2015/05/21 by Katherine Driggs-Campbell, Driggs-Campbell, Katherine, Růžena Bajcsy +1 · 1 citation
Engineering · Psychology · #Autonomous Vehicle Technology and Safety #FOS: Electrical engineering #Human-Automation Interaction and Safety #Systems and Control (eess.SY) #Traffic and Road Safety #Traffic control and management #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1505.05921
openalex publication_date 2015/05/21 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28
In light of growing attention of intelligent vehicle systems, we propose\ndeveloping a driver model that uses a hybrid system formulation to capture the\nintent of the driver. This model hopes to capture human driving behavior in a\nway that can be utilized by semi- and fully autonomous systems in heterogeneous\nenvironments. We consider a discrete set of high level goals or intent modes,\nthat is designed to encompass the decision making process of the human. A\ndriver model is derived using a dataset of lane changes collected in a\nrealistic driving simulator, in which the driver actively labels data to give\nus insight into her intent. By building the labeled dataset, we are able to\nutilize classification tools to build the driver model using features of based\non her perception of the environment, and achieve high accuracy in identifying\ndriver intent. Multiple algorithms are presented and compared on the dataset,\nand a comparison of the varying behaviors between drivers is drawn. Using this\nmodeling methodology, we present a model that can be used to assess driver\nbehaviors and to develop human-inspired safety metrics that can be utilized in\nintelligent vehicular systems.\n