-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathnavigation.html
143 lines (119 loc) · 6.42 KB
/
navigation.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
<!DOCTYPE html>
<html lang="en">
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/0.9.0/p5.js"></script><script src="assets/icon.js"></script>
<head>
<!-- Basic Page Needs
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
<meta charset="utf-8">
<title>Learning to Navigate Sidewalks</title>
<meta name="description" content="">
<meta name="author" content="">
<!-- Mobile Specific Metas
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
<meta name="viewport" content="width=device-width, initial-scale=1">
<!-- FONT
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
<link href="https://fonts.googleapis.com/css?family=Raleway:400,300,600" rel="stylesheet" type="text/css">
<!-- CSS
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
<link rel="stylesheet" href="css/normalize.css">
<link rel="stylesheet" href="css/skeleton.css">
<!-- Google Analytics - Navigation Page only -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-K6KM5YNWJW"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'G-K6KM5YNWJW');
</script>
</head>
<body>
<div class="teaser">
<div class="container">
<div class="row">
<div class="one-full column category">
<img class="u-max-full-width" src="assets/img/nav/small.png">
</div>
</div>
</div>
<div class="container">
<div class="twelve columns">
<div class="row" style="margin-top: 4%">
<h1>Learning to Navigate Sidewalks in Outdoor Environments </h1>
<td>
<center>
<br>
<font size="+2">
<a target="_blank" rel="noopener noreferrer" href="https://initmaks.com/" style="text-decoration: none"><nobr>Maks Sorokin</nobr></a><sup>1</sup>   
<a target="_blank" rel="noopener noreferrer" href="https://www.jie-tan.net" style="text-decoration: none"><nobr>Jie Tan</nobr></a><sup>2</sup>   
<a target="_blank" rel="noopener noreferrer" href="https://ckllab.stanford.edu" style="text-decoration: none"><nobr>C. Karen Liu</nobr></a><sup>3</sup>   
<a target="_blank" rel="noopener noreferrer" href="https://www.cc.gatech.edu/~sha9/" style="text-decoration: none"><nobr>Sehoon Ha</nobr></a><sup>1,2</sup>
</font>
<br>
<br>
<nobr> <sup>1</sup> Georgia Institute of Technology</nobr>    <nobr> <sup>2</sup> Robotics at Google </nobr>    <nobr> <sup>3</sup> Stanford University </nobr><br>
<br>
<strong>IEEE Robotics and Automation Letters (RA-L) 2022</strong>
<br>
</center>
</td>
<h3>Abstract</h3>
<p>
Outdoor navigation on sidewalks in urban environments is the key technology behind important human assistive applications,
such as last-mile delivery or neighborhood patrol. This paper aims to develop a quadruped robot that follows a route plan
generated by public map services, while remaining on sidewalks and avoiding collisions with obstacles and pedestrians.
We devise a two-staged learning framework, which first trains a teacher agent in an abstract world with privileged ground-truth information,
and then applies Behavior Cloning to teach the skills to a student agent who only has access to realistic sensors.
The main research effort of this paper focuses on overcoming challenges when deploying the student policy on a quadruped robot in the real world.
We propose methodologies for designing sensing modalities, network architectures, and training procedures to enable zero-shot policy transfer to
unstructured and dynamic real outdoor environments. We evaluate our learning framework on a quadrupedal robot navigating sidewalks
in the city of Atlanta, USA. Using the learned navigation policy and its onboard sensors, the robot is able to walk 3.2 kilometers
with a limited number of human interventions.
</p>
</div>
<div class="row" style="margin-top: 4%">
<div class="one column"> </div>
<div class="three columns"><h5>Paper: <a href="https://arxiv.org/pdf/2109.05603.pdf">[pdf]</a></h5></div>
<div class="three columns"><h5>Preprint: <a href="https://arxiv.org/abs/2109.05603">[arXiv]</a></h5></div>
<div class="three columns"><h5>Journal: <a href="https://ieeexplore.ieee.org/document/9691818">[IEEE]</a></h5></div>
<div class="one column"> </div>
</div>
<div class="row" style="margin-top: 4%">
<h3>Video overview</h3>
<p>
<iframe width="720" height="400" src="https://www.youtube.com/embed/JsAZy3YETwQ" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</p>
<h3>More Videos:</h3>
<h4>Speedrun</h4>
<p>
<iframe class="center" width="720" height="400" src="https://www.youtube.com/embed/G7mJucBemfw" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</p>
<h4>Obstacle Avoidance</h4>
<p>
<iframe class="center" width="720" height="400" src="https://www.youtube.com/embed/snngvoBzNh8" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</p>
</div>
<div>
<h3>Bibtex</h3>
<pre><code>@ARTICLE{sorokin2022learning,
author={Sorokin, Maks and Tan, Jie and Liu, C. Karen and Ha, Sehoon},
journal={IEEE Robotics and Automation Letters},
title={Learning to Navigate Sidewalks in Outdoor Environments},
year={2022},
volume={7},
number={2},
pages={3906-3913},
doi={10.1109/LRA.2022.3145947}
}</code></pre>
</div>
</div>
</div>
<br><br>
<div class="div_line"></div>
<br><br>
<p style="text-align:center;">
<a href="../index.html"><input class="button-primary" type="submit" value="return to HOME page"></a>
</p>
<br><br>
</body>
</html>