Motivation

• Building robots that learn to recognