Automatic scribing method, device and equipment of scribing robot and storage medium

Abstract

The invention discloses an automatic lineation method, device and equipment for a lineation robot and a storage medium. The method comprises the steps that first position information of the lineation robot is obtained in real time based on a global satellite navigation system; acquiring a road image in front of the marking robot, and analyzing the road image to obtain second position information of the marking robot relative to the road center point; combining the first position information and the second position information to confirm the accurate position of the scribing robot; and controlling the marking robot to automatically travel based on the accurate position and the preset planned path, and executing road marking operation. According to the invention, the position of the marking robot in the road is accurately positioned through the positioning information of the global satellite navigation system and the acquired road image in front of the marking robot, and then the marking robot is controlled to automatically advance according to the preset planned path and execute the marking operation according to the accurate positioning, so that the marking operation can be accurately completed; and the manpower consumption is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of road marking, and in particular to a method, device, equipment and storage medium for automatic road marking by a road marking robot. Background Art

[0002] Compared with the rapid development of highway construction, the development of road marking equipment lags behind to a certain extent. At present, most road marking equipment is still in a semi-automated stage, such as hand-held, self-propelled, vehicle-mounted and other road marking equipment. Due to the imperfect design of the control system of traditional road marking equipment, high requirements are placed on the marking technology of construction personnel. When the marking machine is operating, construction personnel are required to adjust the spraying position and driving speed of the marking machine at all times. There are large subjective factors, so there will be deviations in road markings, uneven markings, inconsistent widths, uneven thicknesses, etc. (especially on curved sections), and construction personnel are often required to rework, with low work efficiency and high labor intensity. With the rapid increase in highway mileage and the continuous increase in labor costs, traditional road marking equipment has gradually been unable to meet the requirements of the rapid development of highway construction at this stage.

[0003] With the development of autonomous driving technology, automatic road marking robots are increasingly used in road construction and maintenance. Automatic road marking robots can complete road marking tasks efficiently and accurately, but their precise navigation and positioning technology is the key to achieving this goal. Most of the automatic road marking robots currently on the market rely on a single RTK (Real stands for Time Kinematic, real-time dynamic positioning technology) positioning technology. RTK positioning technology requires manual marking and staking in the early stage, which is cumbersome to prepare, and the construction efficiency and accuracy are limited by the marking frequency. Summary of the invention

[0004] In view of this, the present application provides a method, device, equipment and storage medium for automatic road marking by a road marking robot to solve the problems of low efficiency and low precision of existing road marking methods.

[0005] In order to solve the above technical problems, a technical solution adopted in the present application is: to provide an automatic marking method for a marking robot, which includes: acquiring the first position information of the marking robot in real time based on a global satellite navigation system; collecting the road image in front of the marking robot, and analyzing the road image to obtain the second position information of the marking robot relative to the center point of the road; confirming the precise position of the marking robot by combining the first position information and the second position information; controlling the marking robot to automatically drive based on the precise position and a preset planned path, and performing road marking operations at the same time.

[0006] As a further improvement of the present application, an image of the road in front of the road marking robot is collected and analyzed to obtain second position information of the road marking robot relative to the center point of the road, including: using an image acquisition device pre-set on the road marking robot to collect the image of the road in front of the road marking robot; inputting the road image into a pre-trained Mask representation RCNN neural network, and using the Mask representation RCNN neural network to identify the edge of the road; and confirming the second position information of the road marking robot relative to the center point of the road based on the position information of the road marking robot itself and the position information of the center point of the road.

[0007] As a further improvement of the present application, after controlling the automatic driving of the marking robot based on the precise position and the preset planned path, it also includes: judging whether the marking robot is offset based on the precise position and the preset planned path; if so, adjusting the forward direction of the marking robot based on the precise position and the preset planned path.

[0008] As a further improvement of the present application, the forward direction of the marking robot is adjusted based on the precise position and the preset planned path, including: calculating the first deviation distance and the first deviation angle between the precise position and the corresponding position in the preset planned path; inputting the first deviation distance and the first deviation angle into the basic motion model of the marking robot to adjust the forward direction of the marking robot.

[0009] As a further improvement of the present application, after controlling the automatic driving of the marking robot based on the precise position and the preset planned path, it also includes: dividing the road image into multiple areas using a pre-trained target detection algorithm; determining whether there are obstacles in the area on the preset planned path; if so, avoiding the obstacles based on a preset obstacle avoidance method.

[0010] As a further improvement of the present application, avoiding obstacles based on a preset obstacle avoidance method includes: based on a depth estimation algorithm, using feature points and texture information in a road image to calculate the distance between the road marking robot and the obstacle; when the distance is less than a preset distance threshold, controlling the road marking robot to stop suddenly; when the distance is greater than or equal to the preset distance threshold, controlling the road marking robot to slow down, using a target detection algorithm to identify the outline of the obstacle, and generating an obstacle avoidance path based on the outline and pre-acquired map data, and controlling the road marking robot to move forward based on the obstacle avoidance path.

[0011] As a further improvement of the present application, after controlling the forward movement of the line marking robot based on the obstacle avoidance path, it also includes: after the line marking robot bypasses the obstacle, calculating the second deviation distance and the second deviation angle between the precise position and the corresponding position in the preset planned path; inputting the second deviation distance and the second deviation angle into the basic motion model of the line marking robot to adjust the forward direction of the line marking robot.

[0012] In order to solve the above technical problems, another technical solution adopted in the present application is: to provide a road marking robot, which includes: an acquisition module, used to acquire the first position information of the road marking robot in real time based on the global satellite navigation system; a collection module, used to collect the road image in front of the road marking robot, and analyze the road image to obtain the second position information of the road marking robot relative to the center point of the road; a confirmation module, used to confirm the precise position of the road marking robot in combination with the first position information and the second position information; a control module, used to control the road marking robot to automatically drive based on the precise position and a preset planned path, and perform road marking operations at the same time.

[0013] In order to solve the above-mentioned technical problems, another technical solution adopted in the present application is: to provide a computer device, which includes a processor and a memory coupled to the processor, wherein program instructions are stored in the memory, and when the program instructions are executed by the processor, the processor executes the steps of the automatic marking method of a marking robot as described in any of the above-mentioned items.

[0014] In order to solve the above technical problems, another technical solution adopted by the present application is: to provide a storage medium storing program instructions capable of implementing any of the above-mentioned automatic marking methods of a marking robot.

[0015] The beneficial effects of the present application are as follows: the automatic line marking method of a line marking robot of the present application obtains the first position information of the line marking robot by utilizing the global satellite navigation system, and then obtains the road image in front of the line marking robot by utilizing the image acquisition device of the line marking robot itself, confirms the second position information of the line marking robot relative to the center point of the road from the road image, fuses the first position information and the second position information to obtain the precise position of the line marking robot on the road, and then controls the line marking robot to automatically drive in combination with the precise position and the preset planned path, and simultaneously performs the road marking operation, which realizes the precise positioning of the line marking robot on the road, and then controls the line marking robot to automatically drive according to the precise positioning and the preset planned path, avoiding a large deviation between the line marking robot's travel route and the preset planned path, thereby ensuring that the line marking operation can be accurately performed, the line marking accuracy is higher, and no preparation work is required in advance, and the line marking efficiency is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of an embodiment of the automatic marking method of a marking robot of the present invention;

[0017] Figure 2 It is a flow chart of another embodiment of the automatic marking method of a marking robot of the present invention;

[0018] Figure 3 is a schematic diagram of the motion model of the line marking robot of the present invention;

[0019] Figure 4 It is a flow chart of another embodiment of the automatic marking method of a marking robot of the present invention;

[0020] Figure 5 It is a functional module schematic diagram of an embodiment of the marking robot of the present invention;

