| Popis: |
When designing experimental studies in the driving domain, the question always rises which driving scenarios to include. This becomes more important as research starts to focus on environment adaptive HMI in automated vehicles. A feasible approach would be to include scenarios of varying levels of complexity. However, to the best of our knowledge, there has not been proper definition of factors that determine the complexity of a complete driving situation. Previous research has already identified infrastructural factors that may increase the complexity of a driving situation but does not include the influence of traffic. This study not only examined the overall environment characteristics that may add to the complexity of a driving situation, but also designed and validated a set of driving scenarios of varying complexities. Five basic road types were adapted with a variety of both infrastructure- and traffic elements from literature to create different levels of complexity. Infrastructural changes to the base scenarios included elements such as obstructions, sharp turns, unsignalised intersections and obstructed view. Traffic changes to the base scenarios included among others traffic density, variety in road user types and (un)predictability of other road users. The developed scenarios were then presented in a survey, with 143 respondents rating each individual scenario on their perceived overall-, infrastructure- and traffic complexity. While both infrastructure- and traffic changes added to the overall complexity of scenarios, the traffic elements showed to have a larger contribution. This study presents: 1) a list of infrastructure- and traffic characteristics that may in- or decrease the overall complexity of a driving situation, and 2) a validated selection of scenarios of varying complexities. The development and validation of scenarios with varying complexity is particularly relevant for studying and developing adaptive driver support systems in partially automated cars. |
