Energy-Optimized Driving with an Autonomous Vehicle in Urban Environments

A high percentage of deceleration and acceleration in urban traffic is caused by the stops at traffic lights. Reducing these required stops combined with an overall smoother traffic flow can reduce the negative effects on the environment in terms of less emissions, noise reduction, and energy consumption. In this paper a cooperative system developed in the research project KOLINE is introduced aiming at the reduction of delay and emissions by adjusting both the signal control and the vehicles' driving strategy. First, the overall system architecture including the wireless communication between traffic signals and vehicles is described. The paper focuses on the vehicles' reaction approaching a signalized intersection and presents a novel approach to optimize the driving strategy based on the traffic information received. The main feature of our approach is the optimization of both longitudinal and lateral guidance of the vehicle simultaneously. Therefore, the surroundings of the vehicles detected by the vehicles' sensors is combined with more precise information on the local traffic state such as tailback length or traffic volume for each lane and forms the basis for proper driving strategies. To handle the complexity of the optimization problem, a metaheuristic considering the possible actions of the vehicle is used to determine the optimal driving strategy in the current driving situation. The optimized strategy is finally realized by a test vehicle capable of driving autonomously in the dense traffic of urban environments.