Research On Hand-Eye Calibration Technology Of Industrial Robot

Giving the visual function to robot and establishing coordinate relations between robot hand and eye is an important milestone, which can make robot reach the intelligent. Robot hand-eye relation is a bridge which can establish a relation between robot and visual sensor. After robot is installed by sensor, robot may perform some specific tasks with visual access to environmental information. This paper mainly research on some key technologies when robot performing tasks-hand-eye calibration technology. Details are as follows:1. Use an improved optimization approach which can optimize intrinsic and extrinsic parameters of camera based on planar target calibration method of Zhengyou zhang with the two-dimensional calibration template used in this paper. Calculate the intrinsic and extrinsic parameters of camera. Analyse and contrast on the pre-reprejection error figure and the reprejection error figure using this optimization approach. We can make a conclusion that this improved optimization algorithm can improve the accuracy of camera parameters greatly by comparing the different figures. Comparing with other optimization algorithms, the step of this improved optimization algorithm is much more specific, and this algorithm is much more specific. At the same time, this improved optimization algorithm has much higher operability in practical engineering environment.2. Summarize the existence and uniqueness conditions for hand-eye equations and specific solution process. Analyze to improve and enhance the hand-eye relationship between methods and measures in the solution process. At the same time, gives an improved hand-eye calibration algorithm at the same time. This algorithm use matrix direct-product and eigenvector research and solve the linear equation of the robot hand-eye transformation matrix., which solve the complexity that is produced by the hand-eye calibration matrix effectively, and simplify the calculation with sufficient accuracy. The concrete calibration steps are given in this paper. 3. Complete the hand-eye calibration experiment. Use the traditional hand-eye calibration algorithm to solve the hand-eye matrix. Then, use the object-oriented programming techniques to solve the hand-eye matrix with the method of improved hand-eye calibration algorithm in the environment of c++ of vc6.0 version. Through calculate the error between the two methods, realize that the error introduced by the orthogonalization is permissibly in the vast majority cases. This improved hand-eye calibration algorithm solve the complexity that is produced by the hand-eye calibration matrix effectively, and simplify the calculation with sufficient accuracy.