What we do

Scientists and engineers of the iCub Project work at the forefront of mechatronics and artificial intelligence research to build robots of humanoid shape. The iCub "crew" members cover the full development cycle from software to mechanical design and from machine learning to neuromorphic chip realization. Our team aims at developing robots that on the one hand can learn and adapt from their mistakes, and on the other are robust to work in real-world practical scenarios. Activities encompass the construction of the hardware of the humanoid robots, that we call "bodyware", and software which will make, one day, machines of intelligence comparable to humans. We call this technology "mindware".

On the bodyware side we developed the iCub and R1 robot platforms. Never tired of new challenges, we are simultaneously addressing the development of the technologies for the next generation of robots based on soft and adaptable materials for sensing, actuation, and computation.

On the mindware side, the iCub team develops visual, auditory and tactile perception skills for our robots as well as the ability to gaze, reach and manipulate objects while walking freely to reach their targets, interacting naturally with the environment and their human teachers.

The iCub Project blends results from various IIT Research Lines by applying the principles of systems engineering and by seeking worldwide collaboration opportunities. Not less importantly, the iCub team is active in several industrial partnerships. The iCub Project represents one of IIT's thrusts in the transfer of robotics technologies to industrial exploitation (see also Technology Transfer).

The iCub Project research areas include innovative mechanical and electronic design, development of software and the its maintenance, sensing and machine learning, control of movement, manipulation, and studies on human-robot interaction. We do not eschew full-blown applications as for example in the field of robot assistance for the elderly or children with autistic spectrum disorders, in the home entertainment sector, and generically in the digital industry domain.