Intelligent scribe method and intelligent scribe robot

Abstract

The application provides a kind of intelligent scribing method and intelligent scribing robot, the method comprises: according to preset trajectory parameter control intelligent scribing robot carries out scribing operation, scribing area image carries out image segmentation, obtains route area image and scribing spray head area image;Route area image is guided to line identification and obtains guide line parameter and route position, and scribing spray head area image is positioned to obtain scribing spray head position;According to guide line parameter determines visual deviation angle, according to route position and scribing spray head position determines spray head adjustment distance;According to visual deviation angle, it is driven to adjust intelligent scribing robot, according to spray head adjustment distance, it is position adjusted scribing spray head.The application is based on visual deviation angle, can automatically drive adjustment intelligent scribing robot, improves the accuracy of intelligent scribing robot scribing movement, based on spray head adjustment distance can automatically position adjustment scribing spray head, effectively improves the accuracy of scribing spray head.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of road detection, in particular to an intelligent marking method and an intelligent marking robot. BACKGROUND

[0002] With the continuous acceleration of urbanization process and the vigorous development of highway traffic, a large number of expressways and urban and rural roads have been built, accompanied by huge road marking drawing and periodic maintenance work, which requires a large amount of manpower and material resources.

[0003] In the existing road marking process, the marking is generally performed by manually pushing a marking vehicle, which is hard and inefficient and prone to large marking deviation, thereby reducing the marking accuracy. SUMMARY

[0004] The present application aims to provide an intelligent marking method and an intelligent marking robot, and aims to solve the problems of low accuracy and high labor intensity of the existing road marking.

[0005] The present application is implemented as follows: an intelligent marking method, the method comprising:

[0006] According to the preset trajectory parameters, the intelligent marking robot is controlled to perform a marking operation, a marking area image is acquired in real time, and the marking area image is subjected to image segmentation to obtain a route area image and a marking nozzle area image;

[0007] The route area image is subjected to guide line recognition to obtain guide line parameters and a route position, and the marking nozzle area image is subjected to nozzle positioning to obtain a marking nozzle position;

[0008] The visual deviation angle is determined according to the guide line parameters, and the nozzle adjustment distance is determined according to the route position and the marking nozzle position;

[0009] The intelligent marking robot is driven and adjusted according to the visual deviation angle, and the marking nozzle on the intelligent marking robot is position-adjusted according to the nozzle adjustment distance.

[0010] Preferably, the guide line recognition of the route area image comprises:

[0011] The route area image is subjected to Gaussian filtering processing to obtain a route filtered image, and the route filtered image is subjected to edge detection to obtain an edge image;

[0012] The edge image is subjected to Hough transformation to obtain guide feature lines, and the inclination angles of the guide feature lines are calculated respectively;

[0013] According to the inclination angle, each guide feature line is classified to obtain a scribe guide line, and parameter information and position information of each scribe guide line are obtained to obtain the guide line parameter and the route position.

[0014] Preferably, according to the inclination angle, each guide feature line is classified, including:

[0015] When θ0+δ≥π / 2, and θ0-δ≤θ i ≤π / 2 or -π / 2≤θ i ≤θ0+δ-π, it is determined that θ i is the same curve as θ0, otherwise it is determined that θ i is not the same curve as θ0.

[0016] When θ0-δ≤-π / 2, and -π / 2≤θ i ≤θ0+δ or π+θ0-δ≤θ i ≤π / 2, it is determined that θ i is the same curve as θ0, otherwise it is determined that θ i is not the same curve as θ0.

[0017] When -π / 2+δ≤θ0≤π / 2-δ, and |θ i -θ0|≤δ, it is determined that θ i is the same curve as θ0, otherwise it is determined that θ i is not the same curve as θ0.

[0018] Wherein, θ0is the inclination angle of the first guide feature line, θ i is the inclination angle of the i+1th guide feature line, and δ is a preset angle error.

[0019] If θ i is the same curve as θ0, the guide feature lines corresponding to θ i and θ0are stored in a first guide line set.

[0020] If θ i is not the same curve as θ0, the guide feature line corresponding to θ i is stored in a second guide line set.

[0021] The scribe guide line is generated according to the first guide line set and the second guide line set.

[0022] Preferably, the visual deviation angle is determined according to the guide line parameter, including:

[0023] The center point of each scribe guide line is obtained to obtain an image trajectory line feature point, and a deviation point is determined according to the pixel value of the route area image.

