CN-116394254-B - Zero calibration method and device for robot and computer storage medium

CN116394254BCN 116394254 BCN116394254 BCN 116394254BCN-116394254-B

Abstract

The application discloses a zero point calibration method, a zero point calibration device and a computer storage medium of a robot, wherein the method comprises the steps of respectively moving the robot from each first teaching point position to a standard point position, and recording the tail end coordinates of the robot corresponding to the movement from the different first teaching point positions to the standard point positions; the method comprises the steps of utilizing the tail end coordinates of a robot to solve the offset estimation value of a tool coordinate system, moving the robot from each second teaching point position to a standard point position respectively, recording the positions of the joints of the robot corresponding to the standard point positions from different second teaching point positions, constructing a nonlinear optimization model by utilizing the positions of the joints of the robot, and utilizing the offset estimation value as an initial value to solve the nonlinear optimization model to obtain zero parameters. The zero point calibration device adopts 2D vision, adopts the mode of aligning the center of the main point of the camera and the mark point to perform zero point calibration on the robot, does not need a high-precision calibration plate, and can reduce the calibration cost while ensuring the high-precision calibration effect.

Inventors

GUO XIAOBIN
CUI YONGQIANG
YAO CHAOZHI
HUANG JUNBIAO
LUO XIN
LUO QI
HUANG GUOHUI

Assignees

深圳众为兴技术股份有限公司

Dates

Publication Date: 20260512
Application Date: 20230423

Claims (10)

1. The zero calibration method of the robot is characterized by comprising the following steps of: moving the robot to a plurality of first teaching points, respectively moving the robot from each first teaching point to a standard point, and recording the tail end coordinates of the robot corresponding to the standard point, which are the points of the robot when the center of the standard point is aligned with the main point of the camera; Solving an offset estimation value of a tool coordinate system by utilizing the tail end coordinates of the robot; moving the robot to a plurality of second teaching points, respectively moving the robot from each second teaching point to the standard point, and recording the positions of the joints of the robot corresponding to the positions from different second teaching points to the standard points; Based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged, constructing a nonlinear optimization model by utilizing the robot joint position; and solving the nonlinear optimization model by using the offset estimation value as an initial value to obtain a zero point parameter, wherein the zero point parameter comprises a zero point offset value and a tool offset value of the robot.
2. The zero calibration method according to claim 1, wherein, The zero point offset value of the robot includes a zero point offset value between joints of the robot.
3. The zero calibration method according to claim 2, wherein, After the zero point parameter is acquired, the zero point calibration method further comprises the following steps: acquiring a first homogeneous transformation matrix from a tool coordinate system to a robot tail end coordinate system and a second homogeneous transformation matrix among joints of the robot; Constructing a third homogeneous transformation matrix of the mark point center under a robot base coordinate system based on the first homogeneous transformation matrix and the second homogeneous transformation matrix; solving the third homogeneous transformation matrix by utilizing the zero parameter to obtain a zero coordinate of the mark point center under the robot base coordinate system; And calculating the error of the zero parameter according to the zero coordinate.
4. The zero calibration method according to claim 1, wherein, The constructing a nonlinear optimization model by utilizing the robot joint position based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged comprises the following steps: acquiring the original gesture of each teaching point position; Acquiring an attitude included angle between every two teaching points based on the original attitude of each teaching point; Performing gesture interpolation on the teaching points by using the gesture included angles to obtain interpolation gestures between the teaching points; And constructing the nonlinear optimization model by utilizing the robot joint position, the original gesture and the interpolation gesture based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged.
5. The zero calibration method according to claim 4, wherein, The original gesture based on each teaching point position obtains a gesture included angle between every two teaching point positions, and the method comprises the following steps: Converting the attitude angle of the original attitude of each teaching point position into a unit quaternion; and acquiring the attitude included angle of the short attitude path based on the unit quaternion of the teaching point positions.
6. The zero calibration method according to claim 1, wherein, The solving the offset estimation value of the tool coordinate system based on the robot end coordinate comprises the following steps: Constructing a homogeneous transformation matrix from the robot end coordinate system to the robot base coordinate system based on the robot end coordinate; Constructing a homogeneous transformation matrix from the tool coordinate system to the tail end coordinate system of the robot; And calculating an offset estimation value of the tool coordinate system by using the homogeneous transformation matrix from the robot end coordinate system to the robot base coordinate system and the homogeneous transformation matrix from the tool coordinate system to the robot end coordinate system.
7. The zero calibration method according to claim 1, wherein, The recording of the robot terminal coordinates corresponding to the standard point position and moving from the different first teaching point positions comprises the following steps: acquiring main point coordinates of a camera and mark point pixel coordinates corresponding to each first teaching point position; acquiring a required offset of the tail end of the robot based on the camera main point coordinates and the mark point pixel coordinates; And acquiring the robot tail end coordinates of the robot tail end at the standard point based on the required offset and the original robot tail end coordinates of the robot tail end at the first teaching point position.
8. The zero calibration method according to claim 7, wherein, The obtaining the required offset of the robot terminal based on the camera principal point coordinates and the marker point pixel coordinates includes: Acquiring a deviation value of the main point coordinates of the camera and the pixel coordinates of the mark points; acquiring a required offset of the robot tail end based on the pixel equivalent of the camera and the deviation value; the zero calibration method further comprises the following steps before the pixel equivalent and the deviation value based on the camera acquire the required offset of the tail end of the robot: photographing at any teaching point position to obtain a first teaching pixel coordinate; Moving the tail end of the robot according to a preset step length and a preset direction, and photographing to obtain a second teaching pixel coordinate; Acquiring a rotation angle of the robot base coordinate system and a camera coordinate system based on the first teaching pixel coordinate and the second teaching pixel coordinate; constructing a free vector transformation equation by utilizing the first teaching pixel coordinates, the second teaching pixel coordinates and the rotation angle; and obtaining the pixel equivalent of the camera by utilizing the joint calculation of a plurality of groups of free vector transformation equations.
9. A zero calibration device of a robot, wherein the zero calibration device comprises a memory and a processor coupled with the memory; wherein the memory is configured to store program data and the processor is configured to execute the program data to implement the zero calibration method according to any one of claims 1 to 8.
10. A computer storage medium for storing program data which, when executed by a computer, is adapted to carry out the zero calibration method according to any one of claims 1 to 8.

