Publication Date

Fall 2022

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Engineering

Advisor

Jerry Z. Gao

Subject Areas

Computer engineering

Abstract

Autonomous vehicles (AVs) offer advantages in many different industries which isleading to an increase in their adoption. Testing and conducting quality assurance on the complex systems are crucial in ensuring safety and reliability. The main challenges with AV testing include modeling diverse scenarios which an AV can encounter in the real world. Compared to real-world testing, simulators provide an efficient and versatile alternative for AV testing. SVL Simulator is one such simulator that integrates with Baidu’s Apollo AV platform. In this paper, a model-based testing framework utilizing semantic trees and 3D test tables is used to model driving scenarios into individual test cases. The framework is used to generate well-defined driving scenarios in SVL Simulator and evaluate the behavior of the Apollo AV platform.

Share

COinS