Attack-Resilient Sensor Fusion for Cooperative Adaptive Cruise Control

Abstract

Cooperative adaptive cruise control (CACC) has the potential to enable vehicle platooning and achieve benefits including improved highway throughput and reduced energy consumption. However, malicious attacks such as sensor jamming or data injection can lead to security vulnerabilities of vehicle platooning and cause catastrophic crashes. We present a novel attack-resilience sensor fusion method for vehicle platooning with CACC, which exploits spatial information provided by multiple vehicles and combines sensor readings to achieve more precise estimation. We demonstrate the feasibility of our method in a set of simulated vehicle platooning experiments with different CACC controllers and malicious attacks.