Influence of a multimodal assistance supporting anticipatory driving on the driving behavior and driver's acceptance

Abstract

This work presents an investigation of a multimodal human-machine interface (HMI) of an anticipatory driver assistance system. The HMI of the system consists of visual indicators displayed in the digital instrument cluster and discrete impulses of an active gas pedal (AGP). The assistance recognizes the upcoming driving situation, informs the driver about its emergence, and suggests a driving action, which execution assures significant reduction in fuel consumption. The experiment is performed in the fixed-base driving simulator. Results show that during assisted drives an average reduction in fuel consumption amounts to 7.5%, in comparison to the drives without assistance. In 50% and 80% of all the cases, participants release the accelerator correspondingly within 1.2 and 2 seconds after receiving the first information. Two thirds of the test subjects grade the concept as "good" and "very good". The participants appreciate AGP discrete feedback especially in rare, unexpected, and potentially critical situations. © 2011 Springer-Verlag Berlin Heidelberg.