[0024] An angle between a line connecting the image trajectory feature point and the deviation point and a vertical center line of the image in the route area image is calculated to obtain the visual deviation angle.

[0025] Preferably, the formula for determining the spray head adjustment distance according to the route position and the spray head position comprises:

[0026] e p = k1(x z -x t )

[0027] wherein e p is the spray head adjustment distance, x z is the horizontal coordinate in the spray head position, x t is the horizontal coordinate of the image trajectory feature point in the route position, and k1 is a first proportional coefficient.

[0028] Preferably, the formula for driving and adjusting the intelligent line marking robot according to the visual deviation angle comprises:

[0029] V r = (k2γ + 1)V l

[0030] wherein V l is the control speed of the left driving wheel on the intelligent line marking robot, V r is the control speed of the right driving wheel on the intelligent line marking robot, k2 is a second proportional coefficient, and γ is the visual deviation angle.

[0031] Preferably, the method further comprises:

[0032] If the crossing line guide lines in the same line marking area image are detected, a guide line intersection point between the crossing line guide lines is obtained, and a motion four-tuple data is queried according to the number of the guide line intersection point.

[0033] The intelligent line marking robot is controlled to adjust the pose according to the rotation angle and the forward distance in the motion four-tuple data, and the spray head is controlled to switch the on-off state according to the line marking identifier in the motion four-tuple data.

[0034] Preferably, the spray head positioning on the line marking head area image comprises:

[0035] The line marking head area image is color-identified, and a region position corresponding to a preset color in the color identification result is determined as the spray head position.

[0036] Another purpose of the embodiment of the present application is to provide an intelligent line marking robot, comprising:

[0037] The vehicle body, the driving device arranged on the side of the vehicle body, and the line marking device connected with the vehicle body, the vehicle body is provided with a control unit, the line marking device comprises a linear guide rail, a stepping motor connected with the linear guide rail, a guide rail sliding table arranged on the linear guide rail, and a line marking nozzle arranged on the guide rail sliding table;

[0038] The control unit is used for controlling the driving device and the line marking nozzle to perform driving operation and line marking operation respectively according to preset trajectory parameters;

[0039] Real-time acquisition of line marking area image, and image segmentation of the line marking area image to obtain route area image and line marking nozzle area image;

[0040] Guide line identification of the route area image to obtain guide line parameters and route position, and nozzle positioning of the line marking nozzle area image to obtain line marking nozzle position;

[0041] Determination of visual deviation angle according to the guide line parameters, and determination of nozzle adjustment distance according to the route position and the line marking nozzle position;

[0042] Driving adjustment of the driving device according to the visual deviation angle, and position adjustment of the line marking nozzle driven by the stepping motor according to the nozzle adjustment distance.

[0043] Preferably, the linear guide rail is provided with a left limiter and a right limiter, the guide rail sliding table is arranged between the left limiter and the right limiter, and the linear guide rail is further provided with a light supplementing lamp.

[0044] According to the embodiment of the present application, guide line identification of the route area image, nozzle positioning of the line marking nozzle area image, effective determination of guide line parameters, route position and line marking nozzle position, automatic determination of visual deviation angle based on guide line parameters, automatic determination of nozzle adjustment distance based on route position and line marking nozzle position, automatic driving adjustment of the intelligent line marking robot based on visual deviation angle, improvement of the accuracy of line marking movement of the intelligent line marking robot, automatic position adjustment of the line marking nozzle based on nozzle adjustment distance, effective improvement of the accuracy of line marking of the nozzle, elimination of the need for manual line marking, improvement of line marking efficiency, and prevention of the phenomenon of large line marking deviation caused by manual line marking. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 is a flowchart of the intelligent line marking method provided by the first embodiment of the present application;

[0046] Figure 2is a schematic diagram of a route filtered image provided by the first embodiment of the present application;

[0047] Figure 3 is a schematic diagram of an edge image provided by the first embodiment of the present application;

[0048] Figure 4 is a schematic diagram of a guide feature line provided by the first embodiment of the present application;

[0049] Figure 5 is a schematic diagram of a visual deviation angle provided by the first embodiment of the present application;

[0050] Figure 6 is a schematic diagram of a smart line marking robot circle motion provided by the first embodiment of the present application;

[0051] Figure 7 is a schematic diagram of an intersection provided by the first embodiment of the present application

