Speed-dependent modulation of swing leg rest length during human walking

Understanding swing leg behavior is essential for modeling human gait and developing control strategies for assistive devices. Many previous studies have captured walking dynamics via a spring-based model. While previous SLIP-based walking models have investigated stiffness modulation with gait speed, other critical parameters, particularly swing leg rest length, have been largely overlooked. In this study, we present a modified swing leg model that incorporates both a variable rest length and a torsional spring at the hip, building on a previously proposed single-mass linear spring model. Simulation results showed that the proposed model effectively replicated the overall trends of swing leg dynamics with significantly improved accuracy compared to previous models. Furthermore, a strong correlation (r > 0.8) was observed between gait speed and the modulation of rest length, similar to the correlation observed for leg stiffness. These biomechanical findings highlight that all spring parameters related to swing leg motion are modulated with gait speed, supporting the use of spring-based modeling in motion planning for bipedal robots and exoskeletons.