| At present,in order to pursue a smarter machine that can perform complex tasks,it is a good solution to give vision to the robot.In order to accomplish the task in such a way,it is necessary to first obtain the relationship between the robot and vision system.This transformation relationship is used to control the motion of robot by the image.From the perspective of cost and practicality,consider the way in which two cameras are combined with a robot.Fixed camera has a large search range,it is called fixed camera hand-eye system.One has a higher precision for the camera on hand,and called cameraon-hand hand-eye system.In order to avoid complicated operations and reduce human interference,rapid deployment can be calibrated after installation without additional human intervention.In view of this structure,this paper studies it and implements an algorithm that can quickly calibration.The main research contents of this paper are as follows:1.The fixed camera has a large field of view and can initially position the target within a large range.It remains stationary with the work plane.Therefore,what is acquired for a fixed camera is the transformation relationship between the two planes.This article uses the end of robot as a marker point.This paper designs a special motion based on the rotation around the end to obtain its coordinates in two planes.Since the coordinates of the end of the robot are known on the robot controller,the image coordinates are calculated by acquiring a plurality of images of the calibration plate that is rotated about the end axis.The transformation relationship can be solved after collecting data at multiple collection points.This method is simple and easy,and it has achieved the expected results in actual experiments.2.The resolution of the camera on the hand is higher,and the target can be detected more finely.It can acquire an image of the position of the end of the robot current time.This paper designs a calibration data acquisition method that first moves on the plane and then moves in the vertical direction to calibrate.The result of the calibration is the transformation relationship between the camera and the end of the robot.In actual use,combine the current robot pose to calculate its actual position in the work plane.This method is very fast because it requires only a few small movements.3.After completing the calibration of the fixed camera hand-eye system and the eyeon-hand hand-eye system,tested the whole system.The purpose of the experiment is to determine the parameters that best satisfy the experimental scenario of this paper within the tolerances of the error,and to compare the time consumption.Through experiments,it is found that several aspects have a greater impact on time consumption.One is the number of sampling points and the step size.The more points,the longer it takes.The second is the step size when collecting data.The longer step,the longer it takes. |
