Precognitive control: open- and closed-loop steering in a lane-change manoeuvre

Abstract

This paper presents two experiments in which steering a lane-change manoeuvre is analysed as a precognitive control task. The suggestion that the dependence on instantaneous visual feedback is low during such a task was tested by analysing steering performance under conditions both with and without visual occlusion. Steering force served as the main independent variable in experiment I, which was performed in a fixed-base driving simulator. A 1 s visual occlusion period during the initial steering-wheel movement to the left was given every other manoeuvre. The results indicate that steering force reduces steering-wheel amplitude variability, this effect also being reflected in the maximum heading angle variability.

Steering-wheel angle amplitude was the main independent variable in experiment II, which was carried out in the instrumented car. In this experiment the duration of the occlusion period was 3 s, i.e. covering the complete pull-out phase of the manoeuvre. The results show that variability in steering-wheel movement amplitude increases about linearly with the amplitude: standard deviations are about 9% of the amplitude. For both experiments I and II the results correspond quite well with those of an earlier reproduction experiment. Regarding the withdrawal of visual feedback it appears that occlusion affects the mutual tuning of steering amplitude and timing, rather than affecting these quantities separately. The vehicle-motion effects resulting from occlusion are relatively small and it can be concluded therefore that a temporary withdrawal of visual feedback has no dramatic effects in the pre-programmed task considered here.

Keywords:

• Driving
• Steering
• Control model
• Information processing