A line patrol navigation robot and navigation control method

Abstract

The present invention belongs to the field of automated positioning and navigation technology, and provides a line-patrolling navigation robot and navigation control method. The robot includes a visual navigation module that causes the robot to travel along an indicator light strip, and an IMU inertial navigation module that implements motion control of the robot based on IMU inertial navigation when the visual navigation module fails to obtain detection line information. The visual navigation module includes a line detection module that obtains the offset S and deflection angle θ of the detection line in the robot's world coordinate system. The present invention is not affected by complex ground conditions and has relatively good adaptability in indoor environments with a large number of people and other mobile vehicles. The combination of an infrared camera and an infrared filter is used to capture infrared light strip line images, making visual navigation relatively less affected by ambient light sources. The intermittent nature of the IMU inertial navigation process is utilized to avoid the cumulative error problem inherent in inertial navigation.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic positioning and navigation, and specifically relates to a comprehensive line patrol navigation robot and a navigation correction method that uses visual navigation as a main method and IMU inertial navigation as a supplement by arranging mutually spaced linear light strips on indoor roofs. Background Art

[0002] With the development of automated positioning and navigation technology, indoor mobile robot navigation technology is becoming increasingly mature, compared to outdoor GPS (Global Positioning System) vehicle navigation technology. The market mechanism is becoming more favorable, and its application prospects are becoming increasingly broad. For example, in automated production workshops in industrial environments, catering corridors in commercial environments, and logistics delivery channels in medical institutions, there is a growing demand for intelligent robots that can automatically travel back and forth along planned fixed routes. This replaces manual labor, saves labor costs, and improves production or transportation efficiency. Therefore, it is extremely necessary to develop a line-patrol navigation method suitable for indoor applications that is low-cost, simple, and easy to implement.

[0003] In response to the above needs, there are currently many relevant solutions at home and abroad.

[0004] There are currently two main types of line patrol robots:

[0005] One approach is to use AGV navigation robots based on magnetic strips, which offer the advantages of flexible path planning and relatively accurate navigation positioning. However, traditional AGV navigation robots with magnetic strips require laying magnetic strips on the navigation surface and then applying protective tape. Magnetic strips themselves have poorer wear resistance than the ground, making them suitable for installation in dry, clean environments. Therefore, they are well-suited for neat, automated industrial production workshops. However, magnetic strip navigation solutions are less adaptable in scenarios with high traffic volumes or other mobile vehicles, where the surface is prone to wear and tear.

[0006] Another type of robot is a line-following robot based on vision. While its navigation accuracy is relatively high, it can be affected by poor visual image quality in indoor environments with complex or dim lighting. Summary of the Invention

[0007] To address the above-mentioned drawbacks, the present invention proposes a line-patrolling navigation robot and navigation control method. The robot is relatively low-cost, less affected by ambient light, and better adaptable to indoor environments where floors are prone to wear and tear. To achieve the above-mentioned objectives, the present invention adopts the following specific technical solutions:

[0008] A line patrol navigation robot includes a visual navigation module for causing the robot to move along an indicator light strip, and an IMU inertial navigation module for implementing motion control of the robot based on IMU inertial navigation when the visual navigation module fails to obtain detection line information;

[0009] The visual navigation module includes a straight line detection module, which is used to obtain the offset S and deflection angle θ of the detection line in the world coordinate system of the robot car.

[0010] Preferably, the indicator light strip is placed in a middle position above the navigation road.

[0011] Preferably, the visual navigation module further comprises a camera acquisition module for acquiring the light band straight line generated by the indicator light band, and an image processing module for processing the acquired light band straight line to obtain a detection straight line.

[0012] Preferably, it also includes a main control module and a drive module;

[0013] The main control module is used to send the obtained linear velocity and angular velocity data of the robot car to the drive module, and the drive module enables the robot car to patrol the line.

[0014] Preferably, the indicator light band may be an ultraviolet light source, a visible light source or a near-infrared light source.

