← Back to all projects

Position Estimation with Dead Reckoning

Project Overview

This project focused on developing a robust algorithm for position estimation. The goal was to combine data from multiple sensors to track an object's position and orientation, even when GPS signals are temporarily lost.

Technical Details

The solution uses sensor fusion techniques, primarily a Kalman filter, to merge data from an Inertial Measurement Unit (IMU), a magnetometer, and a GPS receiver. The IMU provides high-frequency acceleration and rotation data, which is used for dead reckoning. However, IMUs suffer from drift over time. The magnetometer helps correct for rotational drift, and the GPS provides absolute position updates to correct for translational drift. The Kalman filter optimally combines these sources to produce a smooth and accurate position estimate.