Perancangan dan Pembuatan Robot Pelacak Objek Otonom Dengan Teknik Penyepadanan Fitur

Jonathan, Jesse (2023) Perancangan dan Pembuatan Robot Pelacak Objek Otonom Dengan Teknik Penyepadanan Fitur. Bachelor thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (70kB) | Preview
Preview	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (739kB) | Preview
Preview	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB) | Preview
Preview	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (823kB) | Preview
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (546kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (0B)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)
	Text (Chapter6) Chapter6.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (338kB)
Preview	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (334kB) | Preview
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (7MB)

Abstract

Robot telah banyak digunakan apalagi di dunia industri. Robot dapat membuat pekerjaan lebih mudah walaupun masih membutuhkan bantuan manusia, tetapi ada robot autonomous yang dapat menyesuaikan diri tanpa bantuan. Penelitian ini merupakan perancangan robot autonomous tracking objek menggunakan teknik feature matching. Robot dirancang untuk mendeteksi dan melakukan tracking objek. Robot diprogram untuk melihat sekitar melalui kamera laptop dan mencari persamaan feature data dengan feature objek. Saat persamaannya dikatakan tepat maka robot akan mengikuti objek. Semakin objek terlihat maka semakin mudah robot mendeteksi objek berupa logo. Pada penelitian ini ditetapkan akurasi lebih dari 75 % untuk dapat dikatakan robot dapat mendeteksi dengan optimal. Pada penelitian ini semua percobaan akan diuji untuk mencari batas kemampuan robot. Dalam percobaan ditemukan robot memiliki akurasi di atas 75% untuk mendeteksi logo sampai jarak 150 cm. Percobaan logo terhadap rotasi sumbu Z juga diuji dan ditemukan bahwa semakin besar rotasi dan jarak objek dari robot, akurasi akan semakin mengecil. Kemiringan tidak berpengaruh besar karena akurasi masih di atas 96%. Selain itu intensitas juga mempengaruhi akurasi robot. FPS (Frame Per Second) yang didapatkan robot sangat kecil yaitu di rentang 5 hingga 7 FPS sehingga robot tidak bisa mengikuti objek yang bergerak dengan cepat. Semua percobaan tersebut dipengaruhi seberapa jelas objek yang dilihat oleh robot. / Robots have been widely used, especially in the industrial world. Robots can make work easier, although they still need human assistance, but there are autonomous robots that can adapt without help. This research covers the design of an autonomous object tracking robot using feature matching techniques. The robot is designed to detect and track object. The robot is programmed to look around through a laptop camera and find a match between the feature data and the object's feature. When the match is deemed accurate, the robot will follow the object. The clearer the object is, the easier it is for the robot to detect the object, such as a logo. In this research, an accuracy of more than 75% is set as the standard for the robot to detect optimally. In this research, all tests were carried out to find the limits of the robot's capabilities. In the tests, it was found that the robot had an accuracy of over 75% for detecting logos up to a distance of 150 cm. Tests of logo rotation around the Z axis were also carried out and it was found that the larger the rotation and the farther the object is from the robot, the accuracy decreases. Inclination does not have a significant effect as the accuracy is still above 96%. In addition, intensity also affects the accuracy of the robot. The FPS (Frames Per Second) obtained by the robot is very small, ranging from 5 to 7 FPS, so the robot cannot follow fast-moving objects. All of these experiments are affected by how clear the object is seen by the robot.

Actions (login required)

View Item