[0052] Figure 8 is a schematic diagram of another intersection provided by the first embodiment of the present application;

[0053] Figure 9 is Figure 7 a trajectory diagram of (1);

[0054] Figure 10 is Figure 8 a trajectory diagram of (2);

[0055] Figure 11 is a schematic diagram of a guide line intersection provided by the first embodiment of the present application;

[0056] Figure 12 is a schematic diagram of a smart line marking robot provided by the second embodiment of the present application;

[0057] Figure 13 is a top view of a smart line marking robot provided by the second embodiment of the present application;

[0058] Figure 14 is a left view of a smart line marking robot provided by the second embodiment of the present application;

[0059] Figure 15 is a front view of a smart line marking robot provided by the second embodiment of the present application;

[0060] Figure 16 is a system framework diagram of a smart line marking robot provided by the second embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0062] In order to illustrate the technical solutions of the present application, the following will be described by specific embodiments.

[0063] Embodiment one

[0064] Please refer to Figure 1 , the flow chart of the intelligent line marking method provided by the first embodiment of the present application, the intelligent line marking method can be applied to any terminal device or system, the intelligent line marking method comprises the following steps:

[0065] Step S10, controlling the intelligent line marking robot to perform line marking operation according to the preset trajectory parameters, acquiring the line marking area image in real time, and performing image segmentation on the line marking area image to obtain the route area image and the line marking nozzle area image;

[0066] In this step, the preset trajectory parameters are the trajectory positions corresponding to the target guide lines.

[0067] In this step, an image acquisition device is arranged on the intelligent line marking robot, which is used to acquire the line marking area image in real time. The route area image and the line marking nozzle area image are obtained by performing image segmentation on the line marking area image according to the preset image segmentation line. The preset image segmentation line can be set according to requirements, for example, the preset image segmentation line can be set as a horizontal segmentation line or a vertical segmentation line with the center point of the line marking area image as the terminal point. In this step, the upper half of the line marking area image is the route area image, and the lower half is the line marking nozzle area image.

[0068] Step S20, performing guide line identification on the route area image to obtain guide line parameters and route positions, and performing nozzle positioning on the line marking nozzle area image to obtain line marking nozzle positions;

[0069] Optionally, in this step, the guide line identification on the route area image comprises:

[0070] performing Gaussian filtering processing on the route area image to obtain a route filtered image, and performing edge detection on the route filtered image to obtain an edge image; wherein, please refer to Figures 2-3 By performing Gaussian filtering processing on the route area image, the Gaussian noise in the route area image can be effectively eliminated, and by performing edge detection on the route filtered image, the edge information in the route filtered image can be effectively extracted. In this step, the Canny algorithm can be used for edge detection.

[0071] A Hough transform is performed on the edge image to obtain guiding feature lines, and the tilt angle of each guiding feature line is calculated; where, please refer to Figure 4 By performing Hough transform on the edge image, guiding feature lines in the edge image can be effectively extracted;

[0072] The guide feature lines are classified according to the tilt angle to obtain the scribed guide lines, and the parameter information and position information of each scribed guide line are obtained to obtain the guide line parameters and the route position.

[0073] Furthermore, the classification of each guiding feature line according to the tilt angle includes:

[0074] When θ0+δ≥π / 2 and θ0-δ≤θ i ≤π / 2 or -π / 2≤θ i If ≤θ0+δ-π, then determine θ i If θ0 is on the same curve, then θ is determined to be the same curve. i It is not the same curve as θ0;

[0075] When θ0-δ≤-π / 2, and -π / 2≤θ i ≤θ0+δ or π+θ0-δ≤θ i If ≤π / 2, then determine θ i If θ0 is on the same curve, then θ is determined to be the same curve. i It is not the same curve as θ0;

[0076] When -π / 2+δ≤θ0≤π / 2-δ, and |θ i If -θ0|≤δ, then determine θ i If θ0 is on the same curve, then θ is determined to be the same curve. i It is not the same curve as θ0;

[0077] Where θ0 is the tilt angle of the first guiding feature line, θ i δ is the tilt angle of the (i+1)th guiding feature line, where δ is a preset angle error;

[0078] If θ i If θ0 is on the same curve, then θ i The guiding feature line corresponding to θ0 is stored in the first guiding line set;

[0079] If θ i If θ0 is not on the same curve, then θ i The corresponding guiding feature lines are stored in the second guiding line set;

