Promoting excellence in mobility engineering

  1. FISITA Store
  2. Technical Papers

Influence of Automated Brake Application on Take-Over Situations in Highly Automated Driving Scenarios
FISITA2014/F2014-ACD-029

Authors

Gold, Christian*; Bengler, Klaus - Technische Universität München
Lorenz, Lutz - BMW Group Research and Technology

Abstract

A sufficient take-over time is essential to ensure the controllability of highly automated vehicles. Because this time budget is limited by factors like the sensor range, this paper verifies in a driving simulator experiment whether an automated brake application which lengthens the time budget by reducing vehicles’ speed is suitable for improving the controllability of highly automated vehicles.

In the dynamic driving simulator of the BMW Group Research and Technology, which is equipped with a full vehicle mock-up, 48 subjects experienced four different take-over scenarios. The subjects were divided into three groups of 16 subjects each - one group with a brake application of 5.0 m/s², one with 3.5 m/s² and a third without automated braking, representing the baseline condition. The time budget for regaining control and avoiding the critical situation was five seconds in each group, plus the extension obtained through the automated brake application (6.09s / 5.50s). To ensure that subjects were unable to identify the system limit before the take-over request occurred, the limit emerged suddenly and the subjects were visually and motorically distracted by the ISO-standardized SuRT task. Vehicle dynamics, gaze behaviour and video data were recorded. Furthermore, the subjects were interviewed after each situation and completed a questionnaire to assess their subjective evaluation.

No statistical evidence was found that automated brake application enhances take-over quality or shortens take-over time. Nevertheless, the study of outcomes like time to collision, lateral accelerations and accident probability indicate that a positive influence may exist, and that experimental restrictions merely prevented its detection.

KEYWORDS - Self-driving automation, take-over quality, reaction times, high automation, automated braking

Add to basket