Analysis of the coupling of communication network and safety application in cooperative collision warning systems

Abstract

Cooperative collision avoidance systems rely on communication between vehicles to achieve the objective of automated or human-dependent crash avoidance. In this paper we investigate the mutual coupling of communication component and the safety application in cooperative collision warning systems. These systems are warning based collision avoidance systems that are currently under field test. We present a comprehensive co-simulation modeling framework which allows modeling and study of the entire system including vehicle dynamics, communication protocols, and collision detection/warning algorithms. Using this model, we show that in designs where the safety application and communication components are designed separately and agnostic to each other, system performance requires significantly higher network resources. Alternate content- and network-aware design strategies are shown to significantly reduce the required resources, resulting in significant reliability improvements. However, the cost of such strategies is mutual coupling of the performance of safety application and communication components. We show that such coupling can be effectively controlled in desired operation ranges for each component, leading to robust systems. The presented framework introduces a method for the study of a wide spectrum of communication dependent vehicular cyber-physical systems.