Autonomous Driving Fuel Economy (ADFE)

Important results from the project

The purpose of the project is to define and evaluate the impact of automated vehicles on the research area energy efficiency that also extends to noise and emissions. Additionally, the test probe research area was included with the purpose to develop and implement automated driving functionality in vehicles in order to evaluate the performance on real roads with regular customers. The operational design domain (ODD) is restricted to urban highways on the Gothenburg ring road during daylight and good weather/roadway conditions.

Expected long term effects

Adaptive cruise control (ACC) is more energy efficient than manual driving for individual vehicles in naturalistic driving data. The major differences are that ACC generally keeps a longer distance to the lead vehicle, and have lower levels of acceleration and deceleration than manual drivers. A share between 1% and 5% of cautious AVs (similar to ACC) was introduced into the simulation environment. The results show a clear trend of increasing energy consumption due to increased congestion, while the noise levels decrease for the complete traffic system as the share of AVs increase.

Approach and implementation

The impact of automated vehicles on energy consumption, noise, and emissions were investigated using real world measurements and computer simulations. Naturalistic driving data was analyzed by comparing fuel consumption with adaptive cruise control to manual driving. A traffic simulation environment was created in SUMO, an open source microscopic simulation tool. Much effort was put into implementing, verifying and validating the traffic simulation tool consisting of three parts: the traffic environment, vehicle models, and driver models for manual and automated driving.