Design of a 2DoF Ankle Exoskeleton with a Polycentric Structure and a Bi-Directional Tendon-Driven Actuator Controlled Using a PID Neural Network

Lower limb exoskeleton robots help with walking movements through mechanical force, by identifying the wearer’s intention. When robot is lightweight and comfortable to wear, stability of increases, energy can be used efficiently. However, because it difficult implement complex anatomical human body, most are designed simply. Due this, misalignment between movement causes wearer feel uncomfortable, reduced. In this paper, we developed a two degrees freedom (2DoF) ankle subtalar joint talocrural joint, applying four-bar linkage realize simple 1DoF structure mainly for ankles. bidirectional tendon-driven actuators (BTDAs) do not consider difference in length change both cables due dorsiflexion (DF) plantar flexion (PF) during walking, causing misalignment. To solve problem, BTDA was considering cables. Cable-driven systems create uncertainty. Accordingly, adaptive control performed proportional-integral-differential neural network (PIDNN) controller minimize system