Publication Date

2023

Document Type

Dissertation/Thesis

First Advisor

Ferdowsi, Hasan H.

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Electrical Engineering

Abstract

Nowadays, Autonomous Vehicles are in the spotlight as a response to the challenges stemming from population growth, worsening traffic congestion, and rising road accidents. Researchers and companies are concentrating on developing Automated Driving Systems (ADS) to tackle these issues.Of the various components that constitute an autonomous vehicle, perception plays a pivotal role, akin to the function of human eyes and the human brain. Therefore, ensuring precise and reliable perception within Autonomous Vehicles is paramount. This thesis is inspired by the principles of human perception, particularly on the road, and endeavors to replicate this process. Just as a human driver does not scrutinize every detail but swiftly identifies and focuses on the most salient elements, this thesis employs a two-tiered object detection approach. The first tier scans the road, while the second recognizes objects in critical areas. A critical region identifier is designed to pinpoint regions requiring a detailed investigation by second object detection. This identifier leverages a fuzzy logic system and a planned path to identify areas necessitating the attention of second-tier object detection. The results demonstrate an enhancement in accuracy and the efficient allocation of computational resources.

Extent

91 pages

Language

en

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

Download

Share

COinS