-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathindex.html
More file actions
310 lines (287 loc) · 16.3 KB
/
index.html
File metadata and controls
310 lines (287 loc) · 16.3 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
<!DOCTYPE HTML>
<!--
Hyperspace by HTML5 UP
html5up.net | @ajlkn
Free for personal and commercial use under the CCA 3.0 license (html5up.net/license)
-->
<html>
<head>
<title>2020 NYU CUSP Capstone Project</title>
<!-- <style type="text/css">
span{font-size:500%;}
</style>-->
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=no" />
<link rel="stylesheet" href="assets/css/main.css" />
<noscript><link rel="stylesheet" href="assets/css/noscript.css" /></noscript>
</head>
<body class="is-preload">
<!-- Sidebar -->
<section id="sidebar">
<div class="inner">
<nav>
<ul>
<li><a href="#video">Abstract</a></li>
<li><a href="#intro">Introduction</a></li>
<li><a href="#one">Data</a></li>
<li><a href="#two">Methods</a></li>
<li><a href="#team">Team</a></li>
<li><a href="#three">Contact Us</a></li>
</ul>
</nav>
</div>
</section>
<!-- Wrapper -->
<div id="wrapper">
<!-- video -->
<section id="video" class="wrapper style1 fullscreen fade-up">
<div class="inner">
<h2>Applying Multi-Agent RL to SLAM with Graph Pose for Sampled-Data MPC and CPN of Autonomous Drone Swarms</h2>
<h3>2020 NYU CUSP Capstone Project</h3>
<!-- <br />-->
<!-- <div class="fullscreen">-->
<!-- autoplay loop muted-->
<!-- <video autoplay poster="images/video_still.png" controls width="1100" height="600">-->
<video controls width="1000" height="600">
<source src="./assets/drone_swarm.mp4" type="video/mp4">
</video>
<p style="padding-left: 2.0em">An example of 3D simulation. (The code for the 3D simulation is taken and modified from <a href="https://github.com/sgowal/swarmsim">https://github.com/sgowal/swarmsim</a>)</p>
</div>
</section>
<!-- Intro -->
<section id="intro" class="wrapper style1 fullscreen fade-up">
<div class="inner">
<!-- <div class="left-1">-->
<p style="text-align: justify">Developing methods that allow drones to autonomously navigate in different environments has been a topic of extensive research in recent years. One research topic of interest in autonomous drone navigation is to explore the maneuverability and capability of drones to navigate inaccessible environments and situations that might be too risky for human access.
Using a swarm of drones that autonomously navigate a post-catastrophe scenario in order to optimally map the disaster zone, i.e independently and efficiently identify and map the structural damage across a geographic site, has been a problem less explored. Detection and mapping changes across a post-catastrophe site enables a more robust estimation of structural damage.
</p>
<p style="text-align: justify">This project attempted to explore and simulate a reinforcement learning approach to enable drones to perform task assignment and scheduling in order to efficiently maximize coverage for identifying and mapping structural changes within the post-catastrophe environment.
The primary objective of the simulation was to focus on the exploration of ad-hoc decentralized task assignment and scheduling by one or more drone(s) at the edge with minimal connectivity aside from local communication between nearest neighbors.
Other workstreams in the project explored satellite and aerial imagery, seismic structural damage equation models, and generative adversarial networks (GANs) related to the Port-au-Prince 2010 Haiti earthquake site as a use case and attempt to explore methods that might be utilized to identify structural changes from satellite images, using generative synthetic data and estimated fragility equations in order to address uncertainty and ambiguity in the detection of discrepancies in edges related to damage.
