Harmony in heterogeneous traffic flow: A lattice hydrodynamic model with perception diversity and predictive effect coordinated by bilaterally controlled CAVs

Abstract

With the rapid advancement of intelligent network communication and autonomous driving technologies, future highway traffic will be characterized by a heterogeneous mix of Human Driven Vehicles (HDVs), Autonomous Vehicles (AVs) at various automation levels, and Connected Autonomous Vehicles (CAVs). To evaluate the effectiveness of CAVs in mitigating traffic congestion, this paper proposes an extended lattice hydrodynamic model by integrating the concept of “flow control through particles” into mixed heterogeneous traffic flow scenarios. The model incorporates various vehicles' perceptions of upstream traffic conditions based on predictive capabilities. Furthermore, it investigates the impact of CAV Market Penetration Rate (MPR) on the system's critical stability, considering the sensitivity of diverse vehicle operators. Additionally, a car following model with CAV navigation is developed, considering bilateral control to analyze the string stability of the platoon. Through theoretical analysis and numerical simulations, the study demonstrates that the system's stability region expands significantly with increasing prediction horizon, sensing capability, sensitivity, and CAV penetration rate. Moreover, empirical data simulations demonstrate that implementing even a small fraction of the proposed CAV models can significantly reduce traffic congestion and improve overall roadway efficiency.