Tesi etd-02032025-091229 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
SIRNA, ERIK
URN
etd-02032025-091229
Titolo
Design, Implementation, and Testing of a Brain-Controlled Shared Autonomy Framework for a Home Assistance Robot
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA ROBOTICA E DELL'AUTOMAZIONE
Relatori
relatore Prof. Grioli, Giorgio
relatore Prof.ssa Pallottino, Lucia
correlatore Infantone, Giuseppe
relatore Prof.ssa Pallottino, Lucia
correlatore Infantone, Giuseppe
Parole chiave
- Alter-Ego
- Brain-Computer-Interface (BCI)
- Home4.0
- Point cloud filtering
Data inizio appello
18/02/2025
Consultabilità
Non consultabile
Data di rilascio
18/02/2065
Riassunto
This thesis work is part of the Home 4.0 project, which proposes an innovative solution for patients with various disabilities, enabling them to gain greater autonomy in daily activities through the use of robotic devices controlled by BCI (Brain-Computer Interface).
Throughout the different chapters, several implementation ideas for specific tasks that could assist patients will be explored. The goal is to make a humanoid robot autonomous and capable of recognizing the environment in which the patient lives, distinguishing different rooms, and identifying a given target specified by the user.
The robot will be able to recognize and detect various objects in the room, allowing it to retrieve and deliver them to the user when needed. To develop these tasks on the robot, various techniques have been analyzed, including object detection algorithms, point cloud filtering, object pose estimation relative to an inertial robot, grasping strategies, learning methods for different types of trajectories, and indoor navigation algorithms focused on obstacle avoidance.
To provide the user with the most intuitive experience possible, a graphical interface has been implemented to translate signals sent by BCI devices into high-level commands for the robot.
Throughout the different chapters, several implementation ideas for specific tasks that could assist patients will be explored. The goal is to make a humanoid robot autonomous and capable of recognizing the environment in which the patient lives, distinguishing different rooms, and identifying a given target specified by the user.
The robot will be able to recognize and detect various objects in the room, allowing it to retrieve and deliver them to the user when needed. To develop these tasks on the robot, various techniques have been analyzed, including object detection algorithms, point cloud filtering, object pose estimation relative to an inertial robot, grasping strategies, learning methods for different types of trajectories, and indoor navigation algorithms focused on obstacle avoidance.
To provide the user with the most intuitive experience possible, a graphical interface has been implemented to translate signals sent by BCI devices into high-level commands for the robot.
File
Nome file | Dimensione |
---|---|
La tesi non è consultabile. |