[0021] Figure 6 is a schematic structural diagram of an embodiment of a computer device of the present invention;

[0022] Figure 7 It is a schematic diagram of the structure of an embodiment of the storage medium of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] The terms "first", "second", "third" in this application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first", "second", "third" can expressly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. In the embodiments of this application, all directional indications (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the steps or units listed, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0025] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0026] Figure 1 FIG. 1 is a flow chart of an automatic marking method of a marking robot according to an embodiment of the present invention. It should be noted that the method of the present invention is not limited to the method of the present invention if substantially the same results are achieved. Figure 1 The process sequence shown is limited. Figure 1 As shown, the automatic marking method of the marking robot comprises the steps of:

[0027] Step S101: acquiring first position information of a marking robot in real time based on a global satellite navigation system.

[0028] It should be noted that the global satellite navigation system includes one of the Global Positioning System (GPS), the Beidou Satellite Navigation System (BDS), the GLONASS Satellite Navigation System (GLONASS), and the Galileo Satellite Navigation System (Galileo), and this embodiment does not limit this.

[0029] Specifically, the line-marking robot obtains its first position information in the global satellite navigation system in real time through its built-in positioning system. The first position information includes but is not limited to the latitude and longitude, altitude and movement speed of the line-marking robot. It should be understood that the size of the line-marking robot is usually small, far less than the size of a normal small car. Therefore, the position occupied by the line-marking robot relative to the road, especially the road with multiple lanes, is small, and the first position information is the approximate position information obtained through the global satellite navigation system, which can usually only locate the road information where the line-marking robot is located, and cannot locate the specific position of the line-marking robot on the road. Therefore, the first position information can only locate the road where the line-marking robot is located, and cannot accurately locate the specific position of the line-marking robot on the road.

[0030] Step S102: collecting a road image in front of the road marking robot, and analyzing the road image to obtain second position information of the road marking robot relative to the center point of the road.

[0031] Specifically, the road image in front of the marking robot is obtained by taking a monocular camera pre-configured on the marking robot. After acquiring the road image, the road image is analyzed to confirm the position of the road edge in the road image, and then the position of the road center point is confirmed based on the edges on both sides of the road. Then, the center point of the marking robot itself is used as a reference to confirm the distance between the road center point and the center point of the marking robot, as well as whether the marking robot is on the left or right side of the road center point, and obtain the second position information of the marking robot relative to the road center point. The second position information records the specific position of the marking robot on the road.

[0032] Furthermore, step S102 specifically includes:

[0033] 1. Use the image acquisition device pre-installed on the road marking robot to collect the road image in front of the road marking robot.

[0034] Specifically, the image acquisition device includes a monocular camera, which is arranged in front of the top of the road marking robot and is used to capture the road image in front of the road marking robot at a preset frame rate.

[0035] 2. Input the road image into the pre-trained Mask Representation RCNN neural network, and use the Mask Representation RCNN neural network to identify the road edge.

[0036] It should be noted that this embodiment uses the Mask representation RCNN neural network as the architecture to construct a road edge recognition model, which is mainly composed of the following parts: feature extraction module, region proposal module, classification and bounding box regression module and data preprocessing module. Among them, the data preprocessing module is used to crop the original image to an appropriate size and perform preprocessing steps such as normalization; the feature extraction module uses the ResNet network as a feature extractor to extract feature information in the image; the region proposal module is used to generate potential target areas, accurately pool the candidate areas, and maintain spatial consistency. The classification and bounding box regression module is used to determine the object category and the position mask branch generates a pixel-level binary mask for each target instance.

[0037] Among them, before using the Mask Representation RCNN neural network for road edge recognition, the Mask Representation RCNN neural network needs to be trained. The training process is as follows:

[0038] (1) Labeled data: Prepare a large number of labeled road edge datasets, with each image accompanied by the pixel coordinates of the edge line.

