Files in this item

FilesDescriptionFormat

application/pdf

application/pdfMARKOWITZ-THESIS-2021.pdf (45MB)
(no description provided)PDF

application/pdf

application/pdfTDA MS thesis (1).pdf (248kB)
(no description provided)PDF

Description

Title:One millimeter for man, one meter for mankind: establishing millimeter wave radar as a ubiquitous sensor in consumer applications
Author(s):Markowitz, Spencer A.
Advisor(s):Do, Minh N
Department / Program:Electrical & Computer Eng
Discipline:Electrical & Computer Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:M.S.
Genre:Thesis
Subject(s):radar, computer vision, background-foreground separation
Abstract:Within the burgeoning and promising autonomous driving community, millimeter-wave frequency-modulated continuous-wave (FMCW) radars are not used to their fullest potential. Classical, hand-designed target detection algorithms are applied in the signal processing chain and the rich contextual information is discarded. This early discarding of information limits what can be applied in algorithms further downstream. This work seeks to bridge this gap by providing the community with a diverse, minimally processed FMCW radar dataset that is not only RGB-D (color and depth) aligned but also synchronized with inertial measurement unit (IMU) measurements in the presence of ego-motion. Moreover, having time-synchronized measurements allow for verification, automated or assisted labelling of the radar data, and opens the door for novel methods of fusing the data from a variety of sensors. This work presents a system that could be built with accessible, off-the-shelf components within a $1000 budget and an accompanying dataset consisting of diverse scenes spanning indoor, urban and highway driving. Finally, we demonstrate the usefulness of this dataset by performing human tracking and background foreground separation in camera data with radar side information. Specifically, this demonstration applies the emerging technique of algorithm unrolling to yield real-time computation, feed-forward inference, and impressive generalization against traditional deep learning methods.
Issue Date:2021-07-23
Type:Thesis
URI:http://hdl.handle.net/2142/113108
Rights Information:Copyright 2021 Spencer Markowitz
Date Available in IDEALS:2022-01-12
Date Deposited:2021-08


This item appears in the following Collection(s)

Item Statistics