[0015] A control method for a line patrol navigation robot comprises the following steps:

[0016] S1. Obtain the slope k and offset b of the detection line in the new pixel coordinate system;

[0017] S2. Establish the world coordinate system of the robot car and use formula (1) to calculate the offset S and deflection angle θ of the detection line in the world coordinate system of the robot car:

[0018]

[0019] Among them, K is the actual distance represented by the unit pixel;

[0020] s is the intercept of the detection line in the y1-axis direction in the new pixel coordinate system;

[0021] S3. Use the motion control model to calculate and adjust the yaw angular velocity w0 and linear velocity v0 to achieve line-patrolling navigation control of the robot car.

[0022] Preferably, step S1 includes:

[0023] S101. In a traditional pixel coordinate system, clipping a light band straight line image formed by the indicator light band, converting the RGB image into a grayscale image and performing binarization processing, optimizing the edge straight line in the binarized image to obtain a detection straight line;

[0024] S102, filter out the detection lines whose slopes exceed the threshold, and obtain the set of two end points of the further optimized detection lines, opti_Lines.

[0025] The number of detected lines is n, and the coordinate set of the two end points of the detected lines is Lines = (x i1 ,y i1 ,x i2 ,y i2 ) i＝1,2,...,n ;

[0026] S103, using formula (2) to obtain the coordinates of each point of opti_Lines in the new pixel coordinate system:

[0027]

[0028] Where x=(x i1 ,x i2 ) i＝1,2,...,n ,

[0029] y=(y i1 ,y i2 ) i＝1,2,...,n ,

[0030] width×height is the default resolution of the camera;

[0031] S104 , using the opti_Lines point set as fitting data points for the detection line, calling a line fitting function to obtain the slope k and offset b of the detection line in the new pixel coordinate system.

[0032] Preferably, step S3 is divided into four cases:

[0033] When S31, θ>0 and T<0, the robot car goes straight and turns left. At this time, the angular velocity The radius of rotation r is obtained by the following formula:

[0034]

[0035] Where T is the intercept of the detection line on the x-axis of the robot car's world coordinate system;

[0036] N is the distance from the optical center a of the camera to the y-axis of the world coordinate system of the robot car;

[0037] L is the distance from the center o of the driving wheel of the robot car to the optical center a of the camera in the y-axis direction of the world coordinate system of the robot car;

[0038] v0 is the known linear velocity of the robot car;

[0039] When S32, θ<0, T>0, the robot car goes straight and turns right. The calculation method of angular velocity w0 is the same as S31;

[0040] When S33, θ>0, T>0, the robot car needs to go straight and turn left first. Given the angular velocity w1, the turning time t1 is controlled so that w1t1=π-2|θ|. After turning left, it enters S32.

[0041] When S34, θ<0, T<0, the robot car needs to go straight and turn right first. The angular velocity w2 is known, and the turning time t2 is controlled so that w2t2=π-2|θ|. After turning right, it enters S31.

[0042] Preferably, when no detection line information is obtained, the method further includes step S4:

[0043] The main control module uses the IMU angle data of the IMU inertial navigation module received, and uses this angle direction as the standard heading for inertial navigation, so that the robot car continues to move in a straight line until the robot car re-enters the visual navigation stage.

[0044] The present invention can achieve the following technical effects:

[0045] 1. The present invention uses the straight line formed by the infrared light band on the indoor roof for visual line patrol. Compared with the AGV magnetic strip located on the ground, it is not affected by the complex ground conditions and has better adaptability in indoor environments with a large number of people and other mobile vehicles.

[0046] 2. A combination of an infrared camera and an infrared filter is used to capture linear images of the infrared light band, filtering out the visible light source and most other stray light in the room. This not only makes the visual navigation relatively less affected by the ambient light source, but also ensures that the infrared light source will not affect the visible light lighting in the room.

