A task-based design methodology for robotic exoskeletons

Abstract

INTRODUCTION This study is aimed at developing a task-based methodology for the design of robotic exoskeletons. This is in contrast to prevailing research efforts, which attempt to mimic the human limb, where each human joint is given an exoskeleton counter-joint. Rather, we present an alternative systematic design approach for the design of exoskeletons that can follow the complex three-dimensional motions of the human body independent of anatomical measures and landmarks. With this approach, it is not necessary to know the geometry of the targeted limb but rather to have a description of its motion at the point of attachment. METHODS The desired trajectory of the targeted limb has been collected through a motion capture system from a healthy subject. Then, an approximate dimensional synthesis has been employed to specify the size of the mechanism and its location with respect to the limb, while generating the desired trajectory. The procedure for this method, from motion capture to kinematic synthesis to mechanism selection and optimization, is validated with an illustrative example. RESULTS The proposed method resulted an exoskeleton which follows the desired trajectory of the human limb without any need of aligning its joint to the corresponding human joints. CONCLUSION A method to design lower mobility exoskeletons for specific sets of human motion is presented; the approach result an exoskeleton with lesser actuation system while generating complex 3D limb motions, which in turn results a lighter exoskeletons. It also avoids a need to align each robotic joint axis with its human counterpart.