The Effects of an Eco-Driving Assistance System for a City Bus on Driving Style

Abstract

Fuel consumption of a city bus is greatly influenced by driving style, and driving assistance systems (DAS) is a key technology to encourage driving style improvement. In the research reported here, we construct a real-time DAS for a city bus based on a decision tree algorithm. The DAS aims to reduce fuel consumption by encouraging three behaviors: reduced rates of acceleration, reduced depth of acceleration pedal and early upshifting through the gears. Using the proposed DAS, we establish a driver experiment platform with a driving simulator. Then, we design an experiment for 4 non-professional drivers with various driving experience to determine the effects of DAS on their driving style. The driving simulator experiment results show that all the drivers exhibited an observable change in acceleration pedal depth distribution, acceleration pedal positive gradient distribution, and shift timing. These findings indicated that through the real-time guidance on the manipulation of the accelerator pedal position and gear, the drivers could obey the devices from DAS to a certain extent, so the eco-driving device can help drivers change their driving style.