Abstract
Wearable assistive devices' (WADs) development is impeded by traditional actuators' and control paradigms' lack of compliance and adaptability; as such, the central nervous system's (CNS) actuation and control principles have been investigated in order to overcome these limitations in a novel way. A bio-mimetic model of a limb joint, which is antagonistically actuated by two Hill-type muscles, is presented. Limb joint stability and transient response, as functions of co-activation, have been investigated. Three simulations have been carried out: equal (Af(f,f)), unequal (Af(3,f)), and differential (Af(3,Ramp(g))) co-activation. For normalized stimulus frequency range: 1 ≤ f ≤ 3, equal and unequal co-activation leads to increased limb joint damping and reduced transient oscillation. For the equally/unequally co-activated case, the linearized model's dominant complex-conjugate poles become increasingly negative and tend towards the real axis, which indicate that increased co-activation leads to increased joint stability. With respect to differential co-activation, increasing the limb joint's antagonistic muscle's normalized stimulus frequency's (f) rate of change (g) leads to increased angular velocity (ω); however, at a cost of increased overshoot. Differential activation of the antagonistic muscle results in positive angular rotation (θ). Copyright © 2011 by ASME.