Intelligent vehicle power control based on effective roadway types and traffic congestion levels

Abstract

This paper presents a new method for defining standard roadway types used in a machine learning approach for intelligent vehicle power management. The machine learning approach uses a roadway specific energy optimization method to train an intelligent power controller(IPC) for a conventional (non-hybrid) vehicle. Experiments are conducted under the simulation program PSAT to evaluate the effectiveness of the proposed standard drive cycles. The intelligent power controller is implemented in a Ford Taurus model provided by PSAT. The experiments on 11 test drive cycles show that the IPC used the proposed standard drive cycles performed better than the IPC used the 11 Sierra standard drive cycles.