Optimal routing for autonomous vehicles in a mixed congested network considering fairness

The field of transportation has undergone a revolutionary transformation with the advent of Autonomous Vehicles (AVs). The ability to centrally control the route choices of these vehicles presents a promising opportunity to minimize total travel time within networks and effectively alleviate congestion. Recent studies have modelled Human-Driven Vehicles (HDVs) as selfish drivers who prioritize user-optimum routes, while AVs strive to optimize the system by reducing congestion. However, this routing approach may be perceived as unfair by AV users, leading to potential challenges. In this paper, we propose a heuristic framework to solve the dynamic mixed traffic flow problem using a simulation-based simulator and address this issue by providing equitable paths to AVs. These routes not only result in a significant reduction in the total system travel time (TSTT) but also minimize the need for AVs to make excessive compromises.