Reducing Driver Distraction through Software

Advanced Driver Assistance (ADA) systems currently operate within vehicles, offering drivers assistance to either avoid hazardous situations, or information to make travelling easier. However, these devices have the potential to contribute to driver distraction as they require a certain level of driver attention in order to provide a benefit, taking cognitive, visual, auditory, and manual resources away from the main driving task. As these systems become more prolific in the market, the potential number of devices that can operate within a vehicle at any one time increases. Therefore, this paper presents a new in-vehicle architecture to unobtrusively reduce ADA system related distraction. Our approach consists of sensing and assessing the current driving context. The context is gathered by an in-vehicle system which senses and articulates relevant information about the environment, driver and vehicle. We use Bayesian Networks in our architecture to assess driving distraction and to identify an optimal way to interact with the driver. This paper will address the assessment of driver distraction based on contextual information in relation to the vehicle, the environment, and the driver.