Optimization-Based Design of a Lightweight Rehabilitation Exoskeleton of an Upper Extremity for Task-Oriented Treatment

Contemporary physiotherapy requires technological tools to provide effective therapy to the increasing group of patients, neurological, among others. This can be achieved with rehabilitation robots, which can also be exoskeletons - wearable devices mobilizing multiple joints with complex motions representing activities of daily living. To perform the kinesiotherapy conveniently in home-like environments, the exoskeletons need to be relatively lightweight. The paper presents the methodology of decreasing the mass of the exoskeleton design with the human-in-the-loop simulations of motions, followed by multibody dynamics simulations, and finite element method (FEM) multistep optimization. The process includes sequential initial parametric optimization, topology optimization, and final parametric optimization. The steps are used to set initial dimensional and material parameters, extract new geometrical features, and adjust the final geometry dimensions of a new design. The presented case of the SmartEx-Home exoskeleton resulted in a total mass reduction of almost 50% while meeting the criteria of the minimum safety factor and maximum internal stress and strain for all the components. The final design is being manufactured and will be used within the tests with humans, reflecting almost fully automatic passive and active therapy.