Cooperative autonomous traffic organization method for connected automated vehicles in multi-intersection road networks
Yunpeng Wang,Pinlong Cai,Guangquan Lu
DOI: https://doi.org/10.1016/j.trc.2019.12.018
2020-01-01
Abstract:Connected automated vehicles (CAVs) have been currently considered as promising solutions for realization of envisioned autonomous traffic management systems in the future. CAVs can achieve high desired traffic efficiency and provide safe, energy-saving, and comfortable ride experience for passengers. However, in order to practically implement such autonomous systems based on CAVs, there exist several significant challenges to be dealt with, such as coupled spatiotemporal constraints on CAVs' trajectories at unsignalized intersections, multiple objectives for trajectory optimization in road segments, and heterogeneous decision-making behaviors of CAVs in road networks with highly dynamic traffic demand. In this paper, we propose a cooperative autonomous traffic organization method for CAVs in multi-intersection road networks. The methodological framework consists of threefold components: an autonomous crossing strategy based on a conflict resolution approach at unsignalized intersections, multi-objective trajectory optimization in road segments, and a composite strategy for route planning considering heterogeneous decision-making behaviors of CAVs based on social and individual benefit, respectively. Specifically, we first identify a set of potential conflict points of different CAVs' spatial trajectories at the intersection, and then design different minimum safe time headways to resolve conflicts. Under the constraints of entry and exit conditions at adjacent intersections, we propose a multi-objective optimal control model by jointly considering vehicle safety, energy conservation, and ride comfort, and then analytically derive a closed-form solution for optimizing the CAVs' trajectories. Furthermore, with the purpose to adapt dynamic traffic demand, we propose a composite strategy for route planning by coordinating heterogeneous decision-making behaviors of CAVs in road networks. Finally, extensive simulation experiments have been performed to validate our proposed method and to demonstrate its advantage over conventional baseline schemes in terms of global traffic efficiency. Additional numerical results are also provided to shed light on the impact of the proportion of CAVs with heterogeneous decision-making behaviors on the global system performance.