-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
409 lines (371 loc) · 30.6 KB
/
index.html
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
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
<!DOCTYPE HTML>
<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Shuyuan Wang</title>
<meta name="author" content="Shuyuan Wang">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="stylesheet.css">
<link rel="icon" type="image/png" href="images/seal_icon.png">
</head>
<body>
<table style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr style="padding:0px">
<td style="padding:0px">
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr style="padding:0px">
<td style="padding:2.5%;width:63%;vertical-align:middle">
<p style="text-align:center">
<name>Shuyuan Wang</name>
</p>
<p>
I am a PhD student at the <a href="ubc.ca">University of British Columbia (UBC)</a> focusing on control theory and reinforcement learning, where I am supervised by Prof. <a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
and Prof. <a href=https://personal.math.ubc.ca/~loew />Philip Loewen</a>. As part of my PhD program, I had the pleasure of collaborating with <a href="https://www.honeywell.com/us/en/company/sustainability">Honeywell Process Solutions</a>.
<p>
Previously, I had an internship with a wonderful group of people at <a href="http://dev3.noahlab.com.hk/">Huawei Noah's Arc Lab</a>.
I completed my M.Sc. in Automation at <a href="http://en.hit.edu.cn/">Harbin Institute of Technology (HIT)</a>, supervised by <a href="https://homepage.hit.edu.cn/lixianzhang">Prof. Lixian Zhang</a> (Fellow of IEEE), and B.Sc. in Electrical Engineering at <a href="http://en.njtu.edu.cn/">Beijing Jiaotong University (BJTU)</a>.
</p>
<p style="text-align:center">
<a href="[email protected]">Email</a>  / 
<a href="https://github.com/josef-w">Github</a> / 
Linkedin
</p>
</td>
<td style="padding:2.5%;width:40%;max-width:40%">
<a href="img/Frank.jpg"><img style="width:100%;max-width:100%" alt="profile photo" src="img/shuyuan.jpg" class="hoverZoomLink"></a>
</td>
</tr>
</tbody>
</table>
<!-- Expericences -->
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Expericences</heading>
</td>
</tr>
</tbody>
</table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:35px;width:25%;vertical-align:middle">
<a href="img/Honeywell.jpg"><img style="width:80%;max-width:100%" alt="profile photo" src="img/Honeywell.jpg" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>Research co-op</papertitle>
<br>
<em><font color=#FF4000><strong>Department of Sustainability,</strong></font></em>
<br>
<em><font color=#FF4000><strong>Honeywell</strong></font></em>
<br>
<p></p>
<p>
Develop innovative machine learning (ML) technologies to assist Honeywell in solving sustainability challenges in industrial processing and exploring robotic control techniques.
Specifically, integrate reinforcement learning (RL) and control theory to improve sample efficiency, generalization, and stability.
</p>
</td>
</tbody>
</table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:35px;width:25%;vertical-align:middle">
<a href="img/huawei.png"><img style="width:70%;max-width:100%" alt="profile photo" src="img/huawei.png" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>Research intern</papertitle>
<br>
<em><font color=#FF4000><strong>Planning and Control Group,</strong></font></em>
<br>
<em><font color=#FF4000><strong>Huawei Noah's Arc Lab</strong></font></em>
<br>
<p></p>
<p>
1. Developed advanced Hybrid A* planner for the company, fixed the flaw of the original version of the planner and completed the U-turn and L-turn scenarios.
<br>
2. Developed distributed planning algorithm for multi vehicle system, balancing efficiency and overall performance.
<br>
3. Received the letter of appreciation from our customer.
</p>
</td>
</tr>
</tbody>
</table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:40px;width:25%;vertical-align:middle">
<a href="img/tsinghua.jpg"><img style="width:65%;max-width:100%" alt="profile photo" src="img/tsinghua.jpg" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>Research assistant</papertitle>
<br>
<em><font color=#FF4000><strong><a href="http://www.idlab-tsinghua.com/thulab/labweb/index.html">iDLab</a></strong></font></em>
<br>
<em><font color=#FF4000><strong>Tsinghua University</strong></font></em>
<br>
<p></p>
<p>
1. Designed a scheme allowing multi agents to accelerate the exploring speed of Reinforcement Learning(RL). The acceleration is demonstrated to be proportional with the amount of agents.