[0047] 3. There is no need to use expensive hardware or sensors, such as lidar, which makes the cost of the robot car relatively low.

[0048] 4. There is no need to lay magnetic strips or stick protective magnetic tape along the entire navigation path. Only infrared light strips need to be arranged at appropriate intervals, which reduces installation and maintenance costs.

[0049] 5. The intermittent nature of the IMU inertial navigation process and the relatively short distance of each navigation avoid the inherent cumulative error problem of inertial navigation, making navigation relatively accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a system framework diagram of a line patrol navigation robot and a navigation correction method according to an embodiment of the present invention;

[0051] Figure 2 is a camera visual imaging model according to an embodiment of the present invention;

[0052] Figure 3 is an image processing flow chart of an embodiment of the present invention;

[0053] Figure 4 is a schematic diagram of detecting straight lines in a new pixel coordinate system according to an embodiment of the present invention;

[0054] Figure 5 1 is a schematic diagram comparing detection lines in the world coordinate system and the new pixel coordinate system according to an embodiment of the present invention;

[0055] Figure 6 This is a world coordinate system motion control model of a robot car when straight line detection θ>0 and T<0 in a world coordinate system according to an embodiment of the present invention;

[0056] Figure 7 This is a world coordinate system motion control model of a robot car when straight line detection θ<0, T<0 according to an embodiment of the present invention;

[0057] Figure 8 This is a world coordinate system motion control model of a robot car when straight line detection θ>0 and T>0 according to an embodiment of the present invention;

[0058] Figure 9 This is a world coordinate system motion control model of a robot car when straight line detection θ<0, T<0 according to an embodiment of the present invention;

[0059] Figure 10 1 is a schematic diagram of an IMU inertial navigation system according to an embodiment of the present invention;

[0060] Figure 11 FIG. 4 is a flow chart of cruise control according to an embodiment of the present invention.

[0061] Reference numerals:

[0062] Indicator light strip 1, camera 2, robot car 3, gyroscope 4. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not constitute a limitation of the present invention.

[0064] The present invention aims to provide a line-following navigation robot and a navigation correction method. This robot uses visual navigation and, when the robot travels outside the visual navigation control range, utilizes IMU inertial navigation to control the robot's line-following navigation. The following describes the line-following navigation robot and navigation correction method provided by the present invention in detail, using specific embodiments.

[0065] Reference Figure 1 In the system framework shown, under the guidance of the visual navigation module, the robot car 3 moves along the indicator light strip 1. When the visual navigation module does not obtain the detection straight line information of the indicator light strip 1, that is, when the movement position of the robot car 3 is in the middle part of the two straight lines of the light strip, the IMU inertial navigation module is used to control the movement of the robot car 3.

[0066] The visual navigation module includes a camera acquisition module, an image processing module and a straight line detection module. The camera acquisition module captures the light band straight line generated by the indicator light band 1. After processing by the image processing module, the edge of the detection straight line is extracted. The straight line detection module is used to obtain the offset S and deflection angle θ of the detection straight line in the world coordinate system of the robot car.

[0067] In a preferred embodiment of the present invention, the indicator light strip 1 is arranged longitudinally in the middle position of the roof above the indoor navigation road. The indicator light strip 1 adopts an infrared linear light strip with a wavelength of 850nm. Each section of the light strip is about 1m long, the required power supply voltage is 12V, and the center interval of each section of the light strip is about Q meters. Q can be appropriately adjusted according to the actual application scenario.

[0068] Robotic vehicle 3 follows the principle of right-hand traffic. To ensure that the image of the light band is as straight as possible within the fixed field of view of the camera acquisition module, camera 2 is fixed to the left of the center of the upper end of robot vehicle 3 using a fixed bracket during installation, with camera 2 facing vertically towards the roof. An 820±30nm infrared filter is also fixed above camera 2 to filter out visible light and other stray light.

[0069] In another embodiment of the present invention, the indicator light band 1 can also be replaced by other near-infrared light bands with a wavelength above 760nm, and an infrared filter of the corresponding wavelength can be used to filter the visible light source and other stray light; if the application scenario is an unmanned environment, to avoid the harmful effects of ultraviolet light sources on humans, an ultraviolet light band with a wavelength below 400nm can also be used, and an ultraviolet filter of the corresponding wavelength can be used to filter other light sources.

[0070] The IMU inertial navigation module hardware uses a six-axis gyroscope, which is fixed at the center of gravity of the robot car 3. It is used to realize the motion control of the robot car 3 based on the IMU inertial navigation when the visual navigation module fails to obtain the detection line information;

[0071] In a preferred embodiment of the present invention, Figure 1 , also includes a main control module and a driving module. The main control module is used to send the obtained linear velocity and angular velocity data of the robot car 3 to the driving module. The driving module controls the wheel movement of the robot car 3 based on the differentiator principle.

[0072] like Figure 11 As shown in the line patrol control flow chart, the robot car 3 produces cumulative errors when traveling in a straight line, resulting in yaw. At this time, it enters the heading correction mode. The image of the straight line of the near-infrared light band on the roof is obtained by camera 2, and the slope k and offset b of the detection line are processed. The offset S and deflection angle θ of the detection line in the world coordinate system of the robot car are obtained from this. The motion control models for four cases are established, and the line patrol navigation control of the robot car 3 is realized by calculating and adjusting the yaw angular velocity w0 and the known linear velocity v0.

[0073] In a preferred embodiment of the present invention, in step S1, the hardware of the image processing module adopts Raspberry Pi 4B, and processes the image through the Python software platform on the hardware and the corresponding version of the OpenCV library installed.

[0074] The image captured by camera 2 in step S101 is a straight line image formed by the near-infrared light band of the indoor roof. Based on the OpenCV library, the captured video image is processed, referring to Figure 3 The image processing module shown obtains the flow chart of detecting straight line edges:

[0075] First, crop the image. In the traditional pixel coordinate system, the image between the line y = height / 4 and the line y = 3height / 4 is cropped. This means that the video image (width × height) is cropped to width × height / 2. This captures the image's focus interval, reducing the image processing burden and increasing the frame rate without sacrificing image clarity.

[0076] Secondly, the cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) function in the OpenCV library is used to convert the intercepted image represented by the three-color channel mode of R (red), G (green), and B (blue) into an image represented by a single-channel grayscale.

[0077] Again, call the binarization function cv2.threshold(gray,100,255,cv2.THRESH_BINARY) to extract the pixels with grayscale values greater than 100 in the grayscale image and change their grayscale values to 255, that is, to white.

[0078] Finally, the cv2.Canny() function is called to extract the edges of the binarized image based on the Canny operator, that is, to obtain the edges of the white straight lines in the image.

[0079] In step S102, based on the Hough line detection method, the cv2.HoughLinesP() function is called to obtain the detected lines in the image after Canny edge detection. When calling the function, important parameters are set so that the detected lines meet the following three conditions:

[0080] a. The Hough space intersection points required for the detected straight line are at least 40 intersection points of the curves;

[0081] b. The minimum number of points to form a straight line is 30. Lines with insufficient points will be discarded.

[0082] c. The maximum distance between points on a straight line is 5. If the distance is greater than this, they will be considered as two straight lines;

[0083] The function output is a set of coordinates of the end points of multiple detection lines within the line of sight.

[0084] Further optimize the detected straight lines and filter out those with excessively large slopes:

[0085] Assume that the number of detection lines is n, and the coordinate set of the two end points of the detection line is Lines = (x i1 ,y i1 ,x i2 ,y i2 ) i＝1,2,...,n ;

[0086] When x i1 =x i2 When the detected lines are not vertical lines, substitute Lines into formula (4) to obtain the set of deviations of the slopes of each line dev_Slope1,

[0087]

[0088] When x i1 =x i2 When the detected lines are vertical lines, Lines is substituted into formula (5) to obtain the set of deviations of each line dev_Slope2,

[0089]

[0090] The deviation of each straight line is then compared with the set threshold value dev_Treshold = 0.1, and the straight lines exceeding the threshold value are removed to obtain the set of optimized straight line end point coordinates opti_Lines.

