We have been intending to propose a model
connecting body dynamics and sensor feedback to investigate the gait
adaptation mechanisms of decerebrate (thalamic) cats in split-belt
waking[Yanagihara et al. Neuro. Res. 1993].
In our previous study[LivingMachines2020],
the result of early adaptation was not
sufficient.
Recently, we aimed to clarify the mechanisms of
early adaptation, and newly employed the
swing phase adjustment into our old spinal cat model by
referring to the spinal cord half-center CPG model proposed by [Frigon
et al. J. Phys. 2017]. Utilizing this new spinal cat model, we
conducted experiments using Kotetsu of split-belt walking with hindlimbs
while changing the speed ratio (fast belt speed / slow belt speed) from
1.25 to 2.5, which were the similar condition to the one in
experiments of spinal cats by [Frigon et
al. 2017]. As a result, we obtained the pattern of changing
durations of the swing and the stance phases very
similar to the one in spinal cats experiments. In addition,
such pattern of changing durations are quite
similar to the one of early adaptation of thalamic cats[Yanagihara
et al. 1993]. For explanations of the result, please see
the figure.
In this paper, while referring to the analysis of the leg phase sifting in split-belt locomotion using the center of the stance phase (COST) and the center of the single supporting phase (COSS) by [Fujiki, Aoi et al. J. Royal Soc. Interface, 2015], we especially paid attention to the normalized difference between times of COSS and COST (DBCC) and the mechanism of DBCC genration. In order to evaluate the new spinal cat model, we considered the relation between bisupport phases (B1, B2) and DBCC, and the problem of DBCC in view of the posture (lateral motion) stability using the transition of leg loading. For definitions of COST, COSS, DBCC, B1 and B2, please see the figure.