Inverse Modeling for Component Selection of Twisted String Actuators

Abstract

Twisted string actuators (TSAs) have been widely used in numerous mechatronic applications. However, existing literature selects the TSA's components (strings and motors) through ad hoc procedures. The selection of TSA components requires inversion of its mathematical models. This is difficult because many variables are coupled in such a way that prevents direct inversion. This article presents a novel systematic design framework and inversion algorithm for TSAs based on their required performance properties. First, a comprehensive formula to invert the kinetostatic and dynamic models of the TSA is presented. This inversion finds the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TSA-specific properties</i> given the desired performance properties that are <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">common to most actuators</i> . The solution is iteratively solved from a synthesis of existing models that account for string compliance, variable TSA radius, friction, the payload's inertia, and its linear velocity. Second, an algorithm recommends the most suitable motor and string pair from a custom-made database. Third, simulations and experiments evaluated the algorithm's performance under different conditions.