Optimal stiffness tuning for a lower body exoskeleton with spring-supported passive joints

Abstract

This paper presents a framework to optimally tune the stiffness values of spring-supported passive joints that are included in lower body exoskeletons. First, a dynamic model of a combined human-exoskeleton system was created in MSC.ADAMS software. Second, a gradient-descent based algorithm was used to find the optimum value to minimize the ZMP for a range of ankle stiffness values. In order to corroborate the proposed method, simulation experiments were conducted by considering three cases in which different body mass and heights were assigned to the combined human-exoskeleton system. The simulation results indicate that the proposed methodology is effective in order to find the optimum ankle stiffness for the combined human-exoskeleton systems, resulting in reductions in ZMP variations and therefore increasing the balancing ability. As a consequence, it may be possible to reduce the number of active joints in exoskeletons that aim crutch-free 3-D walking motion support.