[0039] (2) Model initialization, loading pre-trained weights, and adjusting the output layer according to task requirements.

[0040] (3) Forward propagation: input the labeled road edge data into the model and calculate the prediction results through the model.

[0041] (4) Loss calculation: The total loss is calculated based on the difference between the pre-labeled true label and the predicted result.

[0042] (5) Back propagation, updating model parameters to minimize the loss.

[0043] (6) Iterative optimization: repeat the above steps until the model converges or reaches the predetermined number of iterations.

[0044] In order to better meet the needs of application scenarios, a comprehensive loss function cross entropy loss is customized to measure classification accuracy. The loss function specifically includes: bounding box regression loss, mask branch loss and custom loss term. The bounding box regression loss uses Smooth L1 Loss to evaluate the deviation between the predicted box and the true box; the mask branch loss uses binary cross entropy loss to evaluate the quality of mask prediction; the custom loss term introduces additional penalty terms, such as applying greater weights to pixels that are close to the edge of the road but not correctly marked, to improve edge detection accuracy.

[0045] 3. Confirm the location information of the road center point based on the road edge.

[0046] Specifically, after identifying the road edge position, the road edge position is substituted into a pre-constructed coordinate system, and the position information of the road center point in the coordinate system is calculated. The pre-constructed coordinate system is constructed with the vertical axis along the road travel direction and the horizontal axis perpendicular to the road travel direction. It should be understood that there are multiple road center points, which form a straight line in the coordinate system.

[0047] 4. Confirm the second position information of the road marking robot relative to the center point of the road based on the position information of the road marking robot itself and the position information of the center point of the road.

[0048] Specifically, the position of the marking robot is represented by the position of the center point of the marking robot itself in the coordinate system, the distance between the center point of the marking robot itself and the nearest center point of the road is calculated, and it is confirmed whether the marking robot is on the left or right side of the center point of the road to obtain the second position information of the marking robot relative to the center point of the road.

[0049] Step S103: Determine the precise position of the marking robot by combining the first position information and the second position information.

[0050] Specifically, by combining the first position information and the second position information, the specific road where the line marking robot is located and the specific position of the line marking robot in the road can be acquired, thereby confirming the precise position of the line marking robot.

[0051] Step S104: Based on the precise position and the preset planned path, the marking robot is controlled to automatically drive and perform the road marking operation at the same time.

[0052] It should be noted that the preset planned path is pre-set and stored in the built-in memory of the marking robot. When the marking robot performs the marking operation, it first reads the preset planned path corresponding to the marking operation, and then controls the marking robot to automatically drive according to the preset planned path and mark the line synchronously to complete the marking operation.

[0053] Specifically, after the precise position of the line marking robot is acquired, the precise position is combined with a preset planned path, so that the line marking robot can be accurately controlled to travel along the preset planned path, thereby avoiding large deviations of the line marking robot during the travel process.

[0054] The automatic marking method for a marking robot of the present embodiment obtains the first position information of the marking robot by using the global satellite navigation system, obtains the road image in front of the marking robot by using the image acquisition device of the marking robot itself, confirms the second position information of the marking robot relative to the center point of the road from the road image, fuses the first position information and the second position information to obtain the precise position of the marking robot on the road, controls the marking robot to automatically drive in combination with the precise position and the preset planned path, and simultaneously performs the road marking operation, which realizes the precise positioning of the marking robot on the road, and then controls the marking robot to automatically drive according to the precise positioning and the preset planned path, avoiding a large deviation between the travel route of the marking robot and the preset planned path, thereby ensuring that the marking operation can be accurately performed, the marking accuracy is higher, and no preparation work is required in advance, and the marking efficiency is high.

[0055] Furthermore, in order to enable the marking robot to quickly respond to external changes during its movement and improve its intelligence level. Figure 2 As shown, after step S104, the following steps are also included:

[0056] Step S201: Determine whether the marking robot is offset based on the precise position and the preset planned path. If yes, execute step S202. If no, continue to execute the step of controlling the marking robot to automatically travel based on the precise position and the preset planned path.

