Multi-automated vehicle coordination using decoupled prioritized path planning for multi-lane one- and bi-directional traffic flow control

Abstract

Within the context of autonomous driving, this paper presents a method for the coordination of multiple automated vehicles using priority schemes for decoupled motion planning for multi-lane one- and bi-directional traffic flow control. The focus is on tube-like roads and non-zero velocities (no complete standstill maneuvers). We assume inter-vehicular communication (car-2-car) and a centralized or decentralized coordination service. We distinguish between different driving modes including adaptive cruise control (ACC) and obstacle avoidance (OA) for the handling of dynamic driving situations. We further assume that any controllable vehicle is equipped with proprioceptive and exteroceptive sensors for environment perception within a particular range field. In case of failure of the inter-vehicle communication system, the controllable vehicles can act as autonomous vehicles. The motivation is the control of a) one-directional multi-lane roads available for automated as well as unautomated objects with potentially, but not necessarily, varying reference speeds, and b) bi-directional traffic flow control making use of all available lanes, allowing, in general, object- and direction-wise variable reference speeds. For the one-directional case, we discuss a suitable deterministic priority scheme for throughput maximization and quickly reaching of a platooning state. For the bi-directional scenario, we derive a binary integer linear program (BILP) for the assignment of lanes to one of the two road traversal directions that can be solved optimally via linear programming (LP). The approach is evaluated on three numerical simulation scenarios.