Abstract
In this paper, the quantitative analysis of a gait as a stochastic process, and its experimental application are presented.
First of all, the gait pattern is described with the statistics of the measured data of the gait cadence and the step length. And the gait pattern is evaluated with the indexes of the symmetry and the state variable variance which are calculated also from the same data.
Next, with the application of an ergodic Markov chain, the gait pattern is analyzed and its change is predicted.
And, with the application of an absorbing Markov chain, the rehabilitation plan can be made on a sound basis.
Furthermore, as a practice of the above, it is shown that a malformed gait may be reformed systematically making a full use of the training system which the authors manufactured for trial.
