Design and Control of Soft Exoskeleton Based on Human Motion Intention

Abstract

Exoskeletons have been utilized to enhance human movement capabilities; however, most of exoskeletons are typically limited by the small-range stiffness of actuators. Furthermore, these exoskeletons are not comfortable to wear due to the rigid structures. This letter proposes a soft exoskeleton with twisted string actuators to assist knee joint movement. The mechanical structure and dynamics model of the exoskeleton robotic system is introduced, where the variable-stiffness actuator with twisted strings is able to provide controllable walking assistance and improve users' comfort. In addition, the users motion intention is detected through biomedical signal, and an assist-as-need control strategy is proposed to enhance assistance efficiency. Finally, the effectiveness of the proposed exoskeleton is verified empirically.