[0080] The generation of the scribe guide line is determined according to the first guide line set and the second guide line set; wherein the corresponding guide feature line in the first guide line set is the same scribe guide line, and each guide feature line in the second guide line set is a guide line different from the first guide line by more than δ;

[0081] In this step, the scribe guide line is:

[0082] y=c0x+c1

[0083] Wherein, x, y are known quantities (image point information extracted by the Hough transform, the Hough transform extracts many groups of points), and c0 and c1 are to be solved;

[0084] For the i-th curve point set {(x i , y i ), (x i+1 , y i+1 )} extracted by the Hough transform, take:

[0085]

[0086] T represents matrix transposition, C i is a column vector, c0 is a first-order term coefficient, c1 is a constant, x is the horizontal coordinate point of the image extracted by the Hough transform, y is the vertical coordinate point of the image extracted by the Hough transform, c0 and c1 are solved by least squares fitting, and the parameter matrix is solved by least squares fitting:

[0087] C i =(c0, c1) T

[0088] The inclination angle is obtained:

[0089] θ i =tan -1 c0, θ i ∈(-π / 2, π / 2)

[0090] For different inclination angles, classification is performed according to the preset angle error δ, taking the inclination angle θ0 corresponding to the first group of line segment point sets as the standard, and taking the inclination angle θ i corresponding to the i+1 group of point sets, and comparing it with θ0, and considering the value range of θ i , taking the straight line and the cross line as examples.

[0091] If θ i is the same curve as θ0, the curve parameter data is recorded in the first guide line set C0, otherwise it is recorded in the second guide line set C A For any m curves in a curve parameter set, take:

[0092]

[0093] wherein m is the total number of curves in the set, C i is the parameter of the i-th curve in the set.

[0094] Preferably, in the embodiment, if x i = x i+1 , c0 is not solved by using the least square method, but c0 is directly given a larger value, such as 5000, and c1 is solved as follows: c1 = -c0 x i + (y i + y i+1 ) / 2.

[0095] Further, the jet positioning on the scribe line jet area image comprises:

[0096] color recognition is performed on the scribe line jet area image, and a region position corresponding to a preset color in the color recognition result is determined as the scribe line jet position.

[0097] The preset color can be set according to requirements, and the preset color in the step is set as red, that is, the region position of red in the scribe line jet area image is determined as the scribe line jet position.

[0098] Step S30, determining a visual deviation angle according to the guide line parameter, and determining a jet adjustment distance according to the route position and the scribe line jet position.

[0099] The visual deviation angle is determined according to the guide line parameter, comprising:

[0100] obtaining a center point of each scribe guide line, obtaining an image trajectory line feature point, and determining a deviation point according to a pixel value of the route area image.

[0101] calculating an included angle between a connecting line between the image trajectory line feature point and the deviation point and an image vertical center line in the route area image, to obtain the visual deviation angle.

[0102] Please refer to Figure 5 , assuming that a pixel size of the route area image is (h, w), the image trajectory line feature point is the center point of the scribe guide line, the deviation point is (h, w / 2), and the visual deviation angle γ is an included angle between a connecting line between the image trajectory line feature point (x t , y t ) and the deviation point (h, w / 2) and the image vertical center line, the direction is from the image vertical center line to the connecting line, the connecting line between the image trajectory line feature point (x t , y t ) and the deviation point (h, w / 2) is taken, x d = h, and yd = w / 2

[0103]

[0104] The monomial linear regression equation is selected for least square fitting to calculate the visual deviation angle γ:

[0105]

[0106] c0 is the coefficient of the first order term of the monomial linear equation;

[0107] D = (X T X) -1 X T Y

[0108] T represents the matrix transpose, and -1 represents the inverse of the matrix;

[0109] D = (d0, d1) T

[0110] If x t = x d = h, d0 and d1 are not solved by the least square method, but d0 is directly given a larger value, such as 5000, and the value of d1 is: d1 = -d0 x x t + (y t + y d ) / 2.

[0111] Optionally, the center position of the marking head is (x z , y z ), and the formula used to determine the head adjustment distance according to the route position and the marking head position comprises:

[0112] e p = k1 (x z - x t )

[0113] wherein e p is the head adjustment distance, x z is the horizontal coordinate in the marking head position, x t is the horizontal coordinate of the image trajectory feature point in the route position, and k1 is the first proportional coefficient, the value of the proportional coefficient k1 being determined by the projection size of the actual physical position on the image.

[0114] In step S40, the intelligent marking robot is driven and adjusted according to the visual deviation angle, and the position of the marking head on the intelligent marking robot is adjusted according to the head adjustment distance;

[0115] Specifically, by controlling the speed of the differential drive wheels (left and right drive wheels) on the intelligent line-marking robot, the drive adjustment effect of the intelligent line-marking robot is achieved, based on the center position of the line-marking nozzle (x... z y z The desired position P of the marking nozzle is determined by adjusting the nozzle distance. E Based on the desired location P E Controlling the stepper motor on the intelligent robot to move the marking nozzle to the desired position P E This allows for position adjustment of the marking nozzle, improving the accuracy of the marking.

[0116] Optionally, in this step, the formula used to adjust the drive of the intelligent line-marking robot based on the visual deviation angle includes:

[0117] The control law is designed as follows:

[0118] f(γ)=V r / V l =k2γ+1

[0119] V r It can be represented as:

[0120] V r =(k2γ+1)V l

[0121] Among them, V l V is the control speed of the left drive wheel on the intelligent line-marking robot. r The control speed of the right drive wheel on the intelligent line-marking robot is k2, which is the second proportional coefficient, and γ is the visual deviation angle. This is achieved by controlling V... l and V r This creates a speed difference between the left and right drive wheels, which is used to control the intelligent line-marking robot to adjust its angle, thereby improving the accuracy of the robot's movement and ultimately enhancing the accuracy of the line marking.

[0122] In this embodiment, the basic motion elements of the line-marking robot are a straight line and a circle, denoted by L(T) and R(θ, d) respectively, where T can be positive or negative, with a positive value indicating the direction of motion relative to the local coordinate system X. R The directions are the same; θ represents the rotation angle, which can be positive or negative, with positive indicating clockwise rotation; d represents the rotation radius, which is the distance from the rotation center O to the local coordinate system circle O. R The distance.

[0123] The motion represented by R(θ, d) is as follows: Figure 6 As shown, trajectory planning can effectively perform line marking operations for different needs on different road sections, such as...Figure 7 and Figure 8 When the intelligent marking robot moves to the intersection shown in Figure 7 , there are multiple possible trajectories (1) and (2); and when the intelligent marking robot moves to the position shown in Figure 8 , multiple pose adjustments are required, and the trajectory graph is obtained by using a planar trajectory planning method, Figure 9 is the trajectory graph in (1) of Figure 7 is the trajectory graph in (2) of Figure 10 is the trajectory graph in (3) of Figure 8 is the trajectory graph in (4) of Figure 9 and Figure 10 In (1) of

[0124] Further, in the embodiment, the method further comprises:

[0125] If it is detected that the intersection of the marking guide lines exists in the same marking area image, the guide line intersection point between the intersecting marking guide lines is obtained, and the motion four-tuple data is queried according to the number of the guide line intersection point. The intersection number encountered for the first time by the intelligent marking robot during marking on the site is 1, the intersection number encountered for the second time is 2, and the number is the node number. The node number of the motion four-tuple data generated according to the execution sequence of the intelligent marking robot marking indeed contains the coordinate information of the guide line intersection point;

[0126] The pose adjustment of the intelligent marking robot is controlled according to the rotation angle and the forward distance in the motion four-tuple data, and the switching of the on-off state of the marking head is controlled according to the marking identifier in the motion four-tuple data;

[0127] wherein the motion four-tuple data is (ID, A, D, P), ID represents the node number, A represents the rotation angle, the value of which corresponds to θ in R(θ, d), d is the distance between the intelligent marking robot local coordinate system circle point O R and the center point O of the marking head, D represents the forward distance set in advance for each node, and P represents whether to mark, which is a binary quantity, and Down and Up are used to represent marking and non-marking, respectively. The guide line intersection point is used to represent the coordinates of the pose adjustment of the intelligent robot. The number of the guide line intersection point is matched with the pre-stored trajectory data record table to obtain the four-tuple data. The trajectory data record table stores the correspondence between the numbers of different guide line intersection points and the corresponding four-tuple data. Preferably, the trajectory data record table stores the numbers of different guide line intersection points, for example, the trajectory data record table includes:

[0128]

[0129]

[0130] In this step, the position of the intelligent line marking robot pose adjustment is judged by visual information, please refer to Figure 11 , when the intelligent line marking robot moves to the node as shown in Figure 11 , two line guide lines and an intersection of guide lines will be detected, and the intelligent line marking robot will query the trajectory data record table to determine the next movement trajectory.

