Automatic centering method for vehicle bottom inspection robot

Abstract

The invention provides an automatic centering method for a vehicle bottom inspection robot, which comprises the following steps of: 1, judging whether the robot is in an initial state or not according to an image acquired by a visual device, if so, executing a step 2, and otherwise, adjusting the robot to enter the initial state; step 2, the robot rotates according to the position of a target license plate in the image acquired by the visual device; 3, the robot translates according to the position of the target license plate in the image acquired by the visual device; and 4, the step 2 and the step 3 are executed repeatedly till the preset condition is met, and automatic centering of the vehicle bottom inspection robot is completed.

Description

Technical Field

[0001] The invention relates to a robot centering method, in particular to an automatic centering method of a vehicle bottom inspection robot. Background Art

[0002] This section merely provides background information related to the present disclosure and is not necessarily prior art.

[0003] Traditional under-vehicle inspection methods such as manual inspections and fixed camera monitoring can meet the needs to a certain extent, but they are often restricted by factors such as large detection blind spots, low efficiency, space limitations, operational difficulties and inspection efficiency. With the rapid development of science and technology, intelligence and automation have become the goals pursued by various industries. In the field of traffic management, comprehensive, efficient and accurate inspection of vehicles has become a key link in improving traffic safety and efficiency. Therefore, it is very important to develop a flexible and efficient under-vehicle inspection robot.

[0004] Traditional underbody inspection robots need to be moved to the front of the vehicle, and then manually rotated and moved multiple times to make the robot's center axis coincide with the center axis of the vehicle chassis before entering the vehicle for inspection, which is time-consuming and laborious. Especially for underbody inspection robots with Ackerman structures, since they cannot be translated left and right, it is more troublesome to manually align the robot with the center axis of the vehicle chassis.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0006] Purpose of the invention: The technical problem to be solved by the present invention is to provide an automatic centering method for an underbody inspection robot in view of the deficiencies in the prior art.

[0007] In order to solve the above technical problems, the present invention discloses an automatic centering method of an underbody inspection robot, comprising the following steps:

[0008] Step 1, judging whether the robot is in an initial state according to an image acquired by a visual device arranged on the robot, if so, executing step 2, otherwise adjusting the robot to enter the initial state;

[0009] Step 2, the robot rotates according to the position of the target license plate in the image acquired by the visual device;

[0010] Step 3, the robot performs translation according to the position of the target license plate in the image acquired by the visual device;

[0011] Step 4, repeat steps 2 and 3 until the preset conditions are met, and the automatic centering of the underbody inspection robot is completed.

[0012] Furthermore, the initial state described in step 1, that is, the distance between the robot and the target license plate is within a preset threshold.

[0013] Furthermore, the step of adjusting the robot to enter the initial state in step 1 includes:

[0014] Step 1-1, judging the image in the field of view of the visual device in the robot, if there is a license plate, executing step 1-2, otherwise executing step 1-3;

[0015] Step 1-2, moving in the direction of the license plate until the distance to the target license plate is within a preset threshold, entering the initial state, and executing step 2;

[0016] Step 1-3, the robot rotates until the license plate appears in the field of view, and then executes step 1-2 again.

[0017] Furthermore, the robot in step 2 rotates, including:

[0018] Step 2-1, assuming that the width of the image acquired by the visual device is x and the height is y, use the image recognition method to obtain that the center point of the target license plate is located at point S, and the upper left corner of the image is the coordinate origin, and the coordinates of point S are (u, v);

[0019] Step 2-2, calculate the rotation angle;

[0020] Step 2-3, the robot rotates according to the rotation angle.

[0021] Further, the calculation of the rotation angle described in step 2-2 is as follows:

[0022]

[0023] Wherein, α is the rotation angle, and β is the horizontal field of view angle of the visual device.

[0024] Furthermore, the robot in step 2 performs translation, including:

[0025] Step 3-1, using an image recognition method to obtain a bounding box of the target license plate with a width of w and a height of h;

[0026] Step 3-2, calculate the translation distance, as follows:

[0027]

[0028] Among them, d is the translation distance, abs() is the absolute value function, maxW2H is the license plate type parameter, and exp is the empirical value parameter;

[0029] Step 3-3, determine the translation direction;

[0030] Step 3-4, the robot performs translation according to the translation distance and the translation direction.

[0031] Furthermore, the determination of the translation direction described in step 3-3 is as follows:

[0032] Step 3-3-1, transforming the image within the bounding box of the target license plate into an HSV image;

[0033] Step 3-3-2, respectively calculating the average depth of the left half of the pixels and the average depth of the right half of the pixels in the HSV image and comparing them;

[0034] Step 3-3-3, if the average depth of the left half of the pixels is less than the average depth of the right half of the license plate, the translation direction is to the right, otherwise the translation direction is to the left.

