Limitations for walking speeds – how accurate are predictions based on a reductionist inverted pendulum model


Meeting Abstract

82.4  Friday, Jan. 7  Limitations for walking speeds – how accurate are predictions based on a reductionist inverted pendulum model HUBEL, TY*; USHERWOOD, J; Royal Veterinary College, UK; Royal Veterinary College, UK thubel@rvc.ac.uk

Reductionist inverted pendulum models of walking predict that maximum walking speed depends on step length, leg length, and gravity. Maximum walking speed is limited by the ability of gravity to provide the centripetal accelerations required for a body vaulting around an arc, and prevent foot take off. Most critical conditions for foot take off occur at extreme leg angles (angle between the leg and ground vertices), at the onset and end of stance, due to the combination of maximum centripetal acceleration and a reduced gravity component along an angled leg. Limitations to leg angles based on conditions at early and late stance lead to a maximum step length related to speed. When combined with limitations in leg swing frequency, predicted walking speeds based on the “Compass gait model” (inverted pendulum assumptions) are close to those observed in humans (Fr=0.5) and ducks (Fr=0.25). The goal of the current study is to investigate the limiting effect of leg swing frequency, gravity and substrate incline on maximum walking speed, and to assess the efficacy of an expanded compass gait model to predict these limits. For slope walking we assume that the angle of the slope decreases the gravity component counteracting the acceleration forces at the extremes of stance and expect to see a corresponding reduction in maximum step length. We predict that the reduction in step length with no change in step frequency will lead to decreasing maximum walking speed with increasing slope angles. We use GPS and accelerometry data from eight individuals walking on level ground and different slope angles at a variety of speeds (on treadmill and in field experiments) in order to test the accuracy of the model.

the Society for
Integrative &
Comparative
Biology