Clemson University
Cyber-Threat Detection Strategies Governed by an Observer and a Neural-Network for an Autonomous Electric Vehicle
Abstract
dc:description.abstractA pathway to prevalence for autonomous electrified transportation is reliant upon accurate and reliable information in the vehicle’s sensor data. This thesis provides insight as to the effective cyber-attack placements on an autonomous electric vehicle’s lateral stability control system (LSCS). Here, Data Integrity Attacks, Replay Attacks, and Denial-of-Service attacks are placed on the sensor data describing the vehicle’s actual yaw-rate and sideslip angle. In this study, there are three different forms of detection methods. These detection methods utilize a residual metric that incorporate sensor data, a state-space observer, and a Neural-Network. The vehicle at hand is a four-motor drive autonomous electric vehicle that is propelled using 4-pole, 3-phase Brushless DC motors. Each motor is controlled using the Direct-Torque control motor control scheme that provides fast output torque response time. This vehicle is controlled via multiple layers of control. A Model Predictive Control Layer is used to discern what lateral trajectory commands minimize the difference between the requested and actual lateral position of the vehicle. These lateral motions are discovered through a Linear-Quadratic Regulator. This study was develop using the MATLAB Simulink environment.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Electrical and Computer Engineering (Holcomb Dept. of)
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Scruggs, Douglas
- Contributors dc:contributor
-
- Behnaz Papari
- Christopher Edrington
- Fatemeh Afghah
Subjects
dc:subject × 8Identifiers
dc:identifier.*- Repository record dc:identifier
- https://open.clemson.edu/all_theses/4186