[0131] If the movement quaternion data at this time is (1, -90, D, Down), the intelligent line marking robot will perform pose adjustment according to the following steps:

[0132] (1) Start pose adjustment, stop the spray head from marking; (2) rotate the intelligent robot counterclockwise by 90° with the center point of the line marking spray head as the center; (3) adjust the horizontal error of the spray head; (4) end pose adjustment, the robot moves D meters along the guide line W1, and the spray head starts marking;

[0133] If the movement quaternion data at this time is (1, 0, D, Up), because the rotation angle is 0, the intelligent line marking robot does not perform pose adjustment, and because the marking value is Up, the spray head stops marking, and the intelligent line marking robot moves along the guide line W2;

[0134] If the movement quaternion data at this time is (1, 90, D, Up), the intelligent line marking robot will perform pose adjustment according to the following steps: (1) Start pose adjustment, stop the spray head from marking; (2) rotate the robot clockwise by 90° with the center point of the line marking spray head as the center; (3) adjust the horizontal error of the spray head; (4) end pose adjustment, the robot moves D meters along the guide line W3, and the marking value is Up, and the spray head does not mark.

[0135] In the embodiment of the application, the guide line parameters, the route position and the line marking spray head position can be effectively obtained by guiding line recognition on the route area image and spray head positioning on the line marking spray head area image, the visual deviation angle can be automatically determined based on the guide line parameters, the spray head adjustment distance can be automatically determined based on the route position and the line marking spray head position, the intelligent line marking robot can be automatically driven and adjusted based on the visual deviation angle, the accuracy of line marking movement of the intelligent line marking robot is improved, the position of the line marking spray head can be automatically adjusted based on the spray head adjustment distance, the accuracy of line marking of the spray head is effectively improved, manual line marking is not needed, the line marking efficiency is improved, and the phenomenon of large line marking deviation caused by manual line marking is prevented.

[0136] Embodiment two

[0137] Please refer to Figures 12-15, is a structural schematic view of the intelligent line marking robot 100 provided by the second embodiment of the present application, comprising a vehicle body 9, a driving device arranged at the side of the vehicle body 9 and a line marking device connected with the vehicle body 9;

[0138] The vehicle body 9 is provided with a control unit, and the line marking device comprises a linear guide rail 6, a stepping motor 1 connected with the linear guide rail 6, a guide rail sliding table 7 arranged on the linear guide rail 6 and a line marking nozzle 4 arranged on the guide rail sliding table 7.

[0139] In the embodiment, the control unit is used for respectively controlling the driving device and the line marking nozzle 4 to perform driving operation and line marking operation according to preset track parameters;

[0140] Real-time line marking area images are acquired, and image segmentation is performed on the line marking area images to obtain route area images and line marking nozzle area images;

[0141] Guide line identification is performed on the route area images to obtain guide line parameters and route positions, and nozzle positioning is performed on the line marking nozzle area images to obtain line marking nozzle positions;

[0142] A visual deviation angle is determined according to the guide line parameters, and a nozzle adjustment distance is determined according to the route positions and the line marking nozzle positions;

[0143] The driving device is driven and adjusted according to the visual deviation angle, and the stepping motor 1 is controlled to drive the line marking nozzle 4 to perform position adjustment according to the nozzle adjustment distance.

[0144] Optionally, the linear guide rail 6 is provided with a left limiter 5 and a right limiter 3, the guide rail sliding table 7 is arranged between the left limiter 5 and the right limiter 3, and the linear guide rail 6 is further provided with a light supplement lamp 2 and a camera.

[0145] The driving device comprises two differential driving wheels 8 and two driven wheels 11, and wheels can also be added or reduced according to specific needs, and the control unit comprises a driver 10, a controller, a battery and other necessary driving units, the driver 10 is used for controlling the stepping motor 1, and the controller is used for controlling the driving motor of the differential driving wheel 8, and the driving motor of the differential driving wheel 8 can adopt a planetary reduction motor.

[0146] The guide rail sliding table 7 is driven by the stepping motor 1 to control the movement of the line marking nozzle 4 to correct the error in the horizontal direction of line marking, the left limiter 5 and the right limiter 3 are used to limit the movement of the guide rail sliding table 7, mainly to prevent the line marking nozzle 4 from interfering with the differential driving wheel 8, and the camera and the light supplement lamp 2 are used to capture road guide line information and position information of the line marking nozzle.

