Research

You can download a pdf that briefly describes my main research activities between 2005 and 2015, here. Similar information is reported hereinafter.

[2014--2016]
I was holding a Marie Sklodowska-Curie Grant (IEF) for the realization of the Project LIMOMAN [PIEF-GA-2013-628315]. The project focused on three main aspects of human motor control that can be combined to improve the ability of current robots in dexterous manipulation: internal models, developmental learning and multisensory integration.
I proposed the concept of "hierarchical, probabilistic and contextual" internal models. A hierarchical organization dealing with different levels of complexity will allow the system to represent from simple grasping to finer manipulation, and to be able to respond properly to the environment depending on the available sensory information. These different levels of complexity are acquired incrementally through motor experience in a developmental way. Different sensory modalities were combined in a probabilistic (Bayesian) fashion, depending on their reliability and the associated computational cost. Real world scenarios were considered, using the iCub humanoid robot.



[2013--2016]
The goal of the EU Project Poeticon++ was to integrate natural language and visual action/object recognition tools with motor skills and learning abilities, in the iCub humanoid robot. My main task was the implementation of learning strategies that allow the co-development of motor and perceptual skills, both through autonomous exploration and through interaction with human partners (exploiting visual and speech cues).

[2010--2012]
Research on intelligent whole-body reaching in humanoid robots, at Takanishi Laboratory, Humanoid Robotics Institute of Waseda University, Tokyo, Japan.
The research goal was to implement learning-based reaching on a full humanoid robot (KOBIAN-R), following bio-inspired principles and supporting autonomous and flexible behaviors.
I studied how the robot could build autonomously a motor representation of its workspace from the goal-directed interaction with the environment, and how such knowledge could be used to plan and execute complex behaviors, such as bimanual reaching and whole-body optimal reaching. I also investigated how to use and learn models of different tools and how to integrate locomotion. Part of my studies were in the framework of the EU Project Robosom.

[2007--2010]
Research on autonomous sensorimotor learning in a humanoid robot, at the Italian Institute of Technology, Genova, Italy.
The research included study on the basis of learning in biological and artificial systems, robot advanced programming and coordinated control of multiple DOFs, robot maintenance and updating (software, firmware, mechanics, electronics, computers network).
I applied my research on the humanoid robot James, that learned autonomously how to move its human-like neck and how to reach for visually detected objects. Part of my studies were in the framework of the EU Projects Contact and RobotCub.

[2005--2006]
Design and realization of tactile sensors for an anthropomorphic robotic hand, at LIRA-Lab laboratory, University of Genova, Italy.
The research included study on materials (elastomers), electronics (data collection board, cable routing, signal processing), software and firmware programming and robot control.

[2005]
Collaboration project with the Robotic Division of a car manufacturing company, concerning the realization and installation of tactile sensors for a humanoid robot, in Brussels, Belgium.

[2005]
Three-months stage (march--june) involving the study and impementation of control strategies for robotic manipulators, at the Department of Robotics and Automation of the Technical University of Sofia, Bulgaria. Different control techniques have been tested on the SCORBOT-ER VII robot (Eshed Robotec).

[2004]
Realization of some units of the 'Lido' laboratory website (http://www.lido.dist.unige.it), University of Genova, Italy.
Website based on ASP technology, interfacing with a SQL-Server database.

[2003]
Design, development and realization of a MasterGlove system for telerobotics, at GRAAL-Tech laboratory, University of Genova, Italy.
The realized device is a sensorized glove able to measure hand joints angle, which has been used to remotely control a robotic hand.