[0057] Specifically, the preset position corresponding to the current precise position of the marking robot is confirmed from the preset planned path, and then the precise position and the preset position are compared to determine whether the distance difference between the two reaches the preset threshold. If it reaches, it is considered that the marking robot has deviated from the preset planned path and needs to be corrected; if it does not reach, it is considered that the current route of the marking robot has not deviated from the preset planned path.

[0058] Step S202: adjusting the forward direction of the marking robot based on the precise position and the preset planned path.

[0059] Specifically, when the line marking robot deviates from the preset planned path, the forward direction of the line marking robot is adjusted so that the line marking robot returns to the preset planned path.

[0060] Furthermore, step S202 specifically includes:

[0061] 1. Calculate the first deviation distance and first deviation angle between the precise position and the corresponding position in the preset planned path.

[0062] Specifically, by substituting the precise position and its corresponding position in the preset planned path into the same preset coordinate system, a first deviation distance and a first deviation angle of the precise position relative to the corresponding position can be calculated.

[0063] 2. Input the first deviation distance and the first deviation angle into the basic motion model of the marking robot to adjust the forward direction of the marking robot.

[0064] Specifically, the basic motion model of the marking robot is modeled as follows: Figure 3 As shown, the parameters are expressed as:

[0065] V R Indicates the right wheel speed, V L Indicates the left wheel speed;

[0066] x, y represent the relative position of the marking robot in the inertial system;

[0067] θ represents the relative position of the marking robot in the inertial coordinate system;

[0068] V lin Indicates the target linear velocity of the marking robot;

[0069] ω represents the target angular velocity of the marking robot;

[0070] L represents the distance between the left and right wheels (wheel spacing);

[0071] r represents the wheel radius;

[0072] d represents the wheel diameter.

[0073] A mathematical model of the mobile kinematics of the marking robot is established. The movement speed is taken as the research variable in the mathematical model, and the speed is converted into a coordinate vector in a general sense by the following equation. The specific formula is shown in the following formula:

[0074]

[0075] The relationship between the speed of the marking robot and the angular velocity of the left and right wheels is expressed as:

[0076]

[0077] If the marking robot rotates n times with its own center point as the origin as a known condition, and θ is used as the angle generated by the two-wheel differential robot spinning with the left and right wheels as the axis, the accumulated length of the left and right wheels rotating during the spinning process is l, which can be expressed as:

[0078]

[0079] For the convenience of application and debugging, let the marking robot spin an integer number of circles, and assume that the spin angle θ is an integer multiple of 2π. Assuming that the robot rotates M circles, the spin angle θ is:

[0080] θ=2πM;

[0081] In order to enable the marking robot to spin to an angle of θ, the encoder function of the DC motor needs to be used for counting in the actual system design. Assuming that the count value of the robot spinning one circle is k, if the motor count value is K when the driving wheel rotates n circles, the value of K is:

[0082] K = nk;

[0083] Combining the expressions of accumulation length l and spin angle θ, we can obtain:

[0084]

[0085] According to the formula listed above, we must first consider the relationship between the speed of each wheel and the structural parameters of the marking robot, such as the axle spacing L between the two wheels and the radius r of the wheel. Secondly, the moving speed of the marking robot is described based on the interaction between the wheel speed and the chassis structure of the marking robot. In the present invention, the motion system of the marking robot is a two-wheel differential moving system. The chassis of the car is composed of a black rectangular composite material plate. Compared with the bottom plate made of aluminum alloy material, the weight is reduced and the driving pressure of the car is reduced, thereby reducing the impact on the size of the motor torque in the control system.

[0086] Furthermore, in order to further improve the automation level of the marking robot, in some other embodiments, Figure 4 As shown, after step S104, the following steps are also included:

[0087] Step S301: Divide the road image into multiple areas using a pre-trained target detection algorithm.

