-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
389 lines (306 loc) · 18 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
<!DOCTYPE html>
<html lang="en">
<head>
<!-- meta -->
<meta charset="utf-8">
<meta content="width=device-width, initial-scale=1.0" name="viewport">
<title>Rea Nkhumise | Roboticist</title>
<meta content="rea, nkhumise, robotics, engineer, innovation, technology, customer centric, digital systems, agile methodology, product design, system architecture, systems engineering." name="keywords">
<meta content="Personal website of Rea Nkhumise." name="description">
<meta content="Rea Nkhumise" name="author">
<!-- Google Fonts -->
<link href="https://fonts.googleapis.com/css?family=Poppins:300,300i,400,400i,500,500i,600,600i,700,700i|Playfair+Display:400,400i,700,700i,900,900i" rel="stylesheet">
<!-- Bootstrap CSS File -->
<link href="lib/bootstrap/css/bootstrap.min.css" rel="stylesheet">
<!-- Libraries CSS Files -->
<link href="lib/ionicons/css/ionicons.min.css" rel="stylesheet">
<link href="lib/owlcarousel/assets/owl.carousel.min.css" rel="stylesheet">
<link href="lib/magnific-popup/magnific-popup.css" rel="stylesheet">
<link href="lib/hover/hover.min.css" rel="stylesheet">
<!-- Main Stylesheet File -->
<link href="css/style.css" rel="stylesheet">
<!-- Responsive css -->
<link href="css/responsive.css" rel="stylesheet">
<!-- Favicon -->
<link rel="shortcut icon" href="images/favicon2.png">
</head>
<body>
<!-- start section navbar -->
<nav id="main-nav">
<div class="row">
<div class="container">
<!-- <div class="logo">
<a href="index.html"><img src="images/logo_name3.png" alt="logo"></a>
</div> -->
<div class="responsive"><i data-icon="m" class="ion-navicon-round"></i></div>
<ul class="nav-menu list-unstyled">
<li><a href="#header" class="smoothScroll">Home</a></li>
<li><a href="#about" class="smoothScroll">About</a></li>
<!-- <li><a href="#why-dx" class="smoothScroll">Why DX</a></li>
<li><a href="#service" class="smoothScroll">Services</a></li> -->
<li><a href="blog-grid.html" class="smoothScroll">Blog</a></li>
<!-- <li><a href="#contact" class="smoothScroll">Contact</a></li> -->
</ul>
</div>
</div>
</nav>
<!-- End section navbar -->
<!-- start section header -->
<div id="header" class="home">
<div class="container">
<div class="header-content">
<h1>I'm <span class="typed"></span></h1>
<!-- <p>Digital Transformation Specialist, Engineer, Roboticist.</p> -->
<p>PhD Researcher in Robotic Learning.</p>
<ul class="list-unstyled list-social">
<li><a href="https://za.linkedin.com/in/rea-nkhumise-756a15aa" target="_blank"><i class="ion-social-linkedin"></i></a></li>
<li><a href="https://github.com/nkhumise-rea" target="_blank"><i class="ion-social-github"></i></a></li>
</ul>
</div>
</div>
</div>
<!-- End section header -->
<!-- start section about us -->
<div id="about" class="paddsection">
<div class="container">
<div class="row justify-content-between">
<div class="col-lg-4 ">
<div class="div-img-bg">
<div class="about-img">
<img src="images/me2.jpg" class="img-responsive" alt="me">
</div>
</div>
</div>
<div class="col-lg-7">
<div class="about-descr">
<!-- <p class="p-heading">I'm a digital transformation (DX) specialist & robotics engineer,</p> -->
<p class="p-heading">About me</p>
<!-- <p class="separator" style="text-align: justify">with expertise in 1) innovation-driven consumer-centric
business models & 2) digital systems architecture. I've a Masters degree in Robotics Engineering and
worked in high-tech industries including launching my own startup companies. I've led disruptive business models and
innovative products from scratch to rollout while administering process efficiencies - using systems engineering
and agile methodologies.</p> -->
<!-- <p class="separator" style="text-align: justify">
I'm Rea Nkhumise, a PhD student in the Computer Science department at the University of Sheffield, United Kingdom.
My PhD project is focused on applying reinforcement learning (RL) in control of robotic manipulators in unstructured environments.
My work involves developing model-based RL algorithms that can improve the autonomy of manipulators, by enabling them to
self-learn new tasks and adapt to perturbations.
</p> -->
<!-- <p class="separator" style="text-align: justify">
As a PhD student in the Computer Science department at the University of Sheffield, United Kingdom,
I specialize in the application of reinforcement learning (RL) to
control robotic manipulators in unstructured environments. My research involves developing
model-based RL algorithms that improve the autonomy of manipulators by enabling them to self-learn
new tasks and adapt to perturbations.