[0035] Furthermore, the preset conditions in step 4 include:

[0036] The rotation angle in step 2 is less than a preset rotation angle threshold, and the translation distance in step 3 is less than a preset translation distance threshold.

[0037] Furthermore, the license plate type parameter in step 3-2 is the ratio of the actual width and height of the target license plate, specifically including:

[0038] For blue license plates, the license plate type parameter value is 440 / 140, and for new energy license plates, the license plate type parameter value is 480 / 140.

[0039] Furthermore, the experience value parameter exp described in step 3-2 is set to 200.

[0040] Beneficial effects:

[0041] 1. The automatic centering method proposed in the present invention can effectively solve the efficiency problem of manually controlling the underbody inspection robot to align with the center axis of the vehicle chassis. As long as the front camera on the robot can see the license plate, it can automatically align with the center axis of the vehicle.

[0042] 2. The automatic centering method of the underbody scanning robot proposed in the present invention is derived from the urgent need for efficient and safe underbody inspection solutions. It utilizes YOLO target detection, image processing and other algorithms to truly realize the automation of underbody inspection and greatly improve the efficiency of underbody inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0044] Figure 1 It is a schematic diagram of the robot in the present invention.

[0045] Figure 2 It is a schematic diagram of the initial state.

[0046] Figure 3 It is a schematic flow chart of the automatic centering method of the present invention.

[0047] Figure 4 This is a schematic diagram of the rotation angle required for the robot to go from S1 to S2 state.

[0048] Figure 5 This is a schematic diagram of the distance the robot needs to move from S2 to S3. DETAILED DESCRIPTION

[0049] The present invention provides a new automatic method for an under-vehicle inspection robot, which realizes the automation of the under-vehicle inspection robot when performing tasks, that is, as long as the robot can see the license plate, after clicking a one-key inspection button, the robot can automatically move and rotate to align with the center axis of the chassis of the inspected vehicle, and then complete functions such as straight-line driving and scanning and drawing in sequence.

[0050] The general idea of ​​the present invention is as follows:

[0051] The wheels of the undercar inspection robot are Mecanum wheels, which can move forward and backward, left and right, and rotate left and right. There is a camera at a 45-degree angle to the horizontal line in front of the undercar scanning robot to identify the license plate. The upper computer inside the robot will obtain the image from the camera and run the YOLO license plate recognition algorithm to mark the position of the license plate in the image and other information. The robot determines the rotation angle and the distance to move left and right based on the position of the license plate in the image, thereby achieving automatic centering.

[0052] Example:

[0053] like Figure 1 As shown, it is a vehicle bottom inspection robot. In the figure, a camera is provided in front of the robot, which is inclined upward with respect to the horizontal line, preferably at 45 degrees.

[0054] like Figure 2 The figure shows the robot in S1 state.

[0055] like Figure 3 The figure shows the entire process from the initial state where the robot does not align with the license plate to the final state where the robot aligns with the license plate. Figure 3It can be seen that the entire process of the robot's automatic centering includes two processes: rotation and left and right translation. The specific steps are as follows:

[0056] Initial state: Set the initial position, the vehicle to be detected is in any direction of the robot; control the robot to move to a position about 1m away from the license plate. If the robot cannot see the license plate, it will rotate to find the license plate and change to S1 state.

[0057] Step 1: The robot rotates, such as Figure 4 As shown, the specific method is as follows:

[0058] Assume the robot is in Figure 3 In the initial state, the image output by YOLO license plate recognition is as follows Figure 4 As shown, the output image width is x, the height is y, the center point of the license plate is located at point S, the upper left corner is the origin of the coordinates, and the coordinates of point S are (u, v); when the robot is in Figure 2 When the robot is in the S2 state, the center point of the license plate is point D, and the coordinates are (x / 2, v). The angle that the robot needs to rotate from the S1 state to the S2 state is α, and the calculation formula is:

[0059]

[0060] Where β is the horizontal field of view of the camera.

[0061] Step 2: The robot moves horizontally. The specific method is as follows:

[0062] Assume the robot is located at Figure 3 In the S2 state, the image output by YOLO license plate recognition should be as follows Figure 5 As shown, at this time, the center point of the license plate should be located at the center of the camera's horizontal viewing angle. Assume that the width of the bounding box marked by YOLO is w and the height is h. If the ratio of w to h is smaller, it means that the robot needs to move a greater distance. On the contrary, when w / h is larger, even close to the width and height ratio of the license plate itself, it means that the robot is closer to the center state S6. Therefore, the robot's left and right movement distance (in millimeters) can be calculated by the following formula:

[0063]

[0064] Among them, abs() is the absolute value function; maxW2H is the ratio of the width and height of the license plate itself. For blue license plates, this value is 440 / 140, and for new energy license plates, this value is 480 / 140; exp is an empirical value, usually 200.

