Attack detection and estimation in cooperative vehicles: A sliding mode observer approach

Platoons of autonomous vehicles are currently being investigated by academic and industrial researchers as a way to increase road capacity and fuel efficiency. In order to fully reach such goals, a platoon must be endowed with cooperative capabilities, such as Cooperative Adaptive Cruise Control (CACC). This technique is based on the vehicles’ sensors and on wireless communication between them, in order to control their longitudinal dynamics. However, the use of wireless communication exposes individual vehicles to cyber-attacks that aim at disrupting the platoon. Detecting and estimating a class of such attacks is the challenge considered in this paper, where an adaptive sliding mode observer is designed for this purpose. Theoretical results on the observer stability and robustness and simulation results are provided.