-
Notifications
You must be signed in to change notification settings - Fork 0
/
kantor-lab.html
50 lines (44 loc) · 2.85 KB
/
kantor-lab.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
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta http-equiv="X-UA-Compatible" content="ie=edge">
<title>Kantor Lab Project</title>
<link rel="stylesheet" href="styles.css">
</head>
<body>
<header>
<h1>Graduate Research Assistant - Kantor Lab</h1>
</header>
<nav>
<a href="index.html">Home</a>
<a href="mailto:[email protected]">Contact</a>
<a href="https://github.com/Samaksh0278">Github</a>
<a href="hobbies.html">Hobbies</a>
</nav>
<section id="KantorLabExperience">
<h2>Graduate Research Assistant</h2>
<p><strong>Institution:</strong> Robotics Institute, School of Computer Science, CMU</p>
<p><strong>Lab:</strong> Kantor Lab</p>
<p><strong>Tools Used:</strong> ROS1, ROS2, Docker, Pytorch, C++, Arduino</p>
<p><strong>Project Timeline:</strong> May 2024 - Aug 2024</p>
<p>
As a Graduate Research Assistant at the Robotics Institute, I developed a system for <strong>3D point cloud modeling of occluded fruits</strong> in orchards using advanced algorithms and robotic techniques.
</p>
<ul>
<li><strong>3D Point Cloud Modeling:</strong> Developed a 3D point cloud system using the <strong>RAFT stereo algorithm</strong> to model occluded fruits in orchard environments, enabling accurate detection even in cluttered scenes.</li>
<li><strong>Segmentation and Tracking:</strong> Implemented <strong>YOLOv8</strong> for fruit segmentation and enhanced the tracking process with an augmented <strong>DeepSORT algorithm</strong> to maintain tracking consistency across frames.</li>
<li><strong>Leafblower Actuation System:</strong> Designed and implemented an innovative <strong>leafblower actuation system</strong> controlled via inverse kinematics on servo motors, which cleared obstructions to reveal occluded fruits.</li>
<li><strong>ROS-Based Docker Deployment:</strong> Deployed the entire solution in a <strong>ROS-based Docker container</strong>, allowing for seamless integration with field robotics systems for real-time orchard monitoring.</li>
<li><strong>Improvement in Detection Accuracy:</strong> Achieved up to <strong>35% improvement</strong> in occluded fruit detection accuracy compared to baseline models, significantly enhancing the system's performance in practical applications.</li>
</ul>
<p>
This project highlights my skills in <strong>computer vision</strong>, <strong>robotics</strong>, and <strong>system integration</strong> for agricultural automation, contributing to advancements in smart farming technologies.
</p>
</section>
<footer>
<p>© 2024 Samaksh Judson. All Rights Reserved.</p>
</footer>
</body>
</html>