Implementation of a acceleration estimator based compensation scheme to increase load data accuracy for a robotic testing system for CPR-manikins

Abstract

Laerdal Medical is a producer of Cardiopulmonary Resuscitation (CPR) training manikins, all of which undergo rigorous endurance and accuracy testing. This work proposes an acceleration estimator based compensation scheme for a industrial robot manipulator product testing system with the intention of increasing load data accuracy for the purpose of product review and calibration. As part of the compensation scheme four different acceleration estimators are implemented and compared. Results indicate that the compensation scheme increases the load data accuracy by 1.5 - 6 % of the reference value depending on compression depth and spring rate. However the accuracy goal of 0.4 [kg] is not reached. The work has also uncovered the presence of position error in the robot. Thus, further improvement to the compensation scheme and positional error compensation is required.