Babybot: Awareness and presence

BabyBot was born at the robotics laboratories of the University of Genoa and is the spearhead of the ADAPT project funded by the European Commission to explore the technological frontiers of artificial intelligence. The appearance of BabyBot resembles, in shape and size, that of a two-year-old child: coordinated by Prof. Giorgio Metta, the researchers of the ADAPT project want to discover the secrets of self-perception, the fascinating psychocognitive process at the basis of self-awareness.

About the project

The Babybot was created by a group of professors and researchers from the Faculty of Engineering of the University of Genoa, which has been widely reported in the Italian and foreign press. To be precise, it was the Lira Lab 7 team that gave life to a robot "baby". This was done in order to better understand the mechanisms of the human brain. Not only does the robot do many things because its "brain" can contain many programs, but it can also act in different ways because it has learned these behaviors by interacting with people, that is, because the robot is equipped with programs that allow it to remember what it sees, hears, touches, and this memory is then transformed into a stimulus for action. An action that becomes unpredictable.

Like a baby

"The robot will allow us to find out, through settings, what are the details of interactions between the baby and strangers, between the baby and objects, between the baby and parents," Metta said. It is based on a mathematical model that views self-awareness as a process of action, cognition, and perception: according to the ADAPT model, a child imagines that objects in an environment are not "real" but part of a general perceptual process.

The new future

Why develop such a project? Because of the many practical applications, from psychology to the development of new robots to the creation of increasingly precise virtual reality environments artificially grafted onto human consciousness.
Teach robots to be human and humans to think like robots? Professor Sandini, one of the project leaders, says: "It happens very often, but it is part of the way we work. We do not program the robot to do something in a certain way. We give an input to which the machine responds autonomously and is able to correct itself. If Babybot is supposed to fall while grasping an object, it learns from its mistake and tries again until it gets it right. Another example: although it is programmed to pay attention only to colored objects, if it is stimulated with gray figures for a certain period of time, it learns to 'react' to them.

On the Way to Artificial Consciousness

The learning mechanism of these models is therefore very similar to that of animals and humans. This stage has already been reached in some prototypes, some of which are, among other things, bioelectronic hybrids, i.e. they also contain neurons of organic origin. The next step, foreshadowed by many science fiction writers and some robotics engineers at work, is the emergence of true artificial consciousness. Accepting a strictly materialistic ontology and taking it to the extreme, many researchers are inclined to believe that consciousness is ultimately scientifically comprehensible and therefore technically reproducible.

Bibliography

• Riccardo Campa, 'Le armi robotizzate del futuro. Intelligenza artificialmente ostile? Il problema etico', Rapporto 2010 STEPI-T-3.
• Hillman, M. "Introduction to the special issue on rehabilitation robotics." Robotica 16, no. 5 (September 1998).
• Pransky, Joanne. "The Pransky interview: Dr Nabil Simaan, Vanderbilt University Professor of Mechanical Engineering, Computer Science and Otolaryngology, Thought Leader in Medical Robotics." Industrial Robot: the international journal of robotics research and application 48, no. 4 (July 29, 2021): 473–77.
• Robotics Society of Japan, Advanced Robotics 10,6, (1996). 4) T.Fukuda and T.Ueyama, Cellullar Robotics and Micro Robotic Systems,World Scientific, Singapore, (1994).
• H.Asama, T.Fukuda, T.Arai and I.Endo eds., Distributed Autonomous Robotic Systems, Springer-Verlag, Tokyo, (1994).
• Asama, Hajime. "Special Issue on Distributed Robotic Systems." Journal of Robotics and Mechatronics 8, no. 5 (October 20, 1996): 395.
• Kamilaris, Andreas, and Nicolò Botteghi. "The penetration of Internet of Things in robotics: Towards a web of robotic things." Journal of Ambient Intelligence and Smart Environments 12, no. 6 (November 26, 2020): 491–512.
• Mason, Matthew T. "Toward Robotic Manipulation." Annual Review of Control, Robotics, and Autonomous Systems 1, no. 1 (May 28, 2018): 1–28.
• Khanna, Omaditya, Ryan Beasley, Daniel Franco, and Simon DiMaio. "The Path to Surgical Robotics in Neurosurgery." Operative Neurosurgery 20, no. 6 (May 13, 2021): 514–20.