Kinematic Accuracy Improvement of a Novel Smart Structure-Based Parallel Kinematic Machine

The complex structure of the parallel kinematic machine (PKM) challenges its high kinematic accuracy. An innovative smart structure-based PKM design method was generated in our previous work as a possible solution to improve its accuracy, wherein a prototype was successfully developed. This paper mainly examined the theoretical and technique methods to improve the kinematic accuracy of the novel smart structure-based PKM to promote its practical application. The developed smart structure-based PKM was first introduced and the procedure of improving its kinematic accuracy was presented. The kinematic error model of this PKM was then established by considering the kinematic property of the machine. Due to the special structure of the smart structure-based PKM, several key preparations, including the installation of the grid encoder, the origin returning of the smart structure chains, and the automatic measurement of the motion information, were performed for kinematic accuracy improvement. On this basis, the PKM's kinematic accuracy was improved by the kinematic calibration method and the smart structure chains. The regularization method was employed to deal with the ill-conditioning problem in the error identification of the PKM, thereby reducing the maximum positioning error of the smart structure-based PKM from 300 to 25 μm through error compensation. An experimental test was performed to verify the existence of nonlinear geometric errors in the actual PKM. A regional error identification and compensation method was proposed to reduce their effects on the result of the kinematic calibration. Finally, the smart structure chains were controlled to further improve the PKM's kinematic accuracy. The experimental results indicated that the smart structure-based PKM achieved micron-level positioning accuracy in its whole workspace by following the proposed kinematic accuracy improvement process.