ALS+PDR: Indoor Pedestrian Dead Reckoning Using a Smartphone Ambient Light Sensor

Abstract

This article proposes an indoor position-estimation method that integrates visible light positioning (VLP) with pedestrian dead reckoning (PDR), using a smartphone's built-in ambient light sensor (ALS) offering lower power consumption than a camera and inertial sensor. In the proposed method, the user's position is first estimated via PDR and the positioning results for areas where VLP using ALS (ALS-VLP) is available are corrected by using pose graphs that resolve simultaneous localization and mapping. Experiments were conducted with eight users walking a route measuring 141.67 m for five laps. The results indicated an average error of 11.30 m when only PDR was used, with a substantial reduction to 2.03 m when the proposed method was used. Limitations and challenges related to practical use scenarios of the proposed method clarified through the experiments are discussed.