Consider a fleet of autonomous vehicles traversing an adversarial terrain that includes obstacles and mines. The goal of the fleet is to ensure that they can complete their mission safely (with minimal casualty) and efficiently (as quickly as possible). In Distributed Coordinated Fleet Management (DCFM), fleet members coordinate with one another while traversing the terrain, e.g., a vehicle encountering an obstacle at a location l can inform other agents so that they can recompute their route to avoid l. In this paper, we consider the problem of cyber-resilient DCFM, i.e., DCFM in an environment where the adversary can additionally tamper with the cyber-communication performed by the fleet members. Our framework, DRiFt, enables fleet members to coordinate in the presence of such adversaries. Our extensive evaluations demonstrate that DRiFt can achieve a high degree of safety and efficiency against a large spectrum of communication adversaries.

Cooperative Adaptive Cruise Control (CACC) is a fundamental connected vehicle application. In CACC, a vehicle coordinates its longitudinal movements to safely and efficiently follow the vehicle in front. The follower vehicle relies on a combination of sensory and communication inputs to identify the position, velocity, and acceleration of the preceding vehicle. Malicious subversion of these inputs can cause catastrophic accidents, string instability, and disruption in the transportation infrastructure. In this paper, we develop a security system, ReCAP, to provide real-time resiliency in CACC against adversarial subversion of both sensory and communication inputs. ReCAP makes use of a combination of techniques based on kinematics and machine learning to detect anomalous inputs, narrow down the source of subversion, and perform mitigation. We provide extensive simulations to demonstrate the effectiveness of ReCAP against a diverse spectrum of attacks under complex, multi-channel adversaries.

Vehicular communication systems and their applications have rapidly grown in recent years with the proliferation of cooperative applications. Unfortunately, vehicular network applications can be susceptible to cybersecurity attacks, disrupting the vehicular ecosystem or even causing fatal injuries. Unfortunately, platforms to enable realistic exploration of these vulnerabilities are limited. We address this critical need through the design of a new exploration platform, VeCAEP, to enable comprehension of communication attacks. VeCAEP permits the user to explore diverse communication attacks and comprehend interactions of different attack parameters and their impacts on the attack. We demonstrate VeCAEP with attacks on Cooperative Adaptive Cruise Control.

With the proliferation of connectivity and smart computing in vehicles, a new attack surface has emerged that targets subversion of vehicular applications by compromising sensors and communication. A unique feature of these attacks is that they no longer require intrusion into the hardware and software components of the victim vehicle; rather, it is possible to subvert the application by providing wrong or misleading information. We consider the problem of making vehicular systems resilient against these threats. A promising approach is to adapt resiliency solutions based on anomaly detection through Machine Learning. We discuss challenges in making such an approach viable. In particular, we consider the problem of validating such resiliency architectures, the factors that make the problem challenging, and our approaches to address the challenges.

Vehicles can utilize their sensors or receive messages from other vehicles to acquire information about the surrounding environments. However, the information may be inaccurate, faulty, or maliciously compromised due to sensor failures, communication faults, or security attacks. The goal of this work is to detect if a lane-changing decision and the sensed or received information are anomalous. We develop three anomaly detection approaches based on deep learning: a classifier approach, a predictor approach, and a hybrid approach combining the classifier and the predictor. All of them do not need anomalous data nor lateral features so that they can generally consider lane-changing decisions before the vehicles start moving along the lateral axis. They achieve at least 82% and up to 93% F1scores against anomaly on data from Simulation of Urban MObility (SUMO) [1] and HighD [2]. We also examine system properties and verify that the detected anomaly includes more dangerous scenarios.

Cooperative Adaptive Cruise Control (CACC) is a fundamental connected vehicle application that extends Adaptive Cruise Control by exploiting vehicle-to-vehicle (V2V) communication. CACC is a crucial ingredient for numerous autonomous vehicle functionalities including platooning, distributed route management, etc. Unfortunately, malicious V2V communications can subvert CACC, leading to string instability and road accidents. In this paper, we develop a novel resiliency infrastructure, RACCON, for detecting and mitigating V2V attacks on CACC. RACCON uses machine learning to develop an on-board prediction model that captures anomalous vehicular responses and performs mitigation in real time. RACCON-enabled vehicles can exploit the high efficiency of CACC without compromising safety, even under potentially adversarial scenarios. We present extensive experimental evaluation to demonstrate the efficacy of RACCON.

Modern System-on-Chip (SoC) designs integrate a number of third party IPs (3PIPs) that coordinate and communicate through a Network-on-Chip (NoC) fabric to realize system functionality. An important class of SoC security attack involves a rogue IP tampering with the inter-IP communication. These attacks include message snoop, message mutation, message misdirection, IP masquerade, and message flooding. Static IP-level trust verification cannot protect against these SoC-level attacks. In this paper, we analyze the vulnerabilities of system level communication among IPs and develop a novel SoC security architecture that provides system resilience against exploitation by untrusted 3PIPs integrated over an NoC fabric. We show how to address the problem through a collection of fine-grained SoC security policies that enable on-the-fly monitoring and control of appropriate security-relevant events. Our approach, for the first time to our knowledge, provides an architecture-level solution for trusted SoC communication through run-time resilience in the presence of untrusted IPs. We demonstrate viability of our approach on a realistic SoC design through a series of attack models and show that our architecture incurs minimal to modest overhead in area, power, and system latency.