[0088] It should be noted that the target detection algorithm can be implemented based on neural network algorithms such as R-CNN network (Regions with Convolutional Neural Networks), YOLO network (You Only Look Once), and SSD network (Single Shot MultiBox Detector).

[0089] Specifically, after a road image is input into a target detection algorithm through a pre-trained target detection algorithm, the target detection algorithm divides the road image into multiple confirmations, and confirms the area on the preset planned path.

[0090] Step S302: Determine whether there are obstacles in the area on the preset planned path. If yes, execute step S303. If no, continue to execute the step of controlling the automatic driving of the marking robot based on the precise position and the preset planned path.

[0091] Step S303: avoiding obstacles based on a preset obstacle avoidance method.

[0092] Specifically, if there are obstacles in the area on the preset planned path, it is necessary to control the marking robot to perform obstacle avoidance operations, so as to avoid the obstacles and continue to perform the marking operation.

[0093] Furthermore, step S303 specifically includes:

[0094] 1. Based on the depth estimation algorithm, the distance between the marking robot and the obstacle is calculated using the feature points and texture information in the road image.

[0095] Specifically, the depth estimation algorithm is implemented using a monocular depth estimation algorithm based on deep learning.

[0096] 2. When the distance is less than the preset distance threshold, the marking robot is controlled to stop urgently.

[0097] Specifically, when the distance is less than a preset distance threshold, it means that the distance between the marking robot and the obstacle is too close to replan the path to avoid the obstacle. Therefore, the marking robot is controlled to stop suddenly to avoid directly hitting the obstacle.

[0098] 3. When the distance is greater than or equal to the preset distance threshold, the marking robot is controlled to slow down, the outline of the obstacle is identified using the target detection algorithm, and an obstacle avoidance path is generated based on the outline and pre-acquired map data, and the marking robot is controlled to move forward based on the obstacle avoidance path.

[0099] Specifically, when the distance is greater than or equal to the preset distance threshold, it means that the distance between the marking robot and the obstacle is far, and there is enough space for the marking robot to re-plan the path to avoid the obstacle. Before re-planning the path, the marking robot must be controlled to slow down to prevent the marking robot from approaching the obstacle quickly. Then, the outline of the obstacle is identified through the target detection algorithm, and the obstacle avoidance path is generated in combination with the map data of the current location of the marking robot. Then, the marking robot is controlled to move forward according to the obstacle avoidance path to avoid the obstacle.

[0100] Further, after the marking robot is controlled to move forward based on the obstacle avoidance path, the method further includes:

[0101] 3.1. After the marking robot bypasses the obstacle, the second deviation distance and the second deviation angle between the precise position and the corresponding position in the preset planned path are calculated.

[0102] 3.2. Input the second deviation distance and the second deviation angle into the basic motion model of the marking robot to adjust the forward direction of the marking robot.

[0103] Specifically, after the marking robot avoids the obstacle, it is still necessary to control the marking robot to return to the preset planned path to continue the marking operation. Therefore, this embodiment calculates the second deviation distance and the second deviation angle between the marking robot and the preset planned path, and then controls the marking robot to move forward according to the second deviation distance and the second deviation angle, so that the marking robot returns to the preset planned path to continue the marking operation.

[0104] Figure 5 Schematic diagram of the functional modules of the line marking robot according to an embodiment of the present invention. Figure 5 As shown, the marking robot 20 includes an acquisition module 21 , a collection module 22 , a confirmation module 23 and a control module 24 .

[0105] An acquisition module 21 is used to acquire first position information of the marking robot in real time based on a global satellite navigation system;

[0106] The acquisition module 22 is used to acquire the road image in front of the road marking robot and analyze the road image to obtain the second position information of the road marking robot relative to the center point of the road;

[0107] A confirmation module 23, used to confirm the precise position of the marking robot in combination with the first position information and the second position information;

[0108] The control module 24 is used to control the road marking robot to automatically travel based on the precise position and the preset planned path, and to perform the road marking operation at the same time.

