Theoretical analysis and numerical optimization of a wearable spring-clutch mechanism for reducing metabolic energy cost during human walking

There is a growing interest in assistive wearable devices for laden walking, with applications to civil hiking or military soldiers carrying heavy loads in outdoor rough terrains. While the solution of powered exoskeleton is known to be heavy and energy consuming, recent works presented wearable light-weight (semi-)passive elements based on elastic springs engaged by timed clutches. In this work, we theoretically study the dynamics of a five-link model of a human walker with point feet, using numerical simulations. We propose a novel mechanism of a spring and two triggered clutches, which enables locking the spring with stored energy while the device's length can change freely. For a given gait of joint angles trajectories, we numerically optimize the spring parameters and clutch timing for minimizing the metabolic energy cost. We show that a cleverly designed device can, in theory, lead to a drastic reduction in metabolic energy expenditure.