Ecological driving on multiphase trajectories and multiobjective optimization for autonomous electric vehicle platoon

Abstract

Autonomous electric vehicles promise to improve traffic safety, increase fuel efficiency and reduce congestion in future intelligent transportation systems. Ecological driving characteristics are first studied to concentrate on energy consumption, the ability to quickly pass its destination, etc. of autonomous electric vehicle plans (AEVPs) to maximize total energy efficiency benefits. To realize this goal, an optimal control model is developed to provide ecological driving suggestions to AEVPs. The Radau pseudospectral method (RPM) is adopted to put the optimal control model into nonlinear programs (NLP), and multiobjective optimization, including safety, economy and fast mobility, is considered, which conditions and constraints such as vehicle dynamics, traffic rules, and energy consumption. To enhance optimal model applicability, two ecological driving procedures are proposed. One procedure is that two-phase trajectory optimization and ecological driving states, such as velocity and acceleration, for the leading vehicle are developed according to RPM characteristics, while the other provides a set of targeted driving states to the following vehicles. The objective of the procedure is to minimize the total energy consumption of AEVPs, while travel comfort and safety are integrated into the schematization by optimization functions. Numerical experiments illustrate significance when ecological driving strategy for AEVPs considers energy consumption characteristics, thereby ensuring total energy consumption efficiency for AEVPs.