Abraham, Alvin (2020) Rancang Bangun Troli Belanja Pengikut Manusia Berbasis Visi Komputer. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
07111640000056-Undergraduate_Thesis.pdf Download (1MB) | Preview |
Abstract
Troli belanja konvensional mempunyai keterbatasan, yaitu memerlukan tenaga manusia untuk mengendalikan gerakannya. Keterbatasan tersebut dapat menimbulkan ketidaknyamanan bagi beberapa pelanggan ketika berbelanja. Oleh karena itu, diperlukan adanya suatu inovasi pada troli belanja sehingga dapat memberikan kenyamanan bagi pelanggan ketika berbelanja. Dalam tugas akhir ini, telah dirancang dan dibangun sebuah troli belanja pengikut manusia berbasis visi komputer. Troli belanja ini dilengkapi dengan sebuah USB webcam serta sebuah komputer papan tunggal Raspberry Pi. USB webcam digunakan untuk mengakuisisi citra dalam bentuk digital sehingga dapat diproses menggunakan Raspberry Pi. Raspberry Pi digunakan untuk menjalankan proses visi komputer dan algoritma kontrol gerak. Proses visi komputer yang digunakan meliputi deteksi warna dan ekstraksi kontur. Nilai dari absis piksel minimum dan maksimum penyusun kontur digunakan untuk mengindikasikan posisi manusia relatif terhadap troli belanja. Kemudian kedua nilai ini diproses menggunakan algoritma kontrol PID untuk mengontrol kecepatan rotasi dari kedua motor yang terpasang pada kedua roda belakang troli belanja. Perbedaan kecepatan rotasi antara roda kanan dan roda kiri digunakan untuk mengendalikan gerakan troli belanja. Hasil yang didapatkan dari pengujian sistem menunjukkan bahwa rancang bangun troli belanja ini dapat bergerak mengikuti manusia secara otomatis sesuai dengan warna aksesoris yang dipakai oleh manusia.
================================================================================================================================
Conventional shopping trolleys have a limitation, which require human energy to control the movement. The limitation can cause inconvenience for certain customers during shopping. Therefore, the innovation on the shopping trolleys is required so that they can provide conveniences for the customers during shopping. In this final project, the human follower shopping trolley based on computer vision has been designed and built. This shopping trolley is equipped with single USB webcam and a single board computer Raspberry Pi. USB webcam is used to acquire images in the digital form so that they can be processed using the Raspberry Pi. Raspberry Pi is used to run the computer vision process and movement control algorithm. The computer vision process used here includes color detection and the contour extraction. The minimum and maximum pixel abscissa value which compose the contour, both are used to indicate the human position relative to the shopping trolley. Then, these values are processed using PID control algorithm to control the rotational velocity of both motors installed on both rear wheels of the shopping trolley. The difference in the rotational velocity between the right wheel and the left wheel is used to control the movement of the shopping trolley. The result obtained from system testing shows that the design of shopping trolley is able to follow human automatically according to the color of accessory worn by the human.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSE 629.892 Abr r-1 2020 |
| Uncontrolled Keywords: | Troli belanja, Visi komputer, Kontrol PID |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TJ Mechanical engineering and machinery > TJ211 Robotics. T Technology > TJ Mechanical engineering and machinery > TJ211.4 Robot motion T Technology > TJ Mechanical engineering and machinery > TJ211.415 Mobile robots |
| Divisions: | Faculty of Electrical Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
| Depositing User: | Alvin Abraham |
| Date Deposited: | 21 Jun 2023 07:43 |
| Last Modified: | 21 Jun 2023 07:43 |
| URI: | http://repository.its.ac.id/id/eprint/72778 |
Actions (login required)
 |
View Item |