Project Overview
Unmanned Aerial Vehicles (UAVs) are essential in various applications, including agriculture, disaster management, and military operations, where location integrity and situational awareness are critical. However, methods that rely on GPS signals face significant challenges in extreme conditions, such as signal unavailability (e.g., navigating crowded urban areas) and signal compromise (e.g., GPS spoofing attacks). This project addresses these challenges with data-driven approaches that minimize reliance on GPS information, ensuring UAV location integrity and awareness.
Our Approach
This project investigates a trajectory-based analysis inspired by the nautical practice of “dead reckoning.” This method estimates the position of a moving object based on its movement data, including previous position, velocity, direction, and elapsed time. Our approach is thus data-driven, utilizing UAV movement updates to offer location integrity and awareness without solely relying on GPS information.
Objectives
We aim to develop a cutting-edge deep analytics framework to provide UAV location checks and estimation to ensure its integrity and awareness using machine intelligence. Our project's key innovations include:
- Movement-based validation and estimation of current coordinates, even without GPS.
- Architectural designs to analyze flight data sequences via trajectory-based analysis.
- One-class learning to build deep sequence models focusing on normal flight data, addressing the scarcity of abnormal data with inconsistent location information.
- On-board inference transforming the deep sequence models into lightweight equivalents to effectively support deep inference on UAVs with recourse-constrained AI engines.
Ultimately, this project provides a robust solution to existing GPS-based methods, ensuring reliable UAV operations in challenging environments.
Project Leadership
Led by Dr. Jinoh Kim, Professor of Computer Science at East Texas A&M University (ETAMU), this project receives support from the National Science Foundation under grant ECCS-2502117. We also collaborate with Dr. Sang-Yoon Chang, Computer Science Department at the University of Colorado, Colorado Springs (UCCS).
Contact Us
For inquiries or further information, please reach out: