Stochastic Optimum Energy Management for Advanced Transportation Network

Abstract Smart and optimal energy consumption in electric vehicles has high potential to improve the limited cruising range on a single battery charge. The proposed concept is a semi-autonomous ecological advanced driver assistance system which predictively plans for a safe and energy-efficient cruising velocity profile autonomously for battery electric vehicles. However, high entropy in transportation network leads to a challenging task to derive a computationally efficient and tractable model to predict the traffic flow. Stochastic optimal control has been developed to systematically find an optimal decision with the aim of performance improvement. However, most of the developed methods are not real-time algorithms. Moreover, they are mainly risk-neutral for safety-critical systems. This paper investigates on the real-time risk-sensitive nonlinear optimal control design subject to safety and ecological constraints. This system improves the efficiency of the transportation network at the microscopic level. Obtained results demonstrate the effectiveness of the proposed method in terms of states regulation and constraints satisfaction.