Mixed-method design for user behavior evaluation of automated driver assistance systems: An automotive industry case

Abstract

Automotive systems are changing rapidly from purely mechanical to smart, programmable assistants. These systems react and respond to the driving environment and communicate with other subsystems for better driver support and safety. However, instead of supporting, the complexity of such systems can result in a stressful experience for the driver, adding to the workload. Hence, a poorly designed system, from a usability and user experience perspective, can lead to reduced usage or even ignorance of the provided functionalities, especially concerning Adaptive Driver Assistance Systems. In this paper, the authors propose a combined design approach for user behavior evaluation of such systems. At the core of the design is a mixed methods approach, where objective data, which is automatically collected in vehicles, is augmented with subjective data, which is gathered through in-depth interviews with end-users. The aim of the proposed methodology design is to improve current practices on user behavior evaluation, achieve a deeper understanding of driver's behavior, and improve the validity and rigor of the named results.