Abstract
In order to improve the measurement accuracy of laser profiler and industrial robot measurement system and reduce the influence of random error in calibration process, a hand eye calibration method based on combinatorial optimization is proposed. Firstly, the laser profiler is positioned by the robot to scan a standard sphere from different postures. Then, combinations from all the measurement data are selected to calculate several transformation matrices and the corresponding profile datasets in base coordinate system can be derived consequently. Since all the profile data come from the same stationary standard sphere, sphere fitting is applied to each dataset and the optimal hand eye transformation matrix can be derived which corresponds to the minimum fitting error. Experimental results show that the novel method can effectively improve the calibration accuracy and sphere loss which is weighted sum of sphere fitting error and radius error is reduced from 1.976 to 0.539 after combinatorial optimization.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
