Please use this identifier to cite or link to this item:
http://univ-bejaia.dz/dspace/123456789/20283
Title: | Développement et réalisation d'un système de suivi de cible. |
Authors: | Dahli, Fares Bouhamdani, Messaoud Mezzah, Samia ; promotrice |
Keywords: | Deep learning : System : Planification |
Issue Date: | 2022 |
Publisher: | Université Abderrahmane Mira- Bejaia |
Abstract: | n this project, the main objective was to make an object tracking system. The goal was finally achieved, not exactly as we wanted, but finally achieved and the system is operational anyway. First, we had to learn about those new technologies and understand its fundamentals. Then start the development et the tests. In the beginning, we wanted to develop a human tracking system, which can be used in surveillance and many other applications. But given our limitations in terms of available hardware, we had to change our project and adapt several times. The size of the robot forced us to abandon the "human" part of the project and replace the target with another one with an adapted size. We also planned to use some techniques (deep learning) in our models, but again, the hardware did not allow us. This project allowed us to improve our technical skills by discovering the field of artificial intelligence and its subfields, the python programming language, the Raspberry Pi development platform and the i2c communication bus. It also allowed us to learn to work in a group, to distribute tasks well and to be complementary. There are several improvements that can be added to this robot. First by solving the already mentioned problems. Then, by adding for example a lidar sensor which will allow the robot to do some mapping and planification, which will make it even more autonomous. |
Description: | Option : Automatique et Informatique industrielle |
URI: | http://univ-bejaia.dz/dspace/123456789/20283 |
Appears in Collections: | Mémoires de Master |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.