DEVELOPMENT OF AN OPEN-SOURCE DRIVING SIMULATOR TO EVALUATE DRIVER BEHAVIOR IN AUTONOMOUS ENVIRONMENTS

Open Access
Author:
Arora, Prashant
Graduate Program:
Mechanical Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 01, 2016
Committee Members:
  • Sean N Brennan, Thesis Advisor
  • Matthew B Parkinson, Committee Member
  • Karen Ann Thole, Committee Member
Keywords:
  • Driving Simulator
  • Autonomous Environments
  • Scenario Authoring
Abstract:
The objective of this thesis is to develop an open-source highway driving simulator setup that allows different levels of autonomy in traffic, exposure to different traffic situations, and enables different simulated driver responses in terms of longitudinal and lateral vehicle control. This thesis is particularly motivated by the recent FHWA interest in the study of human factors while driving in autonomous environments on highways. Technological advancements like Adaptive Cruise Control (ACC) and Cooperative Adaptive Cruise Control (CACC) aim to reduce traffic congestion by providing different levels of autonomy to the driver. However, the driver’s acceptance of these technologies has not been quantified yet and needs further investigation. Driving simulators have gained more attention in the past few years being one of the only tools available to safely test human responses to advanced driving automation or driving-assist situations. Recent advancements in driving simulation technology allow scenario authoring to create dynamic situations, allow multiple simulations to be connected to each other, and provide the ability to connect hardware to simulations to enable hardware-in-the-loop driving evaluations using simulators. Using this modern technology, mixed traffic environments are modeled to enable the assessment of driver behavior in autonomous environments and to understand the need and type of information to be conveyed. The virtual platform is designed to be visually and geometrically realistic using AASHTO highway design guidelines. Traffic simulations are scripted in the scenarios allowing mixed autonomous environment with manual, ACC and CACC vehicles.