol robot, characterized in that: The inspection robot tracking control method according to any one of claims 1 to 7 is used to perform movement, comprising: A head module, multiple joints, at least one body module, and a tail module, the multiple joints are respectively used to connect the head module, the at least one body module, and the tail module, the head module, the at least one body module, and the tail module are respectively provided with longitudinal motors, the longitudinal motors are used to provide driving force for the longitudinal movement of the inspection robot, the at least one body module and the tail module are respectively provided with transverse motors, the transverse motors are used to provide driving force for the transverse movement of the inspection robot.

9. The inspection robot according to claim 8, characterized in that: The head module is provided with an image acquisition device, a laser positioning sensor, and a piezoelectric sensor, wherein the laser positioning sensor is used to determine the environmental distance, and the piezoelectric sensor is used to sense the electrical signal generated by the deformation of the inspection robot; and / or, The first body module included in the at least one body module is provided with an electrochemical sensor, and the electrochemical sensor is used for gas detection; the second body module included in the at least one body module is provided with a power supply device; the third body module included in the at least one body module is provided with at least any one of the following: a temperature sensor, a humidity sensor and a sound sensor; and / or, The tail module is provided with a wireless transceiver device, and the wireless transceiver device is used for the inspection robot to interact with the cloud inspection background.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the inspection robot tracking control method described in any one of claims 1 to 7.

Images