[0147] Please refer to Figure 16The overall framework of the intelligent line marking robot 100 is shown in the figure, wherein two differential driving wheels 8 are driven by driving wheel motors, and the line marking nozzle 4 is driven by a stepping motor 1. The intelligent line marking robot 100 obtains the site environment information (line marking area image) through a visual sensor, makes a control decision according to the relevant information, obtains a visual deviation angle γ and an expected position P of the line marking nozzle 4 E , adjusts the speeds (V l and V r ) of the two differential driving wheels 8 according to γ combined control law, and outputs the corresponding rotation speeds (V Rl and V Rr ) of the corresponding reduction motors based on V l and V r , adjusts the output rotation speed P R of the stepping motor 1 according to P E , and adjusts the position of the line marking nozzle 4.

[0148] In the embodiment, the guide line parameters, the route position and the line marking nozzle position can be effectively obtained through the guide line recognition on the route area image and the nozzle positioning on the line marking nozzle area image. The visual deviation angle can be automatically determined based on the guide line parameters, the nozzle adjustment distance can be automatically determined based on the route position and the line marking nozzle position, the intelligent line marking robot 100 can be automatically driven and adjusted based on the visual deviation angle, the accuracy of the line marking movement of the intelligent line marking robot 100 is improved, the position of the line marking nozzle 4 can be automatically adjusted based on the nozzle adjustment distance, the accuracy of the line marking of the nozzle is effectively improved, the line marking by artificial method is not needed, the line marking efficiency is improved, and the phenomenon of large line marking deviation caused by artificial line marking is prevented.

[0149] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features. The modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. An intelligent line-drawing method, characterized in that, The method includes: The intelligent line-marking robot is controlled to perform line-marking operations according to preset trajectory parameters, and the image of the line-marking area is acquired in real time. The image of the line-marking area is then segmented to obtain the route area image and the line-marking nozzle area image. The guide line is identified in the route area image to obtain the guide line parameters and route position, and the nozzle is located in the line-marking nozzle area image to obtain the line-marking nozzle position. The step of identifying guide lines in the route area image includes: The route region image is subjected to Gaussian filtering to obtain a route-filtered image, and edge detection is performed on the route-filtered image to obtain an edge image; Perform a Hough transform on the edge image to obtain guiding feature lines, and calculate the tilt angle of each guiding feature line. The guide feature lines are classified according to the tilt angle to obtain the line drawing guide lines, and the parameter information and position information of each line drawing guide line are obtained to obtain the guide line parameters and the route position; The visual deviation angle is determined based on the guide line parameters, and the nozzle adjustment distance is determined based on the route position and the nozzle position. The formula used to determine the nozzle adjustment distance based on the route position and the marked nozzle position includes: e p = k 1( x z - x t ) in, e p This refers to the nozzle adjustment distance. x z It is the x-coordinate of the position of the scribing nozzle. x t It is the x-coordinate of the feature point of the image trajectory line in the route location. k 1 is the first proportionality coefficient; The intelligent line-marking robot is driven and adjusted according to the visual deviation angle, and the position of the line-marking nozzle on the intelligent line-marking robot is adjusted according to the nozzle adjustment distance. The formula used to adjust the drive of the intelligent line-drawing robot based on the visual deviation angle includes: V r =( k 2 γ +1) V l in, V l It refers to the control speed of the left drive wheel on the intelligent line-marking robot. V r It is the control speed of the right drive wheel on the intelligent line-marking robot. k 2 is the second proportionality coefficient. γ It is the aforementioned visual deviation angle.