[0109] Optionally, the acquisition module 22 executes the operation of acquiring the road image in front of the line marking robot and analyzing the road image to obtain the second position information of the line marking robot relative to the center point of the road, specifically including: using the image acquisition device pre-set on the line marking robot to acquire the road image in front of the line marking robot; inputting the road image into a pre-trained Mask representation RCNN neural network, and using the Mask representation RCNN neural network to identify the road edge; confirming the position information of the center point of the road according to the road edge; and confirming the second position information of the line marking robot relative to the center point of the road based on the position information of the line marking robot itself and the position information of the center point of the road.

[0110] Optionally, after the control module 24 executes the operation of controlling the automatic driving of the marking robot based on the precise position and the preset planned path, it is also used to: determine whether the marking robot is offset based on the precise position and the preset planned path; if so, adjust the forward direction of the marking robot based on the precise position and the preset planned path.

[0111] Optionally, the control module 24 performs operations to adjust the forward direction of the marking robot based on the precise position and the preset planned path, specifically including: calculating a first deviation distance and a first deviation angle between the precise position and the corresponding position in the preset planned path; inputting the first deviation distance and the first deviation angle into the basic motion model of the marking robot to adjust the forward direction of the marking robot.

[0112] Optionally, after the control module 24 executes the operation of controlling the automatic driving of the marking robot based on the precise position and the preset planned path, it is also used to: divide the road image into multiple areas using a pre-trained target detection algorithm; determine whether there are obstacles in the area on the preset planned path; if so, avoid the obstacles based on a preset obstacle avoidance method.

[0113] Optionally, the control module 24 performs operations to avoid obstacles based on a preset obstacle avoidance method, specifically including: based on a depth estimation algorithm, using feature points and texture information in the road image to calculate the distance between the line marking robot and the obstacle; when the distance is less than a preset distance threshold, controlling the line marking robot to stop suddenly; when the distance is greater than or equal to the preset distance threshold, controlling the line marking robot to decelerate, using a target detection algorithm to identify the outline of the obstacle, and generating an obstacle avoidance path based on the outline and pre-acquired map data, and controlling the line marking robot to move forward based on the obstacle avoidance path.

[0114] Optionally, after the control module 24 executes the operation of controlling the line marking robot to move forward based on the obstacle avoidance path, it is also used to: calculate the second deviation distance and the second deviation angle between the precise position and the corresponding position in the preset planned path after the line marking robot bypasses the obstacle; input the second deviation distance and the second deviation angle into the basic motion model of the line marking robot to adjust the forward direction of the line marking robot.

[0115] For other details about the technical solutions for implementing each module in the marking robot in the above embodiment, please refer to the description of the automatic marking method for the marking robot in the above embodiment, which will not be repeated here.

[0116] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0117] See also Figure 6 , Figure 6 FIG. 1 is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Figure 6As shown, the computer device 30 includes a processor 31 and a memory 32 coupled to the processor 31. The memory 32 stores program instructions. When the program instructions are executed by the processor 31, the processor 31 executes the steps of the automatic marking method of the marking robot described in any of the above embodiments.

[0118] The processor 31 may also be referred to as a CPU (Central Processing Unit). The processor 31 may be an integrated circuit chip having signal processing capabilities. The processor 31 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0119] See also Figure 7 , Figure 7 It is a schematic diagram of the structure of the storage medium of an embodiment of the present invention. The storage medium of the embodiment of the present invention stores program instructions 41 that can realize the automatic marking method of the marking robot, wherein the program instructions 41 can be stored in the above storage medium in the form of a software product, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read means Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes, or computer equipment such as computers, servers, mobile phones, tablets, etc.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed computer equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0121] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for automatic marking by a marking robot, characterized in that: It includes: Acquire the first position information of the marking robot in real time based on the global satellite navigation system; Collecting a road image in front of the road marking robot and analyzing the road image to obtain second position information of the road marking robot relative to a center point of the road; Determining the precise position of the marking robot by combining the first position information and the second position information; The road marking robot is controlled to automatically travel based on the precise position and the preset planned path, while performing the road marking operation.

