Interaction Design and Distraction Detection of Drivers in Automobiles

Abstract

In recent times, usage of electronic devices inside cars while driving has increased due to the introduction of new technologies to keep drivers comfortable and entertained. Systems like satnav ease out the navigation by offering optimised traffic plans at times of complex traffic and road conditions. Though technologies like music, radio and phone facilitate onboard communication and entertainment, they have potential in distracting drivers. The interaction with such technologies while driving takes the attention of drivers away from driving. As distraction of drivers leads to car crashes and fatal accidents, the research community investigated detecting and reducing such distractions. Operating a secondary task while driving is one of the key reasons for driver distraction. It is challenging to detect the inattention blindness of drivers compared to detecting instances of eyes-off road. Recent studies have found that high perceptional load results in increased inattention blindness.

This dissertation investigates methods to reduce driver distraction caused due to operating secondary tasks. It proposes new interactive technologies involving virtual touch and eye gaze tracker to undertake secondary tasks in both head down and head up displays. It also proposes a new machine learning model to estimate cognitive load from ocular parameters and validates with respect to EEG parameters from studies involving professional drivers operating real vehicles. Finally, grounded theory method of qualitative research is used to understand and explore concerns and issues of professional drivers, resolve them, and look for factors contributing to acceptance of the proposed interaction technologies. This dissertation discusses the potential of the proposed methods for applications beyond automotive context. As the proposed systems are tested in simulation and real driving environments, they are considered to be deployed by industries.