[0091] In step S103, in order to conveniently represent the situation when the detected straight line is vertical (to avoid the slope k being infinite), relative to the traditional pixel coordinate system, the following is established: Figure 4 The new pixel coordinate system is shown:

[0092] The center of the original image is the origin o1 of the new pixel coordinate system, the straight line through o1 horizontally to the right is the y1 axis, the y1 axis direction is right, the straight line through o1 vertically downward is the x1 axis, the x1 axis direction is downward,

[0093] The coordinates of each point in opti_Lines are transformed according to formula (2) to form the coordinates in the new pixel coordinate system:

[0094]

[0095] Use the new point set as the fitting data points of the target line and call the line fitting function optimize.curve.fit to obtain the slope k and offset b of the detection line in the new pixel coordinate system.

[0096] In another embodiment of the present invention, the camera acquisition module needs to calibrate the actual distance represented by a single pixel in the image before acquiring the image, and establish Figure 2 The camera visual imaging model shown uses a high-definition, distortion-free USB infrared camera (ignoring minor distortion), assumes the default resolution is width × height, the focal length of camera 2 is f1, the distance from camera 2 to the roof light strip is f2, the actual straight line length of the light strip within the field of view is w2, and the imaging straight line length is w1. Based on the pinhole imaging model, we can obtain equation (6):

[0097]

[0098] Since the distance f2 between camera 2 and the roof light strip is a fixed value, the lens of camera 2 faces the roof to shoot a straight infrared light strip with a known length of L0. Then, the pixel length PL of the straight line in the image is obtained through the screenshot software. Finally, the actual distance K represented by the unit pixel can be calculated using the principle of formula (6):

[0099]

[0100] In a preferred embodiment of the present invention, in step S2, the vision-based motion control method of the robot car 3 requires the use of the main control module and the drive module in the back-end system control module, wherein the main control module hardware adopts the STM32F407 development board, and the drive module adopts the wheel hub servo motor driver.

[0101] First, the STM32F407 obtains the detection line slope k and offset b data sent by the Raspberry Pi 4B through network communication;

[0102] Secondly, establish Figure 6 、 Figure 7 、 Figure 8 、 Figure 9 The world coordinate system motion model of the robot car shown takes the center of the line connecting the driving wheels of the robot car 3 as the origin o, the forward direction of the robot car 3 as the positive direction of the y-axis, the straight line passing through point o and perpendicular to the y-axis as the x-axis, and the left direction of the robot car 3 is the positive direction of the x-axis.

[0103] In addition, in the world coordinate system, the clockwise direction of the x-axis is the positive direction of the deflection angle θ, and the counterclockwise direction is the negative direction.

[0104] Continue to refer to Figure 4 , let the distance s between the intersection of the detection line and the y1 axis and the origin o1 in the new pixel coordinate system. Using formula (1), the slope k and offset b of the detection line in the new pixel coordinate system are used to calculate the offset S and deflection angle θ of the detection line in the world coordinate system of the robot car.

[0105]

[0106] During the conversion calculation, since camera 2 faces the roof, the ground projection of the actual roof infrared light band line in the robot car's world coordinate system and the infrared light band line in the new pixel coordinate system are reversed left to right, such as Figure 5 shown.

[0107] Assuming that the linear velocity of the robot car 3 is v0, when the robot car 3 yaws, it is necessary to calculate the angular velocity at this time based on the detection line and the motion control model to correct the heading.

[0108] In a preferred embodiment of the present invention, step S3 is divided into four cases:

[0109] like Figure 6 As shown in the figure, when θ>0 and T<0, the influence of the linear velocity direction on the motion component in the x-axis direction is ignored. The center o of the line connecting the driving wheels o1 and o2 of the robot car 3 moves at an angular velocity w0 and reaches the position of point o′. That is, the center of the driving wheel is located on the detection line and the x-axis of the robot car's world coordinate system is perpendicular to the detection line.

[0110] Assume that the rotation radius from point o to o' is r, the distance from point o to the optical center a of the front camera 2 of the robot car 3 is L, the distance from point o to the intersection b of the detection line and the robot car's world coordinate system on the x-axis is T, the distance from point a to the y-axis is N, and the distance from point a to the detection line in the x-axis direction is S. The values of r and T are calculated using the following formulas:

[0111]

[0112] According to the relationship between linear velocity and angular velocity Substituting r into the equation, we can get the angular velocity w0. At this time, the robot car 3 moves straight and turns left.

[0113] When θ<0, T>0, such as Figure 7 As shown, Figure 6 Similarly, the angular velocity w0 can be calculated. At this time, the robot car 3 goes straight and turns right.

[0114] When θ>0, T>0, such as Figure 8 As shown, the robot car 3 needs to go straight and turn left first, that is, the two wheels o1 and o2 reach o1' and o'2. The angular velocity w1 is known, and the turning time t1 is controlled so that w1t1=π-2|θ|. After turning left, θ<0, T>0, and it enters Figure 7 In the case of , you need to go straight and turn right again. The calculation method of angular velocity w0 is the same as Figure 7 Same thing.

[0115] When θ<0, T<0, such as Figure 9 As shown, the robot car 3 needs to go straight and turn right first, that is, the two wheels o1 and o2 reach o1' and o'2. The angular velocity w2 is known, and the turning time t2 is controlled so that w2t2=π-2|θ|. After turning right, θ>0, T<0, and it enters Figure 6 In the case of , you need to go straight and turn left, the angular velocity w0 is calculated by the same method as Figure 6 Same thing.

[0116] When T=0 or θ=0, there will always be errors in the movement of the robot car 3. After maintaining the moving state for a period of time, it will inevitably enter the state of Figures 6 to 9 Choose one of the four motion conditions, and then calculate the linear velocity and angular velocity according to the above method.

[0117] Finally, the STM32F407 sends the linear velocity v0 and angular velocity w0 to the robot car 3 drive controller through the serial port to control the robot car 3 to patrol the line.

[0118] In a preferred embodiment of the present invention, the hardware used in the motion control method of the robot car 3 based on IMU inertial navigation includes a main control module based on STM32F407 and an IMU inertial navigation module mainly based on the gyroscope 4.

[0119] like Figure 10As shown, gyroscope 4 is fixed to the center of gravity of robot car 3, with its z-axis perpendicular to the horizontal ground and its x- and y-axes parallel to the robot car's world coordinate system. Gyroscope 4 transmits the IMU angle to the STM32F407 in real time via the serial port. When robot car 3 reaches the middle of the two infrared light lines, the system cannot detect the line. Therefore, the Raspberry Pi 4B sends an interval position mark signal to the STM32F407. Upon receiving the signal, the STM32F407 stores and records the IMU angle data at that time. It then uses this IMU angle direction as the standard heading for inertial navigation, allowing robot car 3 to continue traveling in a straight line. If the system detects the infrared light line again and transmits the angular velocity and linear velocity, the STM32F407 clears the stored IMU angle data to prepare for the next recording. At this point, robot car 3 re-enters the visual navigation phase.

[0120] Because the IMU inertial navigation process is intermittent and the distance of each navigation is relatively short, the inherent cumulative error problem of inertial navigation is avoided and the navigation is relatively accurate.