Description

Zero calibration method and device for robot and computer storage medium Technical Field The application relates to the technical field of robot kinematics parameter calibration, in particular to a zero calibration device of a robot, a zero calibration method thereof and a computer storage medium. Background The absolute positioning accuracy of the robot is a very important technical index for measuring the performance of the robot. The requirements of the industrial field on the absolute positioning accuracy of robots (such as component plug-in, soldering tin and screw driving) are higher and higher. The absolute positioning accuracy of the robot is affected by a plurality of factors, such as part machining accuracy, assembly errors, joint friction and wear, zero point offset or loss, and the like, so that the actual kinematic parameters of the robot deviate from the theoretical values of the controller, and the kinematic parameters of the robot need to be calibrated to improve the absolute positioning accuracy. Among them, zero point offset or loss is the most important factor affecting the absolute positioning accuracy of the robot. The existing robot zero point calibration method has the problems that calibration equipment is expensive, the calibration process is complex, the calibration cost is high, manual operation is needed, and human errors are introduced due to the fact that naked eyes are needed to observe. Disclosure of Invention The application provides a zero calibration method, a zero calibration device and a computer storage medium for a robot. The application adopts a technical scheme that a zero calibration method of a robot is provided, and the zero calibration method comprises the following steps: moving the robot to a plurality of first teaching points, respectively moving the robot from each first teaching point to a standard point, and recording the tail end coordinates of the robot corresponding to the standard point, which are the points of the robot when the center of the standard point is aligned with the main point of the camera; Solving an offset estimation value of a tool coordinate system by utilizing the tail end coordinates of the robot; moving the robot to a plurality of second teaching points, respectively moving the robot from each second teaching point to the standard point, and recording the positions of the joints of the robot corresponding to the positions from different second teaching points to the standard points; Based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged, constructing a nonlinear optimization model by utilizing the robot joint position; and solving the nonlinear optimization model by using the offset estimation value as an initial value to obtain a zero point parameter, wherein the zero point parameter comprises a zero point offset value and a tool offset value of the robot. Wherein the zero point offset value of the robot includes zero point offset values between joints of the robot. After the zero point parameter is acquired, the zero point calibration method further comprises the following steps: acquiring a first homogeneous transformation matrix from a tool coordinate system to a robot tail end coordinate system and a second homogeneous transformation matrix among joints of the robot; Constructing a third homogeneous transformation matrix of the mark point center under a robot base coordinate system based on the first homogeneous transformation matrix and the second homogeneous transformation matrix; solving the third homogeneous transformation matrix by utilizing the zero parameter to obtain a zero coordinate of the mark point center under the robot base coordinate system; And calculating the error of the zero parameter according to the zero coordinate. The constructing a nonlinear optimization model by using the robot joint position based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged comprises the following steps: acquiring the original gesture of each teaching point position; Acquiring an attitude included angle between every two teaching points based on the original attitude of each teaching point; Performing gesture interpolation on the teaching points by using the gesture included angles to obtain interpolation gestures between the teaching points; And constructing the nonlinear optimization model by utilizing the robot joint position, the original gesture and the interpolation gesture based on the relationship that the coordinates of the mark point center under a robot base coordinate system are unchanged. The method for acquiring the attitude included angle between every two teaching points based on the original attitude of each teaching point comprises the following steps: Converting the attitude angle of the original attitude of each teaching point position into a unit quaternion; and acquiring the a