</p> -->
<p class="separator" style="text-align: justify">
As a PhD student in the Computer Science department at the University of Sheffield, UK,
I specialize in applying reinforcement learning (RL) to enable human-robot shared autonomy
in assistive robots. My research focuses on utilizing human eye gaze signals to control a
dual-manipulation setup, aiding individuals with motor impairments in dressing tasks.</p>
<p class="p-heading">Education</p>
<ul style="list-style: none; color: #414141; margin: 5px 0px; padding: 0px;">
<li class="separator" style="margin: 5px 0px;">
PhD in Computer Science (2022 - present), University of Sheffield, United Kingdom
</li>
<li class="separator" style="margin: 5px 0px;">
MSc in Robotics (2014 - 2016), Tennessee Technological University, United States
</li>
<li class="separator" style="margin: 5px 0px;">
BSc in Mechanical Engineering (2010 - 2013), University of the Witwatersrand, South Africa
</li>
</ul>
<p class="p-heading">Interests</p>
<ul style="color: #414141;">
<li class="separator" style="margin: 5px 0px;">
Robotics
</li>
<li class="separator" style="margin: 5px 0px;">
Reinforcement Learning
</li>
<li class="separator" style="margin: 5px 0px;">
<!-- Control Systems -->
Shared Autonomy
</li>
<li class="separator" style="margin: 5px 0px;">
Enterpreneurship
</li>
</ul>
<!-- <p class="separator" style="text-align: justify">with expertise in 1) innovation-driven consumer-centric
business models & 2) digital systems architecture. I've a Masters degree in Robotics Engineering and
worked in high-tech industries including launching my own startup companies. I've led disruptive business models and
innovative products from scratch to rollout while administering process efficiencies - using systems engineering
and agile methodologies.</p> -->
<!-- <p class="separator" style="text-align: justify">
Back to your question. Generally human beings are inefficient at controlling robots.
They're slow and prone to mistakes. However, currently robots in the industry can't operate outside strictly controlled environments,
where there are obstacles and ever-changing conditions (pretty much the real world). So, RL in robotics enables robots to adapt
and function well in the real-world without human involvement. This opens opportunities for robots to either work in
human-hazardous environments (like mining, disaster cleanup, radioactive site, toxic-air filled space, etc.) or work
on tedious/sensitive/human-complex tasks with higher efficiency like surgery, space exploration, search missions, etc. For now, mining, construction, and health industries can benefit greatly.
</p> -->
<!-- <p class="separator"><a class="btn-new" href="Rea_CV.pdf" target="_blank">Download CV</a></p> -->
</div>
</div>
</div>
</div>
</div>
<!-- end section about us -->
<!-- start section why-dx -->
<div id="why-dx" class="text-left paddsection">
<div class="container">
<div class="section-title text-center">
<h2>AUTONOMY IN REHABILITATION & ASSISTIVE ROBOTICS</h2>
</div>
<div class="section-title">
<!-- <p class="separator" style="text-align: justify">
Human beings are inefficient at controlling robots, because they're slow and prone to mistakes.
Nevertheless, robots in the industry can't operate autonomously outside strictly controlled environments. These
environments are characterized by obstacles and ever-changing conditions (pretty much the real world).
Reinforcement Learning in robotics enables robots to learn to adapt
and function well in these unstructured environments without human involvement.
This opens opportunities for robots to work 1)
In Human-hazardous environments like mining, disaster aftermath, radioactive sites, toxic-air filled spaces, etc.
2) On complex tasks like surgery, construction, space exploration, search missions, etc.
</p> -->
<p class="separator" style="text-align: justify">
People with severe motor impairments stand to achieve greater independence through the assistance of
autonomous assistive robots capable of perceiving their surroundings, executing movements, and
processing information. Reinforcement Learning (RL) emerges as a promising avenue for fostering
human-robot shared autonomy in assistive robotics. The primary objective of my research is to capture
control signals derived from human eye gaze to command a dual-manipulation setup, assisting
individuals with motor impairments to dress.
</p>
<p class="separator" style="text-align: justify">
Selecting the most effective shared autonomy strategy for assistive robots relies on assessing the
user's abilities, the learning algorithm, and the complexity of the task. While understanding the
user's capabilities and exploring various learning algorithms have garnered significant attention,
quantifying task complexity remains a pressing challenge. Resolving this issue would shed light on
the suitability of the task’s Markov Decision Process (MDP) model and its likelihood to succeed or
fail with current state-of-the-art RL algorithms.</p>
<p class="separator" style="text-align: justify">
Hitherto, my research has focused on addressing a crucial aspect of quantifying task complexity in RL
by developing an exploration metric, given the fundamental role of exploration in RL. This metric can
complement existing approaches and contribute to the development of improved heuristics for gauging task
complexity. By doing so, we aim to enhance the evaluation and identification of appropriate task MDPs,
leading to the selection of suitable RL algorithms for shared autonomy.