</p>
<!-- </div>-->
<!-- <div class="right-1">-->
<!-- <strong>Challenges we met include:</strong>-->
<!-- <br />-->
<!-- <br />-->
<!-- <article>For Multi-Agent Path Planning part,</article>-->
<!-- <li>Optimally cover the environment</li>-->
<!-- <li>Limited bandwidth and low power drones</li>-->
<!-- <li>Multi-agent collaboration with only local connectivity</li>-->
<!-- <li>Adaptability and resilience to environment and operational changes</li>-->
<!-- <li>Damage equations help prioritization during the path planning and GNN</li>-->
<!-- <br />-->
<!-- <article>For Change Detection of Images part,</article>-->
<!-- <li>Efficiently detect the changes of post-catastrophe environment</li>-->
<!-- <li>Low computing power requirement for the drones</li>-->
<!-- <li>Limited paired images data before and after disasters might affect the diversity for generating synthetic data</li>-->
<!-- <li>Low resolution of output images</li>-->
<!-- </div>-->
</div>
</section>
<!-- One -->
<section id="one" class="wrapper style2 spotlights">
<section>
<a href="#" class="image"><img src="images/d001.jpg" alt="" data-position="center center" /></a>
<div class="content">
<div class="inner">
<h2>Satellite/Aerial Images</h2>
<p>Satellite and aerial images are taken before and after the Haiti earthquake (2010/01/13~2010/01/21) by Google maps and GeoEye and a sample is shown on the left. Click Learn More (redirects to an official webpage hosted by Google) to know more about the dataset.</p>
<ul class="actions">
<li><a href="https://maps.googleblog.com/2010/01/haiti-imagery-layer-now-available.html" class="button">Learn more</a></li>
</ul>
</div>
</div>
</section>
<!-- <section>-->
<!-- <a href="#" class="image"><img src="images/d002.png" alt="" data-position="top center" /></a>-->
<!-- <div class="content">-->
<!-- <div class="inner">-->
<!-- <h2>LiDAR data</h2>-->
<!-- <p>LiDAR point cloud data after the Haiti earthquake (2010/01/21~2010/01/27) by OpenTopography</p>-->
<!-- <ul class="actions">-->
<!-- <li><a href="https://opentopography.org/news/post-january-2010-haiti-earthquake-lidar-data-now-available-opentopography" class="button">Learn more</a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </section>-->
<!-- <section>-->
<!-- <a href="#" class="image"><img src="images/d003.jpg" alt="" data-position="25% 25%" /></a>-->
<!-- <div class="content">-->
<!-- <div class="inner">-->
<!-- <h2>Sensor data</h2>-->
<!-- <p>ROS bag-files, radio signal strength and team communication data collected on earthquake damages by TRADR</p>-->
<!-- <ul class="actions">-->
<!-- <li><a href="http://www.tradr-project.eu/resources/datasets/" class="button">Learn more</a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </section>-->
</section>
<!-- Two -->
<section id="two" class="wrapper style3 fade-up">
<div class="inner">
<h2>What we did?</h2>
<p>There are multiple facets to this project which are elucidated in detail below.</p>
<div class="features">
<section>
<!-- <span class="icon solid major fa-code"></span>-->
<p style="white-space:nowrap"><h3><strong style="font-size:200%">C</strong>hange Detection</h3></p>
<p>The change detection algorithm is used to quantify how much the pre-catastrophe environment has changed due to the disaster from images. </p>
<h3><a href="generic.html">Learn more</a></h3>
</section>
<section>
<!-- <span class="icon solid major fa-lock"></span>-->
<p style="white-space:nowrap"><h3><strong style="font-size:200%">G</strong>enerative Adversarial Networks</h3></p>
<p>GAN is used to produce synthetic data which can increase the amount of paired pre and post-catastrophe images to train the agents (i.e drones).</p>
<h3><a href="GAN.html">Learn more</a></h3>
</section>
<section>
<p style="white-space:nowrap"><h3><strong style="font-size:200%">E</strong>nhancing pre-disaster images</h3></p>
<p>The upsampling methods are attempted to explore emulating the post-disaster image resolution, although fully acknowledging the joint hypothesis problem as a limitation to this approach.</p>
<h3><a href="Enhancing.html">Learn more</a></h3>
</section>
<section>
<!-- <span class="icon solid major fa-cog"></span>-->
<p style="white-space:nowrap"><h3><strong style="font-size:200%">S</strong>imulations & Reinforcement Learning</h3></p>
<p>Custom simulation is built to replicate our tasks, i.e. optimally map the entire environment, in order to train and evaluate the learning algorithm.</p>
<h3><a href="Simulations.html">Learn more</a></h3>
</section>
<section>
<!-- <span class="icon solid major fa-desktop"></span>-->
<p style="white-space:nowrap"><h3><strong style="font-size:200%">D</strong>amage Equations</h3></p>
<p>To evaluate the seismic effects on structures, damage equations are used to make a likelihood estimation of damage in a cell/raster.</p>
<h3><a href="damageEquations.html">Learn more</a></h3>
</section>
</div>
<!-- <ul class="actions">-->
<!-- <li><a href="generic.html" class="button">Learn more</a></li>-->
<!-- </ul>-->
</div>
</section>
<!-- team -->
<section id="team" class="wrapper style1 fullscreen fade-up">
<div class="inner">
<div class="left-2">
<h2>Team:</h2>
<!-- <article><a href="https://www.linkedin.com/in/grohith327/">Rohith Gandhi Ganesan</a></article>-->
<strong style="font-size:1.2em">Rohith Gandhi Ganesan</strong>
<ul class="icons">
<li><a href="https://www.linkedin.com/in/grohith327/" class="icon brands fa-linkedin-in" style="font-size:30px;color: #ffffff;"><span class="label">Linkedin</span></a></li>
<li style="padding-left: 0.1em"><a href="https://github.com/grohith327" class="icon brands fa-github" style="font-size:32px;color: #ffffff;"><span class="label">Github</span></a></li>
<li style="padding-left: 0.2em"><a href="https://medium.com/@grohith327" class="icon brands fa-medium" style="font-size:32px;color: #ffffff;"><span class="label">Medium</span></a></li>
</ul>
<strong style="font-size:1.2em">Yue Jin</strong>
<ul class="icons">
<li><a href="https://www.linkedin.com/in/luckyjin7/" class="icon brands fa-linkedin-in" style="font-size:30px;color: #ffffff;"><span class="label">Linkedin</span></a></li>
<li style="padding-left: 0.1em"><a href="https://github.com/luckyjin7" class="icon brands fa-github" style="font-size:32px;color: #ffffff;"><span class="label">Github</span></a></li>
<li style="padding-left: 0.2em"><a href="https://medium.com/@yuejin" class="icon brands fa-medium" style="font-size:32px;color: #ffffff;"><span class="label">Medium</span></a></li>
</ul>
<strong style="font-size:1.2em">Kunru Lu</strong>
<ul class="icons">
<li><a href="https://www.linkedin.com/in/lukunru/" class="icon brands fa-linkedin-in" style="font-size:30px;color: #ffffff;"><span class="label">Linkedin</span></a></li>
<li style="padding-left: 0.1em"><a href="https://github.com/carajumpshigh" class="icon brands fa-github" style="font-size:32px;color: #ffffff;"><span class="label">Github</span></a></li>
<li style="padding-left: 0.2em"><a href="https://medium.com/@kl3743" class="icon brands fa-medium" style="font-size:32px;color: #ffffff;"><span class="label">Medium</span></a></li>
</ul>
<strong style="font-size:1.2em">Elmon Toraman</strong>
<ul class="icons">
<li><a href="https://www.linkedin.com/in/elmon-toraman/" class="icon brands fa-linkedin-in" style="font-size:30px;color: #ffffff;"><span class="label">Linkedin</span></a></li>
