Tesi etd-09062022-092828
Link copiato negli appunti
Tipo di tesi
Dottorato
Autore
NOCENTINI, OLIVIA
URN
etd-09062022-092828
Titolo
Study and development of AI-based approaches for cloth manipulation capabilities
Settore scientifico disciplinare
ING-IND/34
Corso di studi
Istituto di Biorobotica - PHD IN BIOROBOTICA
Commissione
relatore Prof. CAVALLO, FILIPPO
Parole chiave
- assistive manipulator
- cloth manipulation
- neural networks
Data inizio appello
26/01/2023;
Disponibilità
parziale
Riassunto analitico
According to the World Health Organization forecasts, there are already more than one
billion people over the age of 60, with that figure predicted to climb to 1.4 billion by
2030. As a result, there is a growth in the need for caregivers, which may become
unsustainable for the future society. In this scenario, as the world population ages, the
demand for automated help grows. Service robotics is one area of robotics where robots
have shown significant promise in working closely with people. Hospitals, household
settings, and elderly homes will need intelligent robotic agents in use that perform daily
activities. Cloth manipulation is one such daily activity and represents a challenging
area for a robot.
This thesis details two main aspects: the cloth image classification and the identification
of grasping points for clothing manipulation. Taking into account the cloth classification
task, we first enhance state-of-the-art convolutions neural networks, e.g., LeNET, by
adding additional image processing features to the network structure. Finally, to further
improve the accuracy, we investigated the implementation of multiple convolutional
neural networks (MCNN). The proposed networks are trained and tested on the Fashion-
MNIST dataset.
The second research goal of this thesis focused on finding the grasping points of the
highest wrinkle (from a later point of view) of a folded hospital gown. The wrinkle is
detected using the Generative Grasping Convolutional Neural Network (GGCNN), while
the approach to the cloth by a manipulator is obtained by designing a visual servoing
algorithm that considers the input of the GGCNN.
In conclusion, the results described in this thesis tend to study by deep some AI-based
approaches for cloth manipulation capabilities; in particular, we concentrated on study-
ing the cloth image classification with neural networks and the Fashion-MNIST dataset.
Moreover, we analysed how to identify the first wrinkle of a cloth by combining the
visual servoing approach with a neural network.
billion people over the age of 60, with that figure predicted to climb to 1.4 billion by
2030. As a result, there is a growth in the need for caregivers, which may become
unsustainable for the future society. In this scenario, as the world population ages, the
demand for automated help grows. Service robotics is one area of robotics where robots
have shown significant promise in working closely with people. Hospitals, household
settings, and elderly homes will need intelligent robotic agents in use that perform daily
activities. Cloth manipulation is one such daily activity and represents a challenging
area for a robot.
This thesis details two main aspects: the cloth image classification and the identification
of grasping points for clothing manipulation. Taking into account the cloth classification
task, we first enhance state-of-the-art convolutions neural networks, e.g., LeNET, by
adding additional image processing features to the network structure. Finally, to further
improve the accuracy, we investigated the implementation of multiple convolutional
neural networks (MCNN). The proposed networks are trained and tested on the Fashion-
MNIST dataset.
The second research goal of this thesis focused on finding the grasping points of the
highest wrinkle (from a later point of view) of a folded hospital gown. The wrinkle is
detected using the Generative Grasping Convolutional Neural Network (GGCNN), while
the approach to the cloth by a manipulator is obtained by designing a visual servoing
algorithm that considers the input of the GGCNN.
In conclusion, the results described in this thesis tend to study by deep some AI-based
approaches for cloth manipulation capabilities; in particular, we concentrated on study-
ing the cloth image classification with neural networks and the Fashion-MNIST dataset.
Moreover, we analysed how to identify the first wrinkle of a cloth by combining the
visual servoing approach with a neural network.
File
| Nome file | Dimensione |
|---|---|
Ci sono 1 file riservati su richiesta dell'autore. |
|