<br>
2. Surveyed on Hierarchical Reinforcement Learning (HRL)
</p>
</td>
</tr>
</tbody>
</table>
<!-- RESEARCH -->
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Selected Research</heading>
<p>
My research interests focus on the intersection of control and reinforcement learning (RL), with applications in industrial processes and robotics.
Specifically, I aim to enable RL with model-based control capabilities from the perspective of algorithmic frameworks, to improve sample efficiency, generalization and so on.
</p>
</td>
</tr>
</tbody>
</table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/neuralPS24.jpg"><img style="width:100%;max-width:100%" alt="profile photo" src="img/neuralPS24.JPG" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
<font color=#FF000><strong>New!</strong></font>Empowering Neural Networks with Control and Planning Abilities
</papertitle>
</a>
<br>
<strong>Shuyuan Wang</strong>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
<br>
Accepted by <em> <font color=#FF8080><strong>NeuralPS 2024 Workshop on Behavioral Machine Learning</strong></font></em>
<br>
Paper (coming soon)
<p></p>
<p>
We present a framework for differentiating through iLQR controllers via implicit differentiation, providing an analytical gradient solution with constant backward cost and accurate gradients for end-to-end learning.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/IROS.gif"><img style="width:100%;max-width:100%" alt="profile photo" src="img/IROS.gif" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>Guiding Reinforcement Learning with Incomplete System Dynamics</papertitle>
</a>
<br>
<strong>Shuyuan Wang</strong>,
<a href="https://jingliang-duan.github.io/">Jingliang Duan</a>,
<a href="https://nplawrence.com/">Nathan Lawrence</a>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>,
<a href="https://homepage.hit.edu.cn/lixianzhang">Lixian Zhang</a>
<br>
Accepted by <em> <font color=#FF8080><strong>IEEE/RSJ IROS 2024</strong></font></em>
<br>
<a href="http://arxiv.org/abs/2410.16821">Paper</a> | <a href="https://www.youtube.com/watch?v=xGNNiuYJh98">Video</a>
<p></p>
<p>
We develop a novel framework that integrates partial model knowledge into RL in a decoupled manner. This approach bridges RL and control frameworks without disrupting the RL structure.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/deephankel.gif"><img style="width:100%;max-width:100%" alt="profile photo" src="img/deephankel.gif" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>Deep Hankel matrices with random elements</papertitle>
</a>
<br>
<a href="https://nplawrence.com/">Nathan Lawrence</a>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<strong>Shuyuan Wang</strong>
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
<br>
<em><font color=#FF8080><strong>L4DC 2024</strong></font></em>
<br>
<a href="https://arxiv.org/abs/2404.15512">Paper</a>
<p></p>
<p>
We study the output prediction accuracy from recursively applying the same persistently exciting input sequence to the Hankel-based model.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/auto_NPL.PNG"><img style="width:100%;max-width:100%" alt="profile photo" src="img/auto_NPL.PNG" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
Stabilizing reinforcement learning control: A modular
framework for optimizing over all stable behavior
</papertitle>
</a>
<br>
<a href="https://nplawrence.com/">Nathan Lawrence</a>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<strong>Shuyuan Wang</strong>
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
<br>
<em><font color=#FF8080><strong>Automatica</strong></font></em>
<br>
<a href="https://arxiv.org/abs/2310.14098">Paper</a>
<p></p>
<p>
We introduce a modular framework for RL-based controller design through a 'model-free' realization of the Y-K parameterization.
