Pedestrian Travel Direction Estimation Using Acceleration Ellipse Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile applications struggle to accurately and efficiently estimate pedestrian direction of travel (DoT) without additional hardware or complex software algorithms, particularly distinguishing between walking forward and backward.
Innovation Solution
Utilizing inertial acceleration data from a mobile device's motion sensor, the method computes a frequency spectrum to generate horizontal and vertical frequency components, constructs an acceleration trajectory as an ellipse, determines the maximum radius vector, and disambiguates the angle sign to estimate DoT.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If frequency-domain methods are used to estimate DoT, then computational efficiency is improved, but the complexity of signal processing increases
Solution Approach 1:
The patent transforms the acceleration signal from time-domain to frequency-domain by computing the Fourier transform at the step frequency. This parameter change allows efficient extraction of directional information through frequency spectrum analysis, achieving real-time DoT estimation with computational efficiency while managing processing complexity through focused frequency analysis.
2Measurement precision
If additional hardware is added to improve DoT estimation accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent utilizes the mobile device's existing motion sensors (accelerometer, gyroscope) to estimate DoT, making the device self-sufficient for directional estimation. By processing the acceleration data through frequency-domain analysis and ellipse fitting, the system achieves accurate DoT estimation without requiring additional specialized hardware, thus maintaining measurement precision while avoiding increased hardware complexity.
3Device complexity
If time-domain methods are used for DoT estimation, then algorithm simplicity is improved, but computational efficiency deteriorates
Solution Approach 1:
The patent replaces traditional time-domain signal processing methods with frequency-domain analysis. By computing the Fourier transform of the acceleration signal and analyzing the frequency spectrum at the step frequency, the system achieves real-time DoT estimation with improved computational efficiency. The ellipse fitting method in the frequency domain provides a systematic approach that balances algorithm complexity with processing speed.
Data Source
AI summary
Embodiments are disclosed for real time estimation of DoT. In some embodiments, a method comprises: obtaining inertial acceleration data from a motion sensor of a mobile device worn or held by a user; computing a frequency spectrum of the acceleration data to generate horizontal and vertical frequency components at a step frequency of the user; constructing an acceleration trajectory of the horizontal frequency components in a horizontal reference frame, the horizontal frequency components forming an ellipse in the horizontal reference frame; determining a maximum radius vector of the ellipse; determining an angle to DoT based on the maximum radius vector; disambiguating a sign of the angle; and determining the disambiguated angle as the DoT.