</p>
<p class="separator" style="text-align: justify">
In the next phase of my research, I will experiment with the exploration metric and other heuristic
methods to assess task complexity. This will deepen our understanding of optimising shared autonomy
strategies for real-world assistive robotics. By refining our approaches iteratively, based on
experimental results and user feedback, we aim to enhance assistive robotics, improving independence
and quality of life for those with motor impairments.
</p>
<!-- The field of robotics has come a long way in recent years, but there are still limitations to what robots can do.
One major obstacle is the inefficiency of human control, which is slow and prone to mistakes.
In order to overcome this, researchers have turned to Reinforcement Learning, a type of machine learning
that enables robots to learn from their experiences and improve their performance over time.
One area where Reinforcement Learning has shown particular promise is in unstructured environments,
where robots must navigate obstacles and adapt to changing conditions. In these environments, human
control is often not practical, and robots must be able to operate autonomously. By using Reinforcement Learning,
robots can learn to make decisions and take actions that enable them to function well in these complex situations.
</p>
<p class="separator" style="text-align: justify">
This opens up a whole new world of opportunities for robots, particularly in areas where human involvement is dangerous
or impossible. For example, robots can be used to work in hazardous environments like mines, disaster zones,
and radioactive sites, where human workers are at risk. They can also be used for complex tasks like surgery,
construction, space exploration, and search and rescue missions.
By enabling robots to learn and adapt on their own, Reinforcement Learning is helping to revolutionize the
field of robotics and unlock new possibilities for the future. With continued research and development,
we can look forward to a world where robots are an integral part of our daily lives, making our work easier, safer,
and more efficient.
</p> -->
</div>
<!-- <div class="section-title text-center">
<p class="separator"><a class="btn-new" href="5_reasons.html">Learn More</a></p>
</div> -->
</div>
</div>
<!-- end section why-dx -->
<!-- start sectoion contact -->
<!-- <div id="contact" class="paddsection">
<div class="container">
<div class="contact-block1">
<div class="row">
<div class="col-lg-6">
<div class="contact-contact">
<h2 class="mb-30">GET IN TOUCH</h2>
<ul class="contact-details">
<li><span>Cape Town, South Africa</span></li>
<li><span>+27 66-252-8338 </span></li>
<li><span>+27 62-099-2712 (WhatsApp)</span></li>
<li><span>rea@nkhumise.com</span> </li>
</ul>
</div>
</div>
<div class="col-lg-6">
<form action="mail/contact_me.php" method="post" role="form" class="contactForm">
<div class="row">
<div id="sendmessage">Your message has been sent. Thank you!</div>
<div id="errormessage"></div>
<div class="col-lg-6">
<div class="form-group contact-block1">
<input type="text" name="name" class="form-control" id="name" placeholder="Your Name" data-rule="minlen:4" data-msg="Please enter at least 4 chars" />
<div class="validation"></div>
</div>
</div>
<div class="col-lg-6">
<div class="form-group">
<input type="email" class="form-control" name="email" id="email" placeholder="Your Email" data-rule="email" data-msg="Please enter a valid email" />
<div class="validation"></div>
</div>
</div>
<div class="col-lg-12">
<div class="form-group">
<input type="text" class="form-control" name="subject" id="subject" placeholder="Subject" data-rule="minlen:4" data-msg="Please enter at least 8 chars of subject" />
<div class="validation"></div>
</div>
</div>
<div class="col-lg-12">
<div class="form-group">
<textarea class="form-control" name="message" rows="12" data-rule="required" data-msg="Please write something for us" placeholder="Message"></textarea>
<div class="validation"></div>
</div>
</div>
<div class="col-lg-12">
<input type="submit" class="btn btn-defeault btn-send" id="msg" value="Send message">
</div>
</div>
</form>
</div>
</div>
</div>
</div>
</div> -->
<!-- start sectoion contact -->
<!-- start section footer -->
<div id="footer" class="text-center">
<div class="container">
<div class="socials-media text-center">
<ul class="list-unstyled">
<li><a href="https://za.linkedin.com/in/rea-nkhumise-756a15aa" target="_blank"><i class="ion-social-linkedin"></i></a></li>
<li><a href="https://github.com/nkhumise-rea" target="_blank"><i class="ion-social-github"></i></a></li>
</ul>
</div>
<?php include '/footer.php'; ?>
<div class="credits">
Designed by Rea
</div>
</div>
</div>
<!-- End section footer -->
<!-- JavaScript Libraries -->
<script src="lib/jquery/jquery.min.js"></script>
<script src="lib/jquery/jquery-migrate.min.js"></script>
<script src="lib/bootstrap/js/bootstrap.bundle.min.js"></script>
<script src="lib/typed/typed.js"></script>
<script src="lib/owlcarousel/owl.carousel.min.js"></script>
<script src="lib/magnific-popup/magnific-popup.min.js"></script>
<script src="lib/isotope/isotope.pkgd.min.js"></script>
<!-- Contact Form JavaScript File -->
<script src="contactform/contactform.js"></script>
<!-- Template Main Javascript File -->
<script src="js/main.js"></script>
</body>
</html>