Localization development for CMU’s driverless FSAE team
- Implemented a real-time Extended Kalman Filter to fuse multi-sensor data for localization, validated against rosbag recordings using Foxglove and deployed for autonomous operation
- Developed a simulation tool in Python that tests candidate racing lines using vehicle dynamics models, enabling rapid evaluation and iteration
- Implemented minimum-curvature trajectory optimization, reducing simulated lap times by 20%