Research Topics


Inverse dynamic Learning of the iCub Humanoid Robot

In order to control the iCub humanoid robot a model describing the relation between the actuator inputs, the interactions with the world and body accelerations is required. This model can be obtained using the techniques of rigid body dynamics. However, it is not able to capture complex phenomena acting in the ‘‘real’’ model. Our aim is to obtain a more flexible and accurate model from experimental data using Machine Learning Techniques.

Partners: IIT and DIBRIS.

People: A. Chiuso, M. Zorzi, D. Romeres


Functional Neuroimaging

Neuroimaging includes the use of various techniques to either directly or indirectly image the structure, function of the nervous system. It is a relatively new discipline within medicine and neuroscience-psychology. Functional neuroimaging is the use of neuroimaging technology to measure an aspect of brain function, often with a view to understanding the relationship between activity in certain brain areas and specific mental functions or illness (e.g. Alzheimer's disease). Our aim is to recover those brain areas from experimental data and using system identification techniques. More Details


System Identification

System identification is the discipline which uses statistical methods to build mathematical models of dynamical systems from measured data. The knowledge of the mathematical model describing a dynamical system is important. For instance, one can predict the future evolution of the system. System identification is also fundamental in adaptive control (here the controlled system is not know and thus the controller is designed according to the model found performing system identification).


Graphical Models

A graphical model is a probabilistic model for which a graph expresses the conditional dependence structure (edges) between random variables (nodes). This graphical interpretation is useful to show in a concise way how these random variables interact each other. Graphical models are used in many fields, for instance in econometrics and biology.