GPS Kalman Filter

Python implementation of Extended Kalman Filter (EKF) and Nonlinear Least Squares (NLS) for GPS position estimation using real satellite data.

Installation

uv sync

uv sync --group dev

Usage

uv run python -m gps.cli

uv run python -m gps.cli --filter ekf --sigma 5.0

uv run python -m gps.cli --plot

uv run python -m gps.cli --help

Development

uv run pytest

Results

With the provided dataset ($\sigma^2 = 5.0$ for EKF):

Filter	2D RMS Error	3D RMS Error
EKF	3.49 m	4.00 m
NLS	4.47 m	10.27 m