A method of personal positioning for indoor customer tracking utilizing wearable inertial sensors

Abstract

This paper presents a method for indoor personal positioning especially addressing on a simple gait measurement using a body-mounted inertial sensor, a walk path estimation algorithm, and a trajectory identification technique. Considering utility of on-site customer traffic tracking, the method was designed to be autonomous needless of neither external measures nor cumbersome installations to the environment. The inertial sensor combining three-dimensional accelerometer, angular rate sensors, and geomagnetic sensors was used to assess walking dynamics and to obtain changes of heading directions. To avoid an inherent problem of the walk trajectory estimation by conventional dead-reckoning algorithm, an advanced probabilistic map matching method which employs a Particle filter was developed. The experiment with healthy male adults was evaluated to confirm the utility of the method in a condition of small retail store. The result showed that the proposed method demonstrated sufficiently feasible tacking which delineated personal trajectories and positioning while shopping. The method showed possible availability with the portable instrument and improved estimation accuracy by the advanced probabilistic map matching which identifies proper routings on the basis of the maximum likelihood. The method will be useful to enhance ambulatory personal tracking technique in daily indoor environment.