Control of a passive walker using a depth sensor for user state estimation

Abstract
The motion of a passive type walker is controlled so that it can be used as a support for standing up, sitting down and also to prevent the user from falling down. A depth vision sensor is used to locate the centroid of the user's upper body. A 2D probabilistic model of the location of the centroid is constructed based on the gathered data of the user during normal walking. A set of statistical normality tests assure the validity of such model. The state estimation method is experimented with a passive-type intelligent walker referred to as “RT Walker”, equipped with servo brakes.

This publication has 14 references indexed in Scilit: