The University of Western Ontario

Published purpose

This research Project aims to build an AI-based driver intentionality predictor that will augment and enhance typical advanced driver assistance systems (ADAS) to improve safety for last mile delivery operation. Since driver gaze directly correlates with driver intentionality and precedes driving action, this project will utilize a driver gaze monitoring system to develop an AI-based tool to predict driver intentionality and action. This will significantly improve the safety associated with last mile delivery in urban environments by broadening the time horizon of ADAS systems. This Project intends to employ real-time driver gaze data to predict driver intentionality and action in order to quantify driver alertness and evaluate the safety of the operation in real-time. This data will be used to evaluate and compare different machine learning methods to develop models of driver attentiveness and alertness by predicting and assessing driver maneuvers. In the long term, having models that predict driver intentionality and actions can provide additional information to an ADAS that can be used to alert the driver or take action to ensure the safety of the vehicle and driver or avoid likely accident scenarios.