Additionally, we establish a data-driven stability criterion and provide a probabilistic analysis of models using Hankel matrix structures.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/itsc.gif"><img style="width:100%;max-width:100%" alt="profile photo" src="img/itsc.gif" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
Velocity Planning for Multi-Vehicle Systems via Distributed Optimization
</papertitle>
</a>
<br>
<strong>Shuyuan Wang</strong>,
Hang Yu,
Shuai Yuan,
<a href="http://www.idlab-tsinghua.com/thulab/labweb/dpeople.html?11">Shengbo Eben Li</a>,
<a href="https://scholar.google.com/citations?user=mf7ivQoAAAAJ&hl=zh-CN">Zepeng Ning</a>
<br>
<em><font color=#FF8080><strong>IEEE ITSC 2023</strong></font></em>
<br>
<a href="https://ieeexplore.ieee.org/document/10422113">Paper</a>| <a href="https://youtu.be/GR6BwFTLErw">Video</a>
<p></p>
<p>
We present a distributed velocity planning strategy for multi-vehicle cooperation within the constraints of pre-defined paths, balancing efficiency and overall performance.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/IFAC.PNG"><img style="width:100%;max-width:100%" alt="profile photo" src="img/IFAC.PNG" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
Reinforcement Learning with Partial Parametric Model Knowledge
</papertitle>
</a>
<br>
<strong>Shuyuan Wang</strong>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<a href="https://nplawrence.com/">Nathan Lawrence</a>,
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
<br>
<em><font color=#FF8080><strong>IFAC World Congress 2023</strong></font></em>
<br>
<a href="https://arxiv.org/abs/2304.13223">Paper</a>
<p></p>
<p>
We propose Partial Knowledge Least Squares Policy Iteration (PLSPI),
which utilizes incomplete information from a linear partial model while retaining the data-driven adaptability of RL towards optimal performance.
</p>
</td>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/ifac_npl.PNG"><img style="width:100%;max-width:100%" alt="profile photo" src="img/ifac_npl.PNG" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
A modular framework for stabilizing deep reinforcement learning control
</papertitle>
</a>
<br>
<a href="https://nplawrence.com/">Nathan Lawrence</a>,
<a href="https://personal.math.ubc.ca/~loew">Philip Loewen</a>,
<strong>Shuyuan Wang</strong>,
<a href="https://www.linkedin.com/in/michael-forbes-b322624/?originalSubdomain=ca">Michael Forbes</a>,
<a href="https://dais.chbe.ubc.ca/">Bhushan Gopaluni</a>
<br>
<em><font color=#FF8080><strong>IFAC World Congress 2023</strong></font></em>
<br>
<a href="https://arxiv.org/abs/2304.03422">Paper</a>
<p></p>
<p>
We propose a method for producing stable operators uses a non-recurrent neural network structure,
and formulate a data-driven realization of the Y-K parameterization essentially removing the prior modeling assumption.
</p>
</td>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/SMC.PNG"><img style="width:100%;max-width:100%" alt="profile photo" src="img/SMC.PNG" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle>
Switching Control of A Mecanum Wheeled Mobile Robot for Vision-Based Tracking with Intermittent Image Losses
</papertitle>
</a>
<br>
<a href="https://homepage.hit.edu.cn/lixianzhang">Lixian Zhang</a>,
<strong>Shuyuan Wang</strong>,
Bo Cai, Tianhe Liu, Yiming Cheng
<br>
<em><font color=#FF8080><strong>IEEE SMC 2019</strong></font></em>
<br>
<a href="https://ieeexplore.ieee.org/document/8914160">Paper</a>
<p></p>
<p>
We propose a switching control scheme to tackle tracking problem for a Mecanum wheeled mobile robot (MWMR) with camera in the presence of intermittent image losses.
</p>
</td>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Teaching</heading>
</td>
</tr>
</tbody>
</table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<td style="padding:20px;width:25%;vertical-align:middle">
<a href="img/ubc.png"><img style="width:100%;max-width:100%" alt="profile photo" src="img/ubc.png" class="hoverZoomLink"></a>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<papertitle> TA for CHBE: Introduction of Process Control [Fall 2022 and 2023]</papertitle>
<p>
Led weekly tutorials and office hours. Assisted with assignment and final exam grading.
</p>
</td>
</tr>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:0px">
<br>
<p style="text-align:right;font-size:small;">
Credits to <a href="https://github.com/jonbarron/jonbarron_website">Jon Barron</a> for the website design.
</p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</table>
</body>
</html>