Search for: All records

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 en- vironment 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. 

Emergent vehicles will support a variety of connected applications, where a vehicle communicates with other vehicles or with the infrastructure to make a variety of decisions. Cooperative connected applications provide a critical foundational pillar for autonomous driving, and hold the promise of improving road safety, efficiency and environmental sustainability. However, they also induce a large and easily exploitable attack surface: an adversary can manipulate vehicular communications to subvert functionality of participating individual vehicles, cause catastrophic accidents, or bring down the transportation infrastructure. In this paper we outline a potential direction to address this critical problem through a resiliency framework, REDEM, based on machine learning. REDEM has several interesting features, including (1) smooth integration with the architecture of the underlying application, (2) ability to handle diverse communication attacks within the same underlying foundation, and (3) real-time detection and mitigation capability. We present the vision of REDEM, identify some key challenges to be addressed in its realization, and discuss the kind of evaluation/analysis necessary for its viability. We also present initial results from one instantiation of REDEM introducing resiliency in Cooperative Adaptive Cruise Control (CACC).