Clemson University
Virtual Reality Based Simulation Testbed for Evaluation of Autonomous Vehicle Behavior Algorithms
Abstract
dc:description.abstractValidation of Autonomous Vehicle behavior algorithms requires thorough testing against a wide range of test scenarios. It is not financially and practically feasible to conduct these tests entirely in a real world setting. We discuss the design and implementation of a VR based simulation testbed that allows such testing to be conducted virtually, linking a computer-generated environment to the system running the autonomous vehicle's decision making algorithms and operating in real-time. We illustrate the system by further discussing the design and implementation of an application that builds upon the VR simulation testbed to visually evaluate the performance of an Advance Driver Assist System (ADAS), namely Cooperative Adaptive Cruise Control (CACC) controller against an actor using vehicular navigation data from real traffic within a virtual 3D environment of Clemson University's campus. With this application, our goal is to enable the user to achieve spatial awareness and immersion of physically being inside a test car within a realistic traffic scenario in a safe, inexpensive and repeatable manner in Virtual Reality. Finally, we evaluate the performance of our simulator application and conduct a user study to assess its usability.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- School of Computing
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rastogi, Vibhor
- Contributors dc:contributor
-
- Dr. Andrew Robb, Committee Chair
- Dr. Sabarish V. Babu, Committee Co-Chair
- Dr. James Martin
- Dr. Pierluigi Pisu
Subjects
dc:subject × 3Identifiers
dc:identifier.*- Repository record dc:identifier
- https://open.clemson.edu/all_theses/2773