Security attacks impact for collective perception based roadside assistance: A study of a highway on-ramp merging case

Abstract

Road critical situations like highway on-ramp merging require a high coordination level between vehicles within a short time while they are driving at high velocity. Augmented perception system that can provide information about non-communicating vehicles in timely manner can drastically help vehicles to safely merge on the highway and avoid incidents. However, this reliance on coordination by information exchange between vehicles requires resilience against cyber attacks and against any other driver misbehavior or error. In this paper, we proposed and implemented a perception model that has been integrated in the merging control algorithm proposed in the literature to enhance the capability of the Road Side Unit (RSU) to detect the surrounding objects on the road. Moreover, we developed in this paper several attack models and we measure through simulations their impacts on collective perception-based on-ramp merging control algorithm.