Dipartimento di Ingegneria informatica, automatica e gestionale

The use of virtual reality in motor rehabilitation has gained significant interest in recent years. To track a subject's motion in a virtual reality environment, a real time estimation of his/her movement is necessary. Wearable inertial measurement units represent one of the most suitable technologies to do so. However, it is well known how their recorded signals, once integrated, are affected by unpredictable drift, introducing errors that are hard to cope with. A number of successful methods have been proposed in the recent literature for the reduction of such errors for offline analysis. Unfortunately, those methods may not be used for real-time estimates of the foot displacement. In this paper we propose a method to estimate the foot displacement in real-time using an inertial measurement unit attached to it while mitigating integration errors.