CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Integració de mecanismes de control i visió artificial en una aplicació C# per al Drone Engineering Ecosystem

Estudiants que han llegit aquest projecte:

Director/a: VALERO GARCÍA, MIGUEL

Departament: DAC

Títol: Integració de mecanismes de control i visió artificial en una aplicació C# per al Drone Engineering Ecosystem

Data inici oferta: 08-09-2025     Data finalització oferta: 08-04-2026


Estudis d'assignació del projecte:
    GR ENG SIS TELECOMUN
    GR ENG SIST AEROESP
    GR ENG TELEMÀTICA
Tipus: Individual
 
Lloc de realització: EETAC
 
Paraules clau:
Drone Engineering Ecosystem, drons, C#, Windows Forms, MQTT, WebRTC, reconeixement de veu, reconeixement de gestos, detecció d'objectes, visió artificial, YOLOv8, MediaPipe, streaming de vídeo, csDronLink, aplicació global
 
Descripció del contingut i pla d'activitats:
 
Overview (resum en anglès):
This Final Degree Project is framed within the Drone Engineering Ecosystem, an academic platform focused on the development and integration of functionalities related to drone control. The main objective of the project has been to study, implement and integrate different control, communication and monitoring mechanisms into a single application, using C# as the main programming language and keeping a modular and reusable architecture. Specifically, the work addresses MQTT communication, voice recognition, real-time video streaming, gesture recognition, object detection with neural networks and the development of a global integration application.

The methodology followed an incremental development process. First, the structure of the DEE and the most suitable technologies for each functionality were analysed. Then, independent demonstrators were created to validate each part of the system before integrating them into a final Windows Forms application. The main control logic and user interface were implemented in C#, while the computer vision and video transmission processes were developed in Python in order to use specialised libraries such as OpenCV, MediaPipe, YOLOv8 and aiortc. MQTT was used as a lightweight communication mechanism between processes, WebRTC as the basis for low-latency video streaming and csDronLink as the drone control library. The tests were mainly carried out in simulation, which made it possible to validate the control commands safely.

As a result, a global application has been obtained that can connect to the simulated drone, execute basic flight actions, display real-time video inside the interface, recognise voice commands, interpret hand gestures and receive objects detected using YOLOv8. The application also includes an activity log, image capture and short video recording. The work shows that it is possible to combine different technologies within a single C# environment and provides a functional and documented base that can be reused and extended in future projects within the DEE.

© CBLTIC Campus del Baix Llobregat - UPC