2. The automatic marking method of a marking robot according to claim 1, characterized in that: The collecting of the road image in front of the road marking robot and analyzing the road image to obtain the second position information of the road marking robot relative to the center point of the road includes: Using an image acquisition device pre-set on the road marking robot to acquire a road image in front of the road marking robot; Inputting the road image into a pre-trained Mask Representation RCNN neural network, and using the Mask Representation RCNN neural network to identify road edges; confirming the location information of the center point of the road according to the edge of the road; Second position information of the line marking robot relative to the center point of the road is confirmed based on the position information of the line marking robot itself and the position information of the center point of the road.

3. The automatic marking method of a marking robot according to claim 1, characterized in that: After controlling the marking robot to automatically travel based on the precise position and the preset planned path, the method further includes: Determining whether the marking robot is offset based on the precise position and the preset planned path; If so, the forward direction of the marking robot is adjusted based on the precise position and the preset planned path.

4. The automatic marking method using a marking robot according to claim 3, characterized in that: The step of adjusting the forward direction of the marking robot based on the precise position and the preset planned path includes: Calculating a first deviation distance and a first deviation angle between the precise position and a corresponding position in the preset planned path; The first deviation distance and the first deviation angle are input into a basic motion model of the line marking robot to adjust the forward direction of the line marking robot.

5. The automatic marking method using a marking robot according to claim 1, characterized in that: After controlling the marking robot to automatically travel based on the precise position and the preset planned path, the method further includes: Dividing the road image into a plurality of regions using a pre-trained object detection algorithm; Determine whether there are obstacles in the area on the preset planned path; If so, avoid the obstacle based on a preset obstacle avoidance method.

6. The automatic marking method using a marking robot according to claim 5, characterized in that: The avoiding the obstacle based on a preset obstacle avoidance method includes: Based on a depth estimation algorithm, the distance between the lane marking robot and the obstacle is calculated using feature points and texture information in the road image; When the distance is less than a preset distance threshold, controlling the marking robot to stop urgently; When the distance is greater than or equal to the preset distance threshold, the line marking robot is controlled to slow down, the outline of the obstacle is identified using the target detection algorithm, an obstacle avoidance path is generated based on the outline and pre-acquired map data, and the line marking robot is controlled to move forward based on the obstacle avoidance path.

7. The automatic marking method using a marking robot according to claim 6, characterized in that: After controlling the line marking robot to move forward based on the obstacle avoidance path, the method further includes: After the marking robot bypasses the obstacle, calculating a second deviation distance and a second deviation angle between the precise position and a corresponding position in the preset planned path; The second deviation distance and the second deviation angle are input into a basic motion model of the line marking robot to adjust the forward direction of the line marking robot.

8. A marking robot, characterized in that: It includes: An acquisition module, used for acquiring the first position information of the marking robot in real time based on a global satellite navigation system; A collection module, used for collecting a road image in front of the road marking robot, and analyzing the road image to obtain second position information of the road marking robot relative to the center point of the road; A confirmation module, used to confirm the precise position of the marking robot in combination with the first position information and the second position information; The control module is used to control the road marking robot to automatically travel based on the precise position and the preset planned path, and to perform the road marking operation at the same time.

9. A computer device, characterized in that: The computer device includes a processor and a memory coupled to the processor, wherein program instructions are stored in the memory. When the program instructions are executed by the processor, the processor executes the steps of the automatic marking method of a marking robot as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The device stores program instructions capable of implementing the automatic marking method of a marking robot as described in any one of claims 1 to 7.

Images