[0121] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0122] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

[0123] The above specific embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A line patrol navigation robot, characterized in that: It includes a visual navigation module that enables the robot car to move along the indicator light strip and a device for detecting a straight line based on the visual navigation module when the visual navigation module fails to obtain the detection straight line information. Inertial navigation realizes motion control of robot car Inertial navigation module; The visual navigation module includes a straight line detection module, which is used to obtain the offset of the detection line in the world coordinate system of the robot car. S and deflection angle The visual navigation module further includes a camera acquisition module for acquiring the light strip straight line generated by the indicator light strip, and an image processing module for processing the acquired light strip straight line to obtain the detection straight line; The line patrol navigation robot also includes a main control module and a drive module. The main control module is used to send the obtained linear velocity and angular velocity data of the robot car to the drive module, and the drive module enables the robot car to patrol the line. When the robot car reaches the middle position of the two light strips and cannot obtain the detection line information, the main control module receives the data. Inertial navigation module The angle data is obtained, and the inertial navigation is performed using this angle direction as the standard heading, so that the robot car continues to travel in a straight line until the robot car re-enters the visual navigation stage.

2. The line patrol navigation robot according to claim 1, characterized in that: The indicator light strip is placed in the middle position above the navigation road.

3. The line patrol navigation robot according to claim 1, characterized in that: The indicator light band may be an ultraviolet light source, a visible light source or a near-infrared light source.

4. A control method for a line patrol navigation robot, implemented using the line patrol navigation robot according to any one of claims 1 to 3, characterized in that: The following steps are involved: S1. Get the slope of the detection line in the new pixel coordinate system k and offset b ; S2. Establish the world coordinate system of the robot car and use formula (1) to calculate the offset of the detection line in the world coordinate system of the robot car S and deflection angle : （1） in, K is the actual distance represented by the unit pixel; s The detection line in the new pixel coordinate system is y The intercept in the direction of axis 1; S3. Use the motion control model to calculate and adjust the yaw angular velocity and line speed , realizing the line patrol navigation control of the robot car; step S3 is divided into four cases: S31, >0, T <0, the robot car goes straight and turns left, and the angular velocity , the radius of rotation r is obtained by the following formula: （3） in, T The detection line is in the world coordinate system of the robot car x intercepts on the axes; N The optical center of the camera To the world coordinate system of the robot car y Axis distance; L The center of the driving wheel of the robot car in the y-axis direction of the world coordinate system of the robot car to the camera optical center distance; is the known linear velocity of the robot car; S32, <0, T > 0, the robot car goes straight and turns right, the angular velocity The calculation method is the same as S31; S33, >0, T > 0, the robot car needs to go straight and turn left first. The angular velocity is known. , control the turning time , making , turn left and enter S32; S34, <0, T <0, the robot car needs to go straight and turn right first. The angular velocity is known. , control the turning time , making , turn right and enter S31.

5. The line patrol navigation robot control method according to claim 4, characterized in that: Step S1 includes: S101, in the traditional pixel coordinate system, cutting the light band straight line image formed by the indicator light band, The image is converted into a grayscale image and binarized, and the edge straight line in the binarized image is optimized to obtain the detection straight line; S102: Eliminate the detection straight line whose slope exceeds a threshold value, and obtain a further optimized set of the two end points of the detection straight line. , Among them, the number of detection lines is n , the coordinate set of the two end points of the detection line is ; S103, using formula (2) to obtain the new pixel coordinate system Coordinates of each point: （2） in, , ， × The default resolution of the camera; S104, The point set is used as the fitting data point of the detection line, and the straight line fitting function is called to obtain the slope of the detection line in the new pixel coordinate system. k and offset b .

6. The line patrol navigation robot control method according to claim 4, characterized in that: When the detection straight line information is not obtained, the method further includes step S4: Using the main control module received Inertial navigation module The angle data is obtained, and the inertial navigation is performed using this angle direction as the standard heading, so that the robot car continues to travel in a straight line until the robot car re-enters the visual navigation stage.

Images