Attack Mitigation in Adversarial Platooning Using Detection-Based Sliding Mode Control

Abstract

In this paper, we consider a mitigation strategy to prevent a vehicle controlled by an attacker from causing collisions in a vehicular platoon. An adversarial-aware control scheme, based on sliding mode control using only local sensor information and a decentralized attack detector, is shown to significantly reduce the number and severity of collisions, without the need for inter-vehicle or vehicle-to-infrastructure communication. Simulations demonstrate that collisions are eliminated (or significantly reduced) when the attacker and normal vehicles have same capabilities, and collisions are reduced even with more powerful attackers.