A Novel Zero Velocity Interval Detection Algorithm for Self-Contained Pedestrian Navigation System with Inertial...

Area of Science:

• * Inertial Navigation Systems
• * Signal Processing
• * Biomechanics

Background:

• * Accurate zero velocity interval (ZVI) detection is crucial for pedestrian navigation algorithms using zero velocity update (ZUPT).
• * Traditional methods often struggle with precision and robustness due to fixed detection thresholds.
• * The need for adaptive ZVI detection that accounts for varying pedestrian gait is evident.

Purpose of the Study:

• * To propose a novel adaptive ZVI detection algorithm for enhanced pedestrian navigation.
• * To improve ZVI detection accuracy and reduce false/missed detections.
• * To evaluate the algorithm's impact on pedestrian trajectory positioning precision.

Main Methods:

• * Development of an adaptive ZVI detection algorithm utilizing a smoothed pseudo Wigner-Ville distribution to remove multiple frequencies (SPWVD-RMFI).
• * Real-time extraction of pedestrian gait frequency using SPWVD-RMFI.
• * Calculation of optimal ZVI detection thresholds dynamically based on gait frequency.
• * Implementation of adaptive threshold adjustment linked to gait frequency.

Main Results:

• * The adaptive ZVI detection algorithm significantly reduces false and missed detection rates compared to fixed threshold methods.
• * Experimental validation demonstrates high detection precision and robustness of the proposed algorithm.
• * Pedestrian trajectory positioning experiments show improved performance when using ZVI detected by the adaptive algorithm across different walking speeds.

Conclusions:

• * The proposed adaptive ZVI detection algorithm offers superior precision and robustness for pedestrian navigation.
• * Dynamic threshold adjustment based on gait frequency is key to improving ZVI detection.
• * The algorithm enhances overall pedestrian trajectory positioning accuracy.