Detection of Network and Sensor Cyber-Attacks in Platoons of Cooperative Autonomous Vehicles: a Sliding-Mode Observer Approach

TL;DR

This paper proposes a sliding-mode observer method to detect cyber-attacks on autonomous vehicle platoons, analyzing its effectiveness against various attack types to enhance safety and reliability.

Contribution

It introduces a novel SMO-based approach for cyber-attack detection in vehicle platoons, with theoretical analysis and simulation validation of its robustness.

Findings

The SMO approach effectively detects multiple attack types.

Theoretical properties of attack detectability are established.

Simulations demonstrate the method's ability to prevent undetected harm.

Abstract

Platoons of autonomous vehicles are being investigated as a way to increase road capacity and fuel efficiency. Cooperative Adaptive Cruise Control (CACC) is an approach to achieve such platoons, in which vehicles collaborate using wireless communication. While this collaboration improves performance, it also makes the vehicles vulnerable to cyber-attacks. In this paper the performance of a sliding mode observer (SMO) based approach to cyber-attack detection is analysed, considering simultaneous attacks on the communication and local sensors. To this end, the considered cyber-attacks are divided into three classes for which relevant theoretical properties are proven. Furthermore, the harm that attacks within each of these classes can do to the system while avoiding detection is analysed based on simulation examples.