Files in this item

FilesDescriptionFormat

application/pdf

application/pdfFA20-ECE499-Thesis-Chang, Yu-Ju.pdf (1MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Greenhouse gas emissions prediction of an IoT taxi fleet
Author(s):Chang, Yu-Ju
Contributor(s):Caesar, Matthew
Subject(s):Artificial Neural Network (ANN)
Greenhouse Gas (GHG) Emissions
Internet of Things (IoT)
Vehicles
Transportation
Education
Project-Based Learning
Abstract:With the advancement of technology, more and more things, like vehicles, are connected through the internet. Useful data could be collected from the sensors on the vehicles and transmitted via the internet. The data could then be analyzed. How to make use of the data is an issue. This study focused on developing a greenhouse gas (GHG) emissions prediction model for an autonomous IoT taxi fleet. This study involved extracting data from an autonomous IoT taxi fleet, processing the data, and producing a machine learning model for the data. Traffic simulations were performed to generate data in this study because not many cars on the road today are connected to the internet. The data was processed in Python, and a machine learning model was produced using Tensorflow. The results showed that a vehicle’s sensor data like vehicle position, speed, waiting time, acceleration, fuel and noise could be used to predict a vehicle’s GHG emissions. Another focus of this study was to make the whole procedure an IoT lab that educates future students about IoT and real-world problem solving. Instructive tutorials were developed. The machine learning program was made into a Jupyter Notebook document with educational texts before each code block. Students were recruited to test the educational effectiveness of the lab.
Issue Date:2020-12
Genre:Other
Type:Text
Language:English
URI:http://hdl.handle.net/2142/109142
Date Available in IDEALS:2021-01-04


This item appears in the following Collection(s)

Item Statistics