Objective metrics of comfort: Developing a driving style for highly automated vehicles

This paper addresses the issue of enabling a comfortable highly automated driving style. Two studies have been conducted to identify metrics which can be used to parametrize a high-quality automated driving style for automobiles with regard to safety, functionality and comfort. The studies were set either in an urban and rural or a highway environment. Participants (N = 12 per study) manually drove a round course assuming either an everyday, a comfortable, or a dynamic driving style in randomized order. The obtained results emphasize the importance of maneuver-based analysis. Namely, a variety of maneuver-specific metrics, such as acceleration, jerk, quickness and headway distance in seconds, were identified, which are prerequisites to differentiate between the three driving styles. These metrics seem to be the essential components for the development of comfortable highly automated driving. (C) 2016 Elsevier Ltd. All rights reserved.