GPS & IMU Sensor Fusion for Automotive Dead Reckoning

Overview

Developed a sensor fusion system combining GPS and IMU data through Extended Kalman Filtering for reliable vehicle positioning.

The Challenge

GPS signals can be unreliable in urban environments with multipath effects and signal blockage. IMU sensors accumulate drift over time.

The Solution

Using MATLAB and ROS, implemented Extended Kalman Filter for optimal sensor fusion with real-time trajectory correction algorithms.

Key Results

• Reliable positioning with sub-meter accuracy in GPS-challenged environments
• Real-time performance suitable for automotive applications
• Maintained accurate navigation through urban GPS multipath zones
• Validated EKF fusion against ground truth with consistent drift correction