Device and method for bearing positioning of robot moving platform

Abstract

The invention provides a device and method for bearing positioning of a robot moving platform, and relates to the field of teaching-free welding. Comprising a platform positioning device and a mobile platform. The platform positioning device mainly comprises a shell, a movable lamp panel, an industrial personal computer, a high-precision visual sensor, a high-precision acceleration sensor used for obtaining the pose angle of the visual sensor, a two-axis rotating motor, a power module and a wireless communication module used for data transmission between the positioning device and the platform. The high-precision acceleration sensor and the two-axis rotating motor are well fixed to the visual sensor. And a mobile platform is carried, so that the time cost investment of deploying a ground rail when the work amount of a current welding project is large can be solved, and the method is more flexible compared with the ground rail. According to the teaching-free welding robot, only the lamp panel needs to be fixed, positioning calibration is conducted for the first time, the robot can be normally borne to a designated place, accurate position information is provided, teaching-free welding work can be suitable for more scenes, the autonomous decision-making process is more complete, and therefore the welding efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of mobile welding, which can effectively reduce the cost input in the early stage of welding large workpieces, improve the flexibility of project scenarios, and standardize the production operation process. Background Art

[0002] With the application of industrial robots in various fields, especially in the welding field, robotic automated welding workstations have gradually replaced manual welding. However, in existing welding projects, in the face of large or numerous workpieces, it is generally necessary to pre-lay a ground rail, that is, the robot moves on a fixed ground rail to eliminate the disadvantage that the reach of a single robot arm is insufficient to cover the entire workpiece. This method has high requirements for the production environment space and high upfront time costs. After the production project ends, the materials are difficult to reuse, and it is inconvenient to adjust, making it difficult to meet the production operation requirements of multi-scene high efficiency. Summary of the Invention

[0003] The purpose of the present invention is to assist in positioning devices and methods for carrying a robot mobile platform to solve the problem in the above background art that the existing method of laying a ground rail is difficult to meet the production operation requirements of multi-scene high efficiency.

[0004] To achieve the above purpose, according to one aspect disclosed in the present invention, a device and method for collaborative positioning of a robot platform are provided:

[0005] The main body of the present invention includes Figure 1 and Figure 2 , the space locator main body includes: a high-precision vision sensor 1, a Z-axis control rotating motor 2, a Y-axis control rotating motor 3, an industrial computer main body 4, and a high-precision acceleration sensor 5. The method includes the following steps, characterized in that:

[0006] S1. Visual sensor and light board calibration:

[0007] Fix the positions of the light board and the calibration board, make the visual sensor frame contain two complete parts, change the angle to take multiple groups of pictures of the light board and the calibration board, and obtain the internal parameters and distortion parameters of the visual sensor and the three-dimensional data of the light board through the Zhang-Zhengyou calibration method and the calculation of PnP.

[0008] Fix the positioning device body on the robot mobile platform. The robot has an "eye-in-hand" camera sensor. Fix the robot pose, operate the mobile platform, and keep the visual sensor of the positioning device and the camera sensor on the robot always contain a complete fixed-position calibration board in their frames respectively. Adjust the position of the mobile platform, take multiple groups of pictures, obtain the pose relationship between the two sensors through solving PnP, and calculate the rotation matrix and offset vector between the positioning device and the mobile platform in combination with the robot TCP coordinates and the hand-eye calibration result.

[0009] Among them, a C++ code example for binocular calibration is as follows:

[0010]

[0011]

[0012]

[0013]

[0014]

[0015] S2. Pre-positioning:

[0016] Fix the light board at a suitable position, move the platform to a stop 1 meter directly in front of the light board, operate the software to take a picture of the light board at this time, and mark it as the origin position;

[0017] S3. Obtain the real-time position:

[0018] When the moving platform travels to the pending working area, the vision sensor calculates the rotation matrix and offset vector between the light board and the moving platform based on the light board picture at this time, and then the accurate position of the moving platform can be known;

[0019] Among them, a C++ partial example code for calculating the relationship between the vision sensor coordinate system and the light board coordinate system through the checkerboard picture is as follows:

[0020]

[0021]

[0022]

[0023] S4. Send data to the robot:

[0024] At this time, that is, when the six-axis robot has an external axis, the pose matrix and offset data will be transmitted in real time to the robot system to assist the robot to perform related operations more flexibly.

[0025] As a further improvement of the present invention, step S3 includes the following process:

[0026] Taking the zero position of the robot as the reference point, establish the spatial coordinates of the origin position;

[0027] a. Obtain the real-time image of the light board on the vision sensor, and according to the pixel size of the light board on the sensor and the proportion of the entire graph on the target surface, adjust the focus and ROI of the sensor in real time, and at the same time rotate the motor to keep the light board graph in the center of the picture;

[0028] b. Calculate the pose relationship between the light board and the vision sensor chip at this time. Combine the rotation angle of the vision sensor provided by the high-precision acceleration sensor to calculate the rotation matrix of the light board relative to the origin.

[0029] c. According to the pixel size and area ratio of the light board, calculate the offset vector of the light board relative to the vision sensor by the triangulation model.

[0030] d. Combine the pose relationship between the vision sensor and the camera sensor of the robot hand, and the rotation matrix and offset vector of the mobile platform and the origin position at this time can be calculated, that is, the accurate position of the mobile platform at this moment.

[0031] As a further improvement of the present invention, the wireless communication module in step S4 communicates with the welding robot through a computer communication protocol.

