Project Description: LiDAR Path Following
Various autonomous systems, such as self-driving cars, autonomous ground robots, and drones, are already used in commercial and military services such as self-driving taxis, terrain exploration, and contamination handling. Performing such complex tasks requires a centimeter-level precision in estimating its own position with respect to the surroundings. Previous work has demonstrated that EMI injections can manipulate the LiDAR-based perception in autonomous systems: (Bhupathirajus et al.) (EMI-LiDAR: Uncovering Vulnerabilities of LiDAR Sensors in Autonomous Driving Setting Using Electromagnetic Interference). This project aims to characterize and evaluate the performance of state-of-the-art path following techniques under EMI injection attack.
● Review LiDAR-based path following algorithms used in Autonomous Vehicles and Robots.
● Use and classify APIs of existing simulation environment of LiDAR sensors.
● Develop a Python/C++/ROS-based implementation of path following algorithms.
● Characterizing the algorithm’s behavior with simulated EMI injection attacks.
Each member of a group was required to submit their own project report detailing their contribution.
Instructions to reproduce my Python-based implementation can be found in KITTI-Pole-Localization.txt