Formation control for connected and automated vehicles on multi‐lane roads: Relative motion planning and conflict resolution

Abstract

Existing research has revealed that multi-vehicle coordinated decision making and control can achieve an improvement in both traffic efficiency and driving safety. In the multi-lane scenarios, a typical coordination method is multi-vehicle formation control. The existing formation control methods predefine the formation switching process and have not considered or explained the collision-free behaviour of vehicles in detail. In this paper, a multi-lane formation control strategy for connected and automated vehicles (CAVs) is proposed. The planning framework is bi-level, which can switch the structure of multi-vehicle formation in different scenarios smoothly and effectively. In the upper-level, the relative coordinate system is built to plan the collision-free relative paths for vehicles. In the lower-level, multi-stage trajectory planning and tracking are modelled as an optimal control problem with path constraints. The case study verifies the function of the formation control method in three lane-number-changing scenarios. Large-scale simulations in the lane-drop bottleneck scenario are conducted under different input traffic volumes, and the numerical results indicate that the proposed formation control method reduces congestion and improves both traffic efficiency and fuel economy at high traffic volumes.