Raihan, Lazuardi (2024) Implementasi Algoritma ORB-SLAM3 Pada Sistem Persepsi dan Lokalisasi Autonomous Car Dengan Sensor Fusion LiDAR-Kamera Berbasis Depth Map. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5009201006-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5009201006-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (4MB) | Request a copy |
Abstract
Sistem persepsi dan lokalisasi berperan penting dalam teknologi sistem navigasi autonomous car yang salah satunya memiliki fungsi untuk menjalankan fungsi Simultaneous Localization and Mapping (SLAM). Pada penelitian tugas akhir ini dilakukan implementasi algoritma ORB-SLAM3 yang ditanamkan pada single board computer NVIDIA Jetson Nano B01. ORB-SLAM3 merupakan algoritma SLAM berbasis visual yang memanfaatkan data citra untuk menghasilkan peta lingkungan dalam domain 3D yang mampu bekerja dalam kondisi outdoor. Pada penelitian ini dilakukan perbandingan performansi implementasi algoritma ORB-SLAM3 dengan konfigurasi sensor kamera Intel RealSense D435i dan konfigurasi sensor fusion LiDAR-kamera berbasis depth map. Parameter performansi yang digunakan yaitu absolute trajectory error (ATE), relative pose error (RPE) dan beban komputasi. Evaluasi parameter performasi estimasi lintasan dilakukan dengan menggunakan dataset odometri KITTI. Performansi algoritma ORB-SLAM3 dengan konfigurasi sensor LiDAR-kamera dengan menggunakan dataset KITTI memiliki rentang nilai RMSE ATE 1,3 hingga 8,6 dan nilai RMSE RPE 0,025 hingga 0,071. Implementasi algoritma ORB-SLAM3 pada sistem persepsi dan lokalisasi autonomous car telah berhasil dilakukan menggunakan sensor tunggal kamera Intel RealSense D435i dengan nilai RMSE ATE dan RPE masing-masing sebesar 0,782 dan 0,081. Adapun beban komputasi algoritma ORB-SLAM3 dengan konfigurasi sensor fusion LiDAR-kamera menunjukkan profil komputasi yang lebih ringan dibandingkan dengan konfigurasi hanya sensor kamera.
====================================================================================================================================
The perception and localization system plays an important role in autonomous car navigation technology, one of which is to perform the function of Simultaneous Localization and Mapping (SLAM). In this final project research, the ORB-SLAM3 algorithm is implemented on a single board computer, the NVIDIA Jetson Nano B01. ORB-SLAM3 is a visual-based SLAM algorithm that utilizes image data to generate a 3D map of the environment, capable of operating in outdoor conditions. This research compares the performance of the ORB-SLAM3 algorithm implementation using an Intel RealSense D435i camera sensor configuration and a sensor fusion configuration of LiDAR-camera based on depth maps. The performance parameters used are absolute trajectory error (ATE), relative pose error (RPE), and computational load. The performance evaluation of the trajectory estimation parameters is carried out using the KITTI odometry dataset. The performance of the ORB-SLAM3 algorithm with a LiDAR-camera sensor fusion configuration using the KITTI dataset has an RMSE ATE value range of 1.3 to 8.6 and an RMSE RPE value of 0.025 to 0.071. The implementation of the ORB-SLAM3 algorithm in the autonomous car perception and localization system has been successfully carried out using single Intel RealSense D435i camera sensor with RMSE ATE and RPE values of 0,782 and 0,081. The computational load of the ORB-SLAM3 algorithm with the LiDAR-camera sensor fusion configuration shows that the computational profil are lighter compared to the camera sensor only configuration.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | autonomous car, kamera, LiDAR, ORB-SLAM3, autonomous car, camera, LiDAR, ORB-SLAM3 |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TA Engineering (General). Civil engineering (General) > TA593.35 Instruments, cameras, etc. T Technology > TE Highway engineering. Roads and pavements > TE228.3 Intelligent transportation systems. T Technology > TJ Mechanical engineering and machinery > TJ211 Robotics. |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | Lazuardi Raihan |
| Date Deposited: | 01 Aug 2024 13:11 |
| Last Modified: | 17 Sep 2024 06:29 |
| URI: | http://repository.its.ac.id/id/eprint/109380 |
Actions (login required)
 |
View Item |