2. The intelligent line-drawing method as described in claim 1, characterized in that, The classification of each guiding feature line according to the tilt angle includes: when θ 0+ δ ≥π / 2, and θ 0- δ ≤ θ i ≤π / 2 or -π / 2≤ θ i ≤ θ 0+ δ -π, then determine θ i and θ 0 indicates the same curve; otherwise, it is determined that... θ i and θ 0 is not the same curve; when θ 0- δ When ≤-π / 2, and -π / 2≤ θ i ≤ θ 0+ δ or π+ θ 0- δ ≤ θ i If ≤π / 2, then determine θ i and θ 0 indicates the same curve; otherwise, it is determined that... θ i and θ i Not the same curve; When -π / 2+ δ ≤ θ 0≤π / 2- δ At that time, and | θ i - θ 0|≤ δ Then determine θ i and θ 0 indicates the same curve; otherwise, it is determined that... θ i and θ 0 is not the same curve; in, θ 0 is the tilt angle of the first guiding feature line. θ i For the first i +1 tilt angle of the guiding feature line, δ This is a preset angle error; like θ i and θ If 0 represents the same curve, then... θ i and θ The guide feature line corresponding to 0 is stored in the first guide line set; like θ i and θ If 0 is not on the same curve, then... θ i The corresponding guiding feature lines are stored in the second guiding line set; The drawing guide line is generated based on the first guide line set and the second guide line set.

3. The intelligent line-drawing method as described in claim 1, characterized in that, Determining the visual deviation angle based on the guide line parameters includes: The center point of each guide line is obtained to obtain the feature points of the image trajectory line, and the deviation point is determined according to the pixel value of the image of the route area. The visual deviation angle is obtained by calculating the angle between the line connecting the feature points of the image trajectory line and the deviation point and the vertical center line of the image in the route area image.

4. The intelligent line-drawing method as described in any one of claims 1 to 3, characterized in that, The method further includes: If intersecting guide lines are detected in the same scribbled area image, the intersection point of the guide lines between the intersecting guide lines is obtained, and the motion quadruple data is queried according to the number of the guide line intersection point. The intelligent line-marking robot is controlled to adjust its pose based on the rotation angle and forward distance in the motion quaternion data, and the line-marking nozzle is controlled to switch its on / off state based on the line-marking identifier in the motion quaternion data.

5. The intelligent line-drawing method as described in claim 1, characterized in that, The step of locating the nozzles in the image of the marked nozzle area includes: Color recognition is performed on the image of the scribing nozzle area, and the location of the area corresponding to the preset color in the color recognition result is determined as the location of the scribing nozzle.

6. An intelligent line-drawing robot, characterized in that, include: The vehicle body, the drive device located on the side of the vehicle body, and the marking device connected to the vehicle body are provided. The vehicle body is equipped with a control unit. The marking device includes a linear guide rail, a stepper motor connected to the linear guide rail, a guide rail slide on the linear guide rail, and a marking nozzle on the guide rail slide. The control unit is used to control the drive device and the line-drawing nozzle to perform drive operations and line-drawing operations respectively according to preset trajectory parameters; The image of the marked area is acquired in real time, and the image of the marked area is segmented to obtain the route area image and the marked nozzle area image; The guide line is identified in the route area image to obtain the guide line parameters and route position, and the nozzle is located in the line-marking nozzle area image to obtain the line-marking nozzle position. The step of identifying guide lines in the route area image includes: The route region image is subjected to Gaussian filtering to obtain a route-filtered image, and edge detection is performed on the route-filtered image to obtain an edge image; Perform a Hough transform on the edge image to obtain guiding feature lines, and calculate the tilt angle of each guiding feature line. The guide feature lines are classified according to the tilt angle to obtain the line drawing guide lines, and the parameter information and position information of each line drawing guide line are obtained to obtain the guide line parameters and the route position; The visual deviation angle is determined based on the guide line parameters, and the nozzle adjustment distance is determined based on the route position and the nozzle position. The formula used to determine the nozzle adjustment distance based on the route position and the marked nozzle position includes: e p = k 1( x z - x t ) in, e p This refers to the nozzle adjustment distance. x z It is the x-coordinate of the position of the scribing nozzle. x t It is the x-coordinate of the feature point of the image trajectory line in the route location. k 1 is the first proportionality coefficient; The driving device is adjusted according to the visual deviation angle, and the stepper motor is controlled to drive the marking nozzle to adjust its position according to the nozzle adjustment distance. The formula used to adjust the drive of the intelligent line-drawing robot based on the visual deviation angle includes: V r =( k 2 γ +1) V l in, V l It refers to the control speed of the left drive wheel on the intelligent line-marking robot. V r It is the control speed of the right drive wheel on the intelligent line-marking robot. k 2 is the second proportionality coefficient. γ It is the aforementioned visual deviation angle.

7. The intelligent line-marking robot as described in claim 6, characterized in that, The linear guide rail is equipped with a left limiter and a right limiter, and the guide rail slide is located between the left limiter and the right limiter. The linear guide rail is also equipped with a supplementary light.

Images