A METHOD USING FEET DISTANCE AND INERTIAL SENSORS DATA FOR GAIT EVENTS DETECTION

Gait data are usually normalized in to gait cycle to simplify analysis. A gait cycle start with initial foot contact (stance phase beginning). Swing phase begins with end of foot contact. This requires the accurate determination of timing of foot-contact (FC) and footoff (FO) events. Optoelectronic systems remain the most accurate non-wearable sensors to perform gait analysis in indoor environments [1]. Among wearable sensors, inertial measurement units (IMU) and rangefinders can be used to overcome limitations of widely used foot-switches [2]. We propose a new method to automatically determinate the gait events using data fusion from radio-frequency range-finders and IMUs. Detected gait events are compared with those coming from the reference motion capture system.