Adaptive Step Length Estimation with Awareness of Sensor Equipped Location for PNS

Abstract

In this paper we present an adaptive step length estimation algorithm with the awareness of the inertial measurement unit (IMU) location for pedestrian dead reckoning (PDR). The developed PDR is assumed to be embedded in mobile devices. Thus it is supposed that the IMU can be equipped on two locations; on a waist belt or in a trouser pocket of the pedestrian. In this paper the step length is estimated by a simple linear combination of a constant, walking frequency and acceleration variance. It is investigated that the step length estimation parameters should be changed if PDR is equipped on different parts of the human body. Otherwise it can cause considerable error in the step length estimation. In order to enhance the performance of the step length estimation algorithm, we present to adaptively estimate the step length according to the PDR equipped location. And then the walking distance of the user is calculated in real-time. The performance of the proposed algorithm is verified through the actual walking tests.