Vision-based kinematic calibration of a small-scale spherical parallel kinematic machine

The paper deals with the kinematic calibration of a mini pointing device with a two degrees of freedom spherical motion. Given an error kinematic model, based on a first order approximation approach, a global calibration method that includes the hand-eye pose estimation is proposed. The calibration method is based on an iterative procedure of refinement, whose convergence is proved. Data for calibration are obtained from a vision system in eye-to-hand configuration. A series of tests were carried out in order to find the best experimental setup and to assess the accuracy of the measurement system. Finally, the results of the calibration procedure are discussed.