Background
In welding applications such as energy, ships, rail transit, engineering construction and the like, there are a large number of field welding tasks of workpieces with complex shapes in small batches. The movable welding robot combines a welding system, a mechanical arm and a movable chassis, has the advantages of large movement range and high flexibility degree, and meets the field welding requirement of complex scenes. In the practical application process, as the mobile welding robot faces the problem of unstructured environment of the operation site, the high-efficiency and intelligent welding requirements are difficult to meet in a teaching programming mode. The method of off-line programming is adopted, and the motion trail and the execution file of the welding robot are automatically generated through a software program.
In order to realize the on-site operation offline programming of the mobile welding robot, the welding groove position needs to be determined under a robot tool coordinate system, namely, a conversion matrix from a workpiece coordinate system to the robot tool coordinate system is searched. At present, two calibration strategies are commonly used, namely an 'eye on hand' method, namely, a camera and a mechanical arm are rigidly fixed, and groove position information acquired by the camera is converted into a robot tool coordinate system through a conversion matrix of a pre-calibrated camera coordinate system and a robot tool coordinate system. In practical application, a person is required to control the position and the direction of the mechanical arm, so that the target groove is positioned in the visual field of the camera, and the process is complex in operation, time-consuming and labor-consuming.
Another calibration strategy is the "eye-out-of-hand" method, i.e. fixing a sensor beside the welding table, predetermining the relative positions of the camera coordinate system and the table and robot, and then converting the groove position to the robot tool coordinate system. Because of the uncertainty of the field position of the mobile robot, the method has long field preparation time, low calibration precision and higher requirements on the skill level of operators.
The Chinese patent application with the application number 201811653845.5 discloses a laser camera hand-eye calibration method applied to arc material increase, and a hand-eye relation is determined by using a laser camera. Because the method of 'eyes on hands' is adopted, the robot still needs to be manually controlled to carry out complex movement, and the operation carelessly collides with a workpiece, so that the damage is caused. And the calibration process needs to use a calibration plate, and the precision of the calibration plate directly influences the calibration precision.
The Chinese patent application with the application number 202111051769.2 discloses a welding robot positioning method based on binocular vision and line laser sensing data fusion. The method adopts the mode that eyes are outside hands, and the binocular vision sensor is fixed beside the tool table. In the field application process, as the appearance of the workpiece can shade the visual field of the visual sensor, if only a welding line and a corner position groove on one side can be calibrated, the flexibility is poor in the field application process, and the workpiece can not be applied to complex welding line scenes.
The Chinese patent application with the application number 202210417510.3 discloses an automatic welding hand-eye calibration method of a laser vision guided robot, which uses a three-dimensional calibration plate with three calibration points to assist in completing the hand-eye calibration process, namely determining the relation between a camera coordinate system and a robot tool coordinate system.
In summary, the conventional method requires a calibration board as an intermediary to calculate the hand-eye matrix, i.e. the transformation matrix from the camera coordinate system to the robot tool coordinate system. And manual teaching is often carried out, the process is complex in operation, time-consuming and labor-consuming, high in requirements on skill and knowledge level of operators, and time cost and labor cost are added to field operation of the mobile welding robot.
Disclosure of Invention
The application provides a rapid calibration system and a rapid calibration method for field operation of a mobile welding robot, and aims to shorten the calibration time consumption of the field operation of the mobile robot, improve the calibration precision and reduce the skill requirement on operators.
The technical aim of the application is realized by the following technical scheme:
The quick calibration system for the field operation of the mobile welding robot comprises a mobile welding robot subsystem and a calibration subsystem, wherein the mobile welding robot subsystem comprises a mobile chassis, a multi-axis mechanical arm, a special-shape welding gun, a working platform and a workpiece to be welded;
One end of the multi-axis mechanical arm is arranged on the movable chassis, and the other end of the multi-axis mechanical arm is connected with the welding gun with the special shape;
the handheld three-dimensional scanner and the display are connected with the computer, and the computer controls the welding site three-dimensional point cloud model and the welding robot tool coordinate system to be displayed on the display.
