Exploring the Feasibility of Inertial Hand Tracking Based on one Sensor

Abstract

This paper presents our work in progress in which we use an inertial sensor to develop a hand tracking system. The main difficulty of this approach comes from the fact that the system is affected by drift errors that has an effect on the estimations of position and the angle around the vertical axis (yaw). It uses Kalman filtering to reduce the error of the estimations while the hand is steady. It also explores the use of the covariance error selective reset technique (CESR) which was employed successfully to track the movements of the foot using an inertial sensor. Preliminary results as well as several strategies that can contribute to improve the behavior of the tracking system are also presented.