SmartPDR: Smartphone-Based Pedestrian Dead Reckoning for Indoor Localization

TLDR

Indoor pedestrian tracking extends location‑based services to indoor environments where GPS is rarely available, yet existing Wi‑Fi methods are costly and require infrastructure, motivating infrastructure‑free solutions. We present SmartPDR, a smartphone‑based pedestrian dead reckoning system that tracks users using only inertial sensors. SmartPDR employs standard dead‑reckoning on off‑the‑shelf smartphones without additional devices, and was evaluated in several buildings. Experimental results show that despite sensor noise and complex movements, SmartPDR achieves reasonable indoor location accuracy, demonstrating its practical viability.

Abstract

Indoor pedestrian tracking extends location-based services to indoor environments where GPS signal is rarely detected. Typical indoor localization method is Wi-Fi-based positioning system, which is practical showing accuracy and extending coverage. However, it involves significant costs of installing and managing wireless access points. A practical indoor pedestrian-tracking approach should consider the absence of any infrastructure or pretrained database. In this paper, we present a smartphone-based pedestrian dead reckoning, SmartPDR, which tracks pedestrians through typical dead reckoning approach using data from inertial sensors embedded in smartphones. SmartPDR does not require any complex and expensive additional device or infrastructure that most existing pedestrian tracking systems rely on. The proposed system was implemented on off-the-shelf smartphones and the performance was evaluated in several buildings. Despite inherent localization errors from low-cost noisy sensors and complicated human movements, SmartPDR successfully tracks indoor user's location, which is confirmed from the experimental results with reasonable location accuracy. Indoor pedestrian tracking system using smartphone inertial sensors can be a promising methodology validating its practical usage through real deployment.