A rapid calibration method for field operations of a mobile welding robot, comprising:
s1, an operator operates a mobile welding robot subsystem and adjusts the distance between a welding gun with a special shape and the starting point of a groove of a workpiece to be welded;
s2, an operator acquires a three-dimensional point cloud model comprising a welding gun with a special shape and a workpiece to be welded through a handheld three-dimensional scanner;
s3, processing the three-dimensional point cloud model, identifying a tool coordinate system, and constructing a target coordinate system corresponding to the groove position;
S4, obtaining a conversion matrix through the tool coordinate system and the target coordinate system, and obtaining the relative position of the groove under the tool coordinate system through the conversion matrix, wherein the relative position is the calibration position.
The rapid calibration system and the rapid calibration method for the field operation of the mobile welding robot have the advantages that operators do not need to have skilled robot kinematics knowledge and robot control skill proficiency, and under the condition of using a handheld three-dimensional scanner, the pose relation between the mobile welding robot tool coordinate system and the welding groove is obtained, so that the operation is simple. Meanwhile, a high-precision calibration plate or target surface is not required to be prepared, the relation between a camera coordinate system and a robot tool coordinate system is not required to be calibrated, and the steps are few. In addition, the external marker of the welding gun is identified through the algorithm, so that the method is simple, quick and accurate.
Detailed Description
The technical scheme of the application will be described in detail with reference to the accompanying drawings.
The application discloses a rapid calibration system for field operation of a mobile welding robot, which is shown in fig. 1 and comprises a mobile welding robot subsystem and a calibration subsystem, wherein the mobile welding robot subsystem comprises a mobile chassis 1, a multi-axis mechanical arm 2, a special-shape welding gun 3, a working platform 4 and a workpiece to be welded 5, one end of the multi-axis mechanical arm 2 is arranged on the mobile chassis 1, the other end of the multi-axis mechanical arm is connected with the special-shape welding gun 3, and the workpiece to be welded 5 is arranged on the working platform 4.
The calibration subsystem comprises a handheld three-dimensional scanner 6, a computer 8, a display 9, a welding site three-dimensional point cloud model 10 and a welding robot tool coordinate system 11, wherein the handheld three-dimensional scanner 6 and the display 9 are connected with the computer 8, and the computer 8 controls the welding site three-dimensional point cloud model 10 and the welding robot tool coordinate system 11 to be displayed on the display 9.
Specifically, the distance between the special-shape welding gun 3 and the starting point of the groove of the workpiece 5 to be welded is 50mm-300mm.
Specifically, as shown in fig. 3, the welding gun 3 with special shape is a cylinder, and the front end of the welding gun is provided with a triangular pyramid-shaped marker, and the tip of the triangular pyramid-shaped marker points to the X t axis of the tool coordinate system.
As shown in fig. 2, the method for quickly calibrating the field operation of the mobile welding robot according to the application comprises the following steps:
s1, an operator operates the movable welding robot subsystem, and adjusts the distance between a welding gun with a special shape and the starting point of a groove of a workpiece to be welded.
Specifically, step S1 includes:
s11, the operator manipulates the mobile chassis 1 to move the mobile chassis 1 to the vicinity of the work platform 4.
And S12, an operator operates the multi-axis mechanical arm 2 to move the welding gun 3 with the special shape to the starting point of the groove of the workpiece 5 to be welded.
And S13, adjusting the distance between the welding gun 3 with the special shape and the starting point of the groove. Specifically, the position of the multi-axis mechanical arm 2 is roughly adjusted to enable the special-shape welding gun 3 to be near the starting point of the groove of the workpiece 5 to be welded, and the distance is generally controlled within the range of 50mm-300 mm.
S2, an operator acquires a three-dimensional point cloud model comprising the welding gun 3 with the special shape and the workpiece 5 to be welded through the handheld three-dimensional scanner 6.
Specifically, an operator is required to perform a scanning operation in the working range of the three-dimensional scanner, and the working distance is 0.4m to 1.0m for example of Artec Eva three-dimensional scanner. An operator performs surrounding scanning on the special-shape welding gun 3 and the workpiece 5 to be welded through the handheld three-dimensional scanner 6, and a three-dimensional point cloud model comprising the special-shape welding gun 3 and the workpiece 5 to be welded is obtained.
And S3, processing the three-dimensional point cloud model, identifying a tool coordinate system, and constructing a target coordinate system corresponding to the groove position.
Specifically, as shown in fig. 4, step S3 includes:
and S31, the software program of the computer 8 reads the position coordinate data of each point in the three-dimensional point cloud model, and then establishes a world coordinate system.
S32, the algorithm of the computer 8 identifies the outline circle of the front end part of the welding gun 3 with the special shape, and the circle center O t(xo,yo,zo is obtained through the circle fitting algorithm).
And S33, identifying the triangular pyramid-shaped marker at the front end of the special-shape welding gun 3, and fitting the outer boundary of the triangular pyramid-shaped marker on the same end face with the special-shape welding gun 3 to obtain a boundary straight line intersection point C t(xc,yc,zc of the triangular pyramid-shaped marker.
S34, connecting O t and C t, determining O tCt as an X t axis of a tool coordinate system, and expressing the direction vector of the X t axis as follows:
The calculation is carried out on a cylindrical central axis Z t of the main body of the welding gun 3 with the special shape, and the direction vector of the Z t axis is expressed as follows: Wherein, Wherein, the point D (x d,yd,zd) and the point E (x e,ye,ze) are both positioned on the outline circle of the front end part of the special-shape welding gun 3;
the Y t axis being determined according to the right-hand rule, i.e Then the tool coordinate system { T } = { X t,Yt,Zt }, is obtained.
S35, constructing a target coordinate system { G }, { G } = { X g,Yg,Zg } by taking a groove starting point as an origin N g, wherein the Z g axis is the direction vertical to the working platform 4, and the direction vector isThe X g axis is the direction of the starting point of the groove to the end point of the groove;
S4, obtaining a conversion matrix through the tool coordinate system and the target coordinate system, and obtaining the relative position of the groove under the tool coordinate system through the conversion matrix, wherein the relative position is the calibration position.
Specifically, step S4 includes:
S41, the origin of the tool coordinate system { T } is represented as a point O t(xo,yo,zo, the origin of the target coordinate system { G } is represented as N g(xn,yn,zn), then the translation vector of the tool coordinate system { T } to the target coordinate system { G } Is (x n-xo,yn-yo,zn-zo).
S42, throughThe unit vector representing the main axis direction of the target coordinate system { G } is calculated by using the tool coordinate system { T } as the reference coordinate system to obtain three unit vectorsAnd
S43 according toIs arranged in 3*3 matrix, then the rotation is converted into matrixCalculations were performed, expressed as:
Wherein scalar r ij represents the component of each vector projected in the direction of the central axis of its reference frame;
s44, according to the translation vector Rotating a transformation matrixObtaining the pose relation of the starting point of the groove corresponding to the tool coordinate system { T }, and obtaining a conversion matrix to be expressed as:
s45, calibrating the relative position of the groove through the conversion matrix.
FIG. 5 is a schematic view of the effect of the method of the present application after rapid calibration. In conclusion, the rapid calibration method and the rapid calibration system do not need to prepare a high-precision calibration plate or target surface, do not need to calibrate the relation between a camera coordinate system and a robot tool coordinate system, and have few steps, so that external markers of the welding gun can be identified simply, rapidly and accurately.
The foregoing is an exemplary embodiment of the application, the scope of which is defined by the claims and their equivalents.