Trustworthy Artificial and Embodied Intelligence

Reliable AI and robotic systems, able to operate in dynamic and critical scenarios 

The activity focuses on frontier aspects of Artificial Intelligence (AI) and on the ability of artificial agents (e.g. robots) to develop intelligent behaviours as emerging from the interactions with the environment, through the body and the sensory-motor system of the agent itself (Embodied Intelligence). 

Within the TAEI FoReLab, a widespread, reliable and integrated use of both are pursued and promoted. 

The research includes the design of behaviours of agents keeping into consideration the environment and the constraints in terms of body, perception, motor and brain systems to intelligently operate in (at least partially) unknown, not structured, dynamical and not predictable scenarios (e.g. multi domain - air-ground-water - search and rescue or environmental monitoring). 

Since humans’ trust in a technology strongly relies on understanding how it works, particular attention is given to the development of algorithms for learning explainable models, also through distributed and privacy-preserving approaches (Human-centered Trustworthy AI).

Technologies

Abel

Abel, a new generation hyper realistic humanoid robot

Abel is conceived to be a research platform for social interaction, emotion modeling, and studies on embodied intelligence. The robotics platform can implement and test theories coming from neuroscience, psychology and sociology, with very promising applications in therapy and diagnosis of mental illness, learning disabilities, autism spectrum, and dementia.
federated_XAI

Federated Learning of explainable AI models

Federated learning has been introduced for preserving data privacy. We are developing a framework and specific algorithms for enabling federated learning of explainable AI models.
Explainable AI

Explainable models by design

We are developing algorithms for learning from data accurate models, such as decision trees and rule-based systems, which are explainable by design.
Ego

Avatar Embodied Intelligence for human-robot interaction

Humans can understand and interact with “intelligent” soft robotic systems reproducing human ability to smoothly operate in the environment, thanks to advanced haptic sensors and interfaces to measure and display physical contact information.
digital twins

Digital Twins

Our research investigates the interplay between DTs and networks under different perspectives: to replicate a network itself and to instantiate DTs in an effective and scalable fashion. Another research activity focuses on DTs of cyber-physical systems.
robot_team_2

Embodied Intelligence for cooperative robot operations in unknown environments

Teams of heterogeneous robots capable of intelligently operate in (at least partially) unknown, not structured, dynamical and not predictable scenarios.
speech_analysis

Speech Processing

Speech processing can extract audio features characterizing specifically each speaker. These features describe temporal intonation, e.g. prosody, or voice quality characteristics.
cuore_cervello

Predictive models to study peripheral and central nervous systems

Future systems for human-machine interaction will be improved both in the design and in the working phases by improved models of human affective state and behaviour.
haptics_verde

Wearable Haptics for Telerobotics

We propose the usage of a wearable haptic interface for softness rendering as a lump display. The device is integrated in a teleoperation architecture that simulates a robot-assisted surgical palpation task of leiomyomas.
Haptics_1

Augmented Reality through Wearable Feel- Through Haptics

We developed an approach that allows, for the first time, to modulate the perceived softness of real objects using a Feel-through wearable that uses a thin fabric as interaction surface.