Real-time estimation of human´s intended walking speed for treadmill-style locomotion interfaces

This paper addresses the problem of estimating human´s intended walking speed. Compared with previous researches on walking speed estimation, we predict the walking intention before gait action. In this paper, we find a composite force index which is significantly correlated with the intended walking speed. We did two experiments. One gives a conclusion that intended walking speed has strong linear correlation with the proposed force index; The other shows that the coefficients of the linear relation vary small, guaranteeing the tolerance of individual variation. Finally, we built a treadmill-style locomotion interface. Compared with the normal cases of treadmill control, the tested subject does not have to follow the speed of treadmill, but can actively changes the speed of treadmill by his (or her) feet. This locomotion interface not only shows the validity of the proposed solution, but also provides a promising human machine interface (HMI) for entertainment, healthcare and rehabilitation.