Date of Award:
Master of Science (MS)
Electrical and Computer Engineering
Ryan M. Gerdes
Highway systems world wide continue to see an ever increased number of vehicles and subsequently a rise in congested traffic. This results in longer commute times, wasted energy as vehicles idle in stop and go traffic, and increases the risk of accidents. In short, increased congestion costs time and money. These issues have prompted much research into Automated Highway Systems (AHS). In AHS vehicles using computer algorithms can safely travel at much smaller inter-vehicle distances than human drivers are capable of. This increases the capacity of existing highway systems. Sensors aboard each vehicle make this possible by monitoring their surroundings. Vehicles equipped with Adaptive Cruise Control (ACC) are capable of this type of close proximity travel. ACC packages are becoming common as a standard package on many mid-priced vehicles.
Another form of automation, Cooperative Adaptive Cruise Control (CACC), which utilizes wireless communication between vehicles, has been proposed and will likely become available within the next couple decades. CACC allows each vehicle to communicate their intended speed or position changes to surrounding vehicles, further decreasing the possibility of collisions.
These automation methods are proposed to reduce driver stress, increase highway throughput, and decrease accident rates. However, the fact that vehicles are being automated creates new opportunities for malicious individuals to wreak havoc on society.
This research investigates the possibility that some vehicles on the highway might be under the control of malicious individuals who have modified their automated control systems to negatively affect vehicles around them. These malicious actors might also exploit the wireless communication of CACC vehicles and hack their control algorithms, causing them to become unstable. These hacked vehicles could become passive participants in the attack unbeknownst to the driver of the vehicle. The result of such attacks could be congested traffic, rapid changes in acceleration causing drivers discomfort, or multi-vehicle collisions. Such attacks could effectively negate the benefits of implementing AHS.
The goal of this work is to bring to light possible weaknesses in the proposed systems so they can be rectified before becoming an issue to the public at large.
Dunn, Daniel D., "Attacker-Induced Traffic Flow Instability in a Stream of Automated Vehicles" (2015). All Graduate Theses and Dissertations. 4455.