[0032] As a further improvement of the present invention, the wireless communication module can instantaneously receive the vision sensor position information and attitude information of the high-precision acceleration sensor module and communicate with the robot.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] The present invention does not require a specific working environment, has no customized requirements for the size of the production workpiece, can perform rapid standardized production, and has a graphical operation interface for spatial positioning operations, making it easier for non-programmers to operate the mobile platform and obtain position information to complete the specified welding engineering project. It can operate in multiple scenarios and can easily handle the switching between different projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The disclosure of the present invention will be described with reference to the accompanying drawings. It should be understood that the drawings are only for illustrative purposes and are not intended to limit the protection scope of the present invention. In the drawings, the same reference numerals are used to refer to the same components. Among them:

[0036] Figure 1 is a schematic diagram of the overall structure of the present invention, showing the layout of the core components of the spatial locator body and the mobile platform;

[0037] Figure 2 is a schematic diagram of the light board structure of the present invention, specifically presenting the geometric features and the distribution of positioning feature points of the indicator light board;

[0038] Figure 3 is a schematic diagram of the graphical operation interface of the present invention, showing the real-time positioning data visualization interface and the control module interaction panel;

[0039] Reference numerals in the figure: 1. High-precision vision sensor; 2. Z-axis rotation motor; 3. Y-axis rotation motor; 4. Industrial control computer; 5. High-precision acceleration sensor; Detailed implementation mode

[0040] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0041] In one embodiment, the method includes the following steps:

[0042] S1. Operate two sensors to take pictures, calibrate the pose relationship, and record the internal parameters and distortion coefficients of the cameras. Here, it includes the parameter coefficients of the vision sensors themselves, as well as the rotation matrix and offset vector between the two sensors;

[0043] S2. Set the origin of the vision sensor, that is, the origin of the mobile platform. The specific method is to place the mobile platform equipped with the vision sensor about 1 meter in front of the auxiliary positioning device, obtain the image information of the auxiliary positioning device at this time through the vision sensor, and combine the parameters of S1 to obtain the position and attitude information of the vision sensor in the dot matrix light source coordinate system by solving the PnP method;

[0044] S3. By comparing the real-time position calculation with the origin position, the accurate position information of the mobile platform can be calculated. The specific calculation method is the same as that of S2;

[0045] S4. Through the information transmission of the internal wireless signal module, the robot can obtain its own real-time position information, so as to perform safe and efficient welding work.

Claims

1. A device and method for positioning a mobile platform carrying a robot, comprising a space positioning instrument body in FIG1 and an indicator light board in FIG2. The space positioning instrument body comprises: High-precision visual sensor 1, Z-axis control rotation motor 2, Y-axis control rotation motor 3, industrial computer body 4, high-precision acceleration sensor 5. The method The method comprises the following steps, characterized in that: S1. Calibration of visual sensor and light board: Fix the positions of the light board and the calibration board so that the visual sensor frame contains the two complete contents. Change the angle to take multiple sets of light board and calibration board pictures. Use Zhang Zhengyou's calibration method and the PnP calculation method to obtain the visual sensor internal parameters and distortion parameters, and the light board 3D data. The body of the fixed positioning device is on the mobile platform carrying the robot. The robot has a camera sensor with "eyes in hands". The robot's posture is fixed, and the mobile platform is operated. The visual sensor of the positioning device and the camera sensor on the robot are always included in the frame of the complete fixed position calibration plate. The position of the mobile platform is adjusted, and multiple groups of pictures are taken. The posture relationship between the two groups of sensors is obtained by solving PnP. Combined with the robot TCP coordinates and the hand-eye calibration results, the rotation matrix and offset vector between the positioning device and the mobile platform are calculated. S2, pre-positioning: Fix the light board in a suitable position, park the mobile platform 1 meter in front of the light board, use the operating software to take a picture of the light board at this time, and mark it as the origin position; S3. Get real-time location: When the mobile platform moves to the undetermined work area, the visual sensor calculates the rotation matrix and offset vector of the light board and the mobile platform based on the light board image at that time, and can know the exact position of the mobile platform at that time; S4. Send data to the robot: At this time, the six-axis robot has an external axis, which will transmit the pose matrix and offset data to the robot system in real time, assisting the robot to perform related operations more flexibly.

2. The device and method for positioning a mobile platform carrying a robot according to claim 1, characterized in that: Step S3 includes the following process: Taking the robot zero position as the reference point, establish the spatial coordinates of the origin position; a. Obtain the real-time image of the light board on the visual sensor, adjust the sensor focus and ROI in real time according to the pixel size of the light board on the sensor and the proportion of the entire image on the target surface, and rotate the motor at the same time to keep the light board image in the center of the frame; b. Calculate the position relationship of the light board relative to the visual sensor chip at this time, and calculate the rotation matrix of the light board relative to the origin by combining the visual sensor rotation angle provided by the high-precision acceleration sensor; c. According to the pixel size and aspect ratio of the light board, the offset vector of the light board relative to the visual sensor is calculated by the triangulation model; d. Combining the posture relationship between the visual sensor and the camera sensor of the robot hand, the rotation matrix and offset vector of the mobile platform and the origin position can be calculated, that is, the exact position of the mobile platform at this moment.

3. The device and method for positioning a mobile platform carrying a robot according to claim 1, characterized in that: Step S4 includes the following advantages: Compared with the single direction limitation of the conventional robot's track-type external axis, the mobile platform can provide direction and angle information, thereby greatly increasing the flexibility of the "external axis". When faced with a sloped site, the installation process of conventional ground rails can be quite tricky. The positioning device of the mobile platform has a built-in two-axis rotating motor that can control the indicator light board within the visual sensor frame. The rotation angle can be captured by the acceleration sensor at the same time, thereby collaboratively providing a more accurate judgment of the mobile platform position information at the target location. The internal circuit of the mobile platform is specially designed to transmit the information of the acceleration sensor, visual sensor and rotating motor in real time and efficiently, making the entire operation process unobstructed, thus achieving real-time positioning.

Images