Graduation Semester and Year
Summer 2024
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Aerospace Engineering
Department
Mechanical and Aerospace Engineering
First Advisor
Animesh Chakravarthy
Second Advisor
Kamesh Subbarao
Third Advisor
Alan Bowling
Fourth Advisor
Shuo Linda Wang
Fifth Advisor
William Beksi
Abstract
The study of potential cyber-attacks in different domains is an active area of research. Given that systems are becoming more and more interconnected, cyber physical systems that operate infrastructure and/or plants can make these assets more vulnerable and open to different attack vectors. The primary focus of this research is the modeling, analysis and detection of cyber-attacks on platoons of autonomous cars and swarms of UAVs. In this work, we consider scenarios wherein an attacker may hack into a subset of vehicles in a multi-vehicle system and make subtle modifications in their parameters. Due to the interconnected nature of the multi-vehicle system, these hacked vehicles (referred to as malicious vehicles) are subsequently able to modify the behavior of the entire system. The multi-vehicle systems considered in this research include vehicle platoons on highway stretches, vehicles driving on interconnected road networks, as well as flying UAV swarms. The heterogeneous two species (normal and malicious vehicles) multi-vehicle systems are modeled using macroscopic PDE (Partial Differential Equation) models. Such models can describe collective phenomena such as the evolution of high-density regions as well as the propagation of traffic waves in multi-vehicle systems in a computationally efficient manner. These models are subsequently analyzed using both analytical and machine learning techniques such as Gaussian Process Regression (GPR) methods, to detect the occurrence of malicious attacks, and quantify the number and distribution of malicious vehicles in the multi-vehicle system. These analyses are verified and supported by non-linear PDE simulations.
Keywords
Traffic Models, UAV Swarms, Partial Differential Equations, Cyber attacks, Fault detection, Machine learning, Multi-Vehicle Systems, Traffic Control
Disciplines
Multi-Vehicle Systems and Air Traffic Control | Partial Differential Equations
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Kashyap, Abhishek, "Analysis and Detection of Cyber Attacks in Multi Vehicle Systems using Macroscopic Models" (2024). Mechanical and Aerospace Engineering Dissertations. 250.
https://mavmatrix.uta.edu/mechaerospace_dissertations/250
Included in
Multi-Vehicle Systems and Air Traffic Control Commons, Partial Differential Equations Commons