<li style="padding-left: 0.1em"><a href="https://github.com/elle1811" class="icon brands fa-github" style="font-size:32px;color: #ffffff;"><span class="label">Github</span></a></li>
<li style="padding-left: 0.2em"><a href="https://medium.com/@et1811" class="icon brands fa-medium" style="font-size:32px;color: #ffffff;"><span class="label">Medium</span></a></li>
</ul>
</div>
<div class="right-2">
<h2>Advisors:</h2>
<strong style="font-size:1.2em">Dr. David K A Mordecai, Professor</strong>
<br />
<strong style="font-size:1.2em">Dr. Giuseppe Loianno, Professor</strong>
<br />
<br />
<ul class="logotypes">
<a href="https://wp.nyu.edu/riskeconlab/"><img src="images/icon002.png" alt=""></a>
<a href="https://cusp.nyu.edu/"><img src="images/icon003.jpg" alt=""></a>
<a href="https://wp.nyu.edu/arpl/"><img src="images/icon004.jpg" alt=""></a>
</ul>
<br />
<article>Support: RiskEcon® Lab for Decision Metrics @ Courant NYU</article>
<article style="padding-left:4.0em">Agile Robotics and Perception Lab (ARPL) @ Tandon NYU</article>
</div>
<br />
<p style="font-size:0.9em">We would like to sincerely thank Joseph Bullock (AI Researcher at UNGP and PhD Researcher at Durham University), Dr. Debra Laefer (New York University) and Dr. Ufuk Hancilar (Bogazici University) for their helpful feedback and suggestions to this project, and thank our project advisors for their guidance throughout the project. Any remaining errors are our own.</p>
</div>
</section>
<!-- Three -->
<section id="three" class="wrapper style1 fade-up">
<div class="inner">
<h2>For more information:</h2>
<br />
<div class="split style1">
<section>
<form method="post" action="#">
<div class="fields">
<div class="field half">
<label for="name">Name</label>
<input type="text" name="name" id="name" />
</div>
<div class="field half">
<label for="email">Email</label>
<input type="text" name="email" id="email" />
</div>
<div class="field">
<label for="message">Message</label>
<textarea name="message" id="message" rows="5"></textarea>
</div>
</div>
<ul class="actions">
<li><a href="" class="button submit">Send Message</a></li>
</ul>
</form>
</section>
<section>
<ul class="contact">
<li>
<h3>Address</h3>
<span>370 Jay Street, 13th Floor<br />
Brooklyn, NY 11201<br />
USA</span>
</li>
<li>
<h3>Email</h3>
<a href="#">yuejin.yz@gmail.com</a>
</li>
<li>
<h3>Phone</h3>
<span>(646)997-0500</span>
</li>
<!-- <li>-->
<!-- <h3>Social</h3>-->
<!-- <ul class="icons">-->
<!-- <li><a href="#" class="icon brands fa-twitter"><span class="label">Twitter</span></a></li>-->
<!-- <li><a href="#" class="icon brands fa-facebook-f"><span class="label">Facebook</span></a></li>-->
<!-- <li><a href="#" class="icon brands fa-github"><span class="label">GitHub</span></a></li>-->
<!-- <li><a href="#" class="icon brands fa-instagram"><span class="label">Instagram</span></a></li>-->
<!-- <li><a href="#" class="icon brands fa-linkedin-in"><span class="label">LinkedIn</span></a></li>-->
<!-- </ul>-->
<!-- </li>-->
</ul>
</section>
</div>
</div>
</section>
</div>
<!-- Footer -->
<footer id="footer" class="wrapper style1-alt">
<div class="inner">
<ul class="menu">
<li>Copyright © 2020 All rights reserved.</li>
</ul>
</div>
</footer>
<!-- Scripts -->
<script src="assets/js/jquery.min.js"></script>
<script src="assets/js/jquery.scrollex.min.js"></script>
<script src="assets/js/jquery.scrolly.min.js"></script>
<script src="assets/js/browser.min.js"></script>
<script src="assets/js/breakpoints.min.js"></script>
<script src="assets/js/util.js"></script>
<script src="assets/js/main.js"></script>
</body>
</html>