Publication Date

5-2-2020

Document Type

Essay

First Advisor

Ferdowsi, Hasan

Degree Name

B.S. (Bachelor of Science)

Department

Department of Electrical Engineering

Abstract

Autonomous vehicle technology has been in development for the past few decades. A system that authorizes self-driving cars to safely and successfully switch lanes must be established to further advance the technology. Lateral control systems have been developed but are not efficient as they are too simplistic and never perform a lane change or too complex and take too much time to compile data. The Autonomous Vehicle Lateral Control System proposes a lateral control system implemented in an open source autonomous vehicle simulator called CARLA. The simulator allows for total lateral autonomy when keeping and changing lanes. The project furthers research at Northern Illinois University and offers a cost-effective solution for fully autonomous lateral control. In the simulation, the vehicle is placed on a track and drives at a safe speed within a lane. The control system takes data from virtual sensors representing an on-board tracking camera, depth camera, and LIDAR (Light Detection and Ranging). It uses the data it collects to execute a variety of algorithms. Once an object is detected in the path of the vehicle, the algorithms decide if the lane next to the vehicle is available and if it is permissible to perform a safe and steady lane change. Upon a successful lane change, the vehicle resumes operation as before. The lateral control system used in CARLA can be applied to a 1:10 scale remote-controlled (RC) car and has the potential to be scaled to full size vehicles.  The outcome of the Autonomous Lateral Control System provided here would impact the market for autonomous vehicles by bringing the cost of autonomy down for full size vehicles, while increasing the level of reliability and efficiency of autonomy.

Extent

49 pages

Language

eng

Publisher

Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type

Text

Share

COinS