Understanding Collective Stability of ACC Systems: From Theory to Real-World Observations

Abstract: Autonomated vehicles (AVs) are expected to have a profound impact on society, with high expectations for their potential benefits. One key anticipated benefit is the reduction of traffic congestion and stop-and-go waves, which negatively affect fuel efficiency, travel time, and environmental sustainability. This paper presents a comprehensive review and meta-analysis of the impact of AVs on the longitudinal collective stability of vehicle single file. We focus on adaptive cruise control (ACC) systems, a widely used precursor technology to fully autonomous driving. ACC controllers have been studied extensively in both theoretical and practical contexts, making it a valuable starting point for analysis. Our study systematically differentiates the findings from models and simulations, controlled experiments, and empirical observations to provide a structured overview of existing research. Although some results in the literature are contradictory, three key insights emerge from our analysis: i) String stability is highly dependent on the chosen time gap and reaction/response time. ii) ACC systems currently implemented in commercial vehicles are not string stable, as manufacturers prioritise individual comfort and smooth driving, resulting in high system response times. iii) Although cooperative ACC (CACC) systems are theoretically the most effective solution to ensure string stability, their widespread implementation in the near future remains uncertain. Instead, improvements in autonomous control algorithms should be considered to enhance system performance.