A Two-Step Kinematic Calibration Method Based on Biological Characteristics for Exoskeleton

Abstract

Exoskeleton is presented for applications including, but not limited to, human motion capture. In this article, we present a two-step kinematic calibration method for exoskeleton, which reduces the computational complexity and improves the motion capture precision. Our planning method calibrates the position errors and the orientation errors in different steps according to biological characteristics. Then, the least square method is used to calculate the calibration results. At the first step, we calibrate the orientation errors only based on the structure of the geometric errors calibration matrix. And at the second step, the calibration results of the first step will be used to update the kinematic parameters and then the position errors are calibrated. Due to the proposed two-step kinematic calibration method, the kinematic parameters could be better calibrated in the embedded system. The performance of the proposed two-step kinematic calibration method, including the computational complexity and the motion capture precision, is compared with the traditional one-step kinematic calibration method. The simulations demonstrate that the motion capture effect is significantly improved by 8.7% comparing with the traditional one-step kinematic calibration method.