An Indoor Localization Method for Pedestrians Base on Combined UWB/PDR/Floor Map

Area of Science:

• Indoor Localization
• Ubiquitous Computing
• Sensor Fusion

Background:

• Accurate indoor pedestrian localization remains a challenge, particularly in complex environments with limited infrastructure.
• Existing methods often struggle with signal obstruction, multipath effects, and the need for extensive calibration.
• Integrating multiple sensor modalities and map information is crucial for robust indoor positioning.

Purpose of the Study:

• To propose a novel indoor pedestrian localization scheme combining Ultra Wideband (UWB), Pedestrian Dead Reckoning (PDR), and Floor Map data.
• To address positioning challenges in narrow corridors and scenarios with partial UWB anchor availability.
• To achieve high-precision, decimeter-level localization through sensor fusion.

Main Methods:

• Developed a robust UWB positioning algorithm using Tukey weight factor and pathological parameter to handle ill-conditioned scenarios.
• Introduced a heading angle computation strategy for PDR, aligning with virtual directions derived from UWB positioning on a floor map.
• Implemented an Extended Kalman Filter (EKF) for fusing UWB, PDR, and floor map data for enhanced localization accuracy.

Main Results:

• The proposed UWB algorithm effectively solves positioning problems in narrow corridors with co-elevation anchors.
• The EKF-based fusion of UWB, PDR, and floor map data achieved reliable decimeter-level positioning accuracy.
• The system demonstrated robust performance even when a portion of UWB data was unavailable.

Conclusions:

• The proposed UWB/PDR/Floor Map fusion scheme provides a reliable and accurate solution for indoor pedestrian localization.
• The method is effective in challenging indoor environments and resilient to partial sensor data loss.
• The system's decimeter-level accuracy has significant implications for various indoor navigation and tracking applications.