[0065] In addition, it is also necessary to determine the direction in which the robot is moving. First, in order to accurately separate the license plate from the image output by YOLO license plate recognition, opencv is used to transform the original RGB image into an HSV image, and then extract Figure 5 The image in the red frame is calculated respectively for the average depth of the pixels of the left half of the license plate and the average depth of the pixels of the right half of the license plate. If the average depth of the pixels of the left half of the license plate is less than the average depth of the pixels of the right half of the license plate, it means that the robot needs to move to the right, otherwise, the robot moves to the left.

[0066] Step 3: Repeat steps 1 and 2 until the centering is completed. The specific method is as follows:

[0067] When the robot's moving distance and rotation angle are both less than a certain threshold, the robot is considered to have completed automatic centering. Otherwise, the previous rotation and translation process will be repeated until automatic centering is completed. When automatic centering is completed, the robot can perform the following functions such as straight-line driving and scanning.

[0068] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, the invention content of the automatic centering method of a vehicle underbody inspection robot provided by the present invention and some or all steps in each embodiment can be executed. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0069] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of computer programs and their corresponding general hardware platforms. Based on such an understanding, the technical solutions in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of computer programs, i.e., software products, which can be stored in a storage medium and include several instructions for enabling a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, an MCU or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0070] The present invention provides a method and idea for the automatic centering of a vehicle underbody inspection robot. There are many methods and approaches to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A method for automatically centering a vehicle underbody inspection robot, characterized in that: The following steps are involved: Step 1, judging whether the robot is in an initial state according to an image acquired by a visual device arranged on the robot, if so, executing step 2, otherwise adjusting the robot to enter the initial state; Step 2, the robot rotates according to the position of the target license plate in the image acquired by the visual device; Step 3, the robot performs translation according to the position of the target license plate in the image acquired by the visual device; Step 4, repeat steps 2 and 3 until the preset conditions are met, and the automatic centering of the underbody inspection robot is completed.

2. The automatic centering method of a vehicle underbody inspection robot according to claim 1, characterized in that: The initial state described in step 1, that is, the distance between the robot and the target license plate is within a preset threshold.

3. The automatic centering method of the underbody inspection robot according to claim 2, characterized in that: The step 1 of adjusting the robot to enter the initial state includes: Step 1-1, judging the image in the field of view of the visual device in the robot, if there is a license plate, executing step 1-2, otherwise executing step 1-3; Step 1-2, moving in the direction of the license plate until the distance to the target license plate is within a preset threshold, entering the initial state, and executing step 2; Step 1-3, the robot rotates until the license plate appears in the field of view, and then executes step 1-2 again.

4. The automatic centering method of the underbody inspection robot according to claim 3 is characterized in that: The robot in step 2 rotates, including: Step 2-1, assuming that the width of the image acquired by the visual device is x and the height is y, use the image recognition method to obtain that the center point of the target license plate is located at point S, and the upper left corner of the image is the coordinate origin, and the coordinates of point S are (u, v); Step 2-2, calculate the rotation angle; Step 2-3, the robot rotates according to the rotation angle.

5. The automatic centering method of the underbody inspection robot according to claim 4, characterized in that: The calculation rotation angle described in step 2-2 is as follows: Wherein, α is the rotation angle, and β is the horizontal field of view angle of the visual device.

6. The automatic centering method of the underbody inspection robot according to claim 5, characterized in that: The robot performs translation in step 2, including: Step 3-1, using an image recognition method to obtain a bounding box of the target license plate with a width of w and a height of h; Step 3-2, calculate the translation distance, as follows: Among them, d is the translation distance, abs() is the absolute value function, maxW2H is the license plate type parameter, and exp is the empirical value parameter; Step 3-3, determine the translation direction; Step 3-4, the robot performs translation according to the translation distance and the translation direction.

7. The automatic centering method of the underbody inspection robot according to claim 6, characterized in that: The determination of the translation direction described in step 3-3 is as follows: Step 3-3-1, transforming the image within the bounding box of the target license plate into an HSV image; Step 3-3-2, respectively calculating the average depth of the left half of the pixels and the average depth of the right half of the pixels in the HSV image and comparing them; Step 3-3-3, if the average depth of the left half of the pixels is less than the average depth of the right half of the license plate, the translation direction is to the right, otherwise the translation direction is to the left.

8. The automatic centering method of the underbody inspection robot according to claim 7, characterized in that: The preset conditions described in step 4 include: The rotation angle in step 2 is less than a preset rotation angle threshold, and the translation distance in step 3 is less than a preset translation distance threshold.

9. The automatic centering method of the underbody inspection robot according to claim 8, characterized in that: The license plate type parameter in step 3-2 is the ratio of the actual width and height of the target license plate, specifically including: For blue license plates, the license plate type parameter value is 440 / 140, and for new energy license plates, the license plate type parameter value is 480 / 140.

10. The automatic centering method of the underbody inspection robot according to claim 9, characterized in that: The experience value parameter exp described in step 3-2 is set to 200.

Images