Influence of tasks on the take-over performance during highly automated driving

Status:

completed 09/2022

Aims:

Highly automated driving allows, to a limited extent, the driving task to be handed over to the vehicle. During this time, drivers can turn their attention to other tasks, but must be able to take over control again at any time. In the context of business-related driving, the question arises as to which tasks the freed-up time can be spent on without endangering road safety and causing strain. The aim was to assess characteristics of naturalistic tasks in terms of their suitability as a non-driving related task during business-related driving. The assessment was based on the safety of takeover reactions and the psychological demands.

Activities/Methods:

A meta-analysis and a study in a driving simulator were used to investigate the extent to which non-driving related tasks influence the safety of takeovers from highly automated driving and impose psychological demands. As a non-driving related task, a cognitively demanding route planning task was developed that allows drivers to be distracted for an extended period of time. In addition, test scenarios were developed for the takeover situation that are moderately urgent, unpredictable and critical, and required complex maneuvering decisions.

Results:

Non-driving tasks affect the safety of takeover from highly automated driving, especially when the need for takeover has to be recognized without a takeover request. Based on the study results and the results of other meta-analyses, a classification of task characteristics was proposed that provides initial guidance for risk assessment of workplaces involving highly automated driving.

Last Update: