A Prospective Model for Predicting Driver Workload, and Safety Related Distraction, using Heterogeneous Data and Machine Learning

There is an increasing amount of potentially distracting information being made available to drivers. However, applications (e.g., for navigation, work, entertainment, etc.) need to be aware of driver workload so that they don’t create unsafe distraction. This work is important because distracted driving is a major risk factor for road accidents and the threat of distraction is slowing down the implementation of entertainment and productivity applications in vehicles. The goal of this research will be to develop a system that can predict driver workload. This type of prediction is needed so that future AI systems can schedule the availability of technology interactions (e.g., reading out text messages) so that they do not distract the driver at a dangerous time. This research will differ from previous research in that it will focus on predicting future, rather than current workload, to support planning of travel paths and scheduled activities.

Faculty Supervisor:

Mark Chignell

Student:

Partner:

Huawei Technologies Canada Co., Ltd.

Discipline:

Engineering

Sector:

Information and cultural industries; Professional, scientific and technical services

University:

University of Toronto

Program: