GA-PDR: Using Gait Analysis for Heading Estimation in PDR-Based Indoor Localization System
<- Publications
GA-PDR: Using Gait Analysis for Heading Estimation in PDR-Based Indoor Localization System
Renjie Wu,
Boon Giin Lee,
Matthew Pike,
Wan-Young Chung,
Xiaoqing Chai,
Lionel Nkenyereye
IECON 2023 - 49th Annual Conference of the IEEE Industrial Electronics Society | 2023
| View on Publisher's Website
Abstract
Indoor positioning in firefighting environments presents unique challenges due to low visibility and signal interference from smoke. Conventional pedestrian dead reckoning (PDR) methods struggle with unpredictable user gaits, leading to inaccurate heading estimation. This paper introduces a gait analysis–based PDR (GA-PDR) approach to enhance localization accuracy. By analyzing inertial measurement unit (IMU) data, the proposed method classifies step patterns, including forward movement, left and right turns, and full rotations. Additionally, a redundant turn elimination technique refines heading estimation by mitigating false positives. The GA-PDR system is evaluated using an experimental dataset collected under smoke-filled conditions, demonstrating significantly reduced loop closure errors compared to traditional PDR methods. The results highlight the effectiveness of GA-PDR in providing accurate indoor localization, particularly in hazardous firefighting scenarios where conventional techniques fail.