A Three-level Control Architecture for Autonomous Vehicle Driving in a Dynamic and Uncertain Traffic Environment

This paper proposes a novel control architecture for autonomous vehicle driving in a dynamic and uncertain tra c environment. The architecture is composed of three levels: (1) the operational level deals with a reactive control of a vehicle in a short time cycle; (2) the tactical level decides proper maneuvers based on prediction of future states using probabilistic trafc models; (3) the meta-tactical level, which is the feature of the architecture, timely activates an appropriate tactical-level planning procedure according to both the history of maneuvers and the current tra c condition. A utility-based maneuver evaluation method is also described. The proposed architecture was tested on a highway driving simulator in various tra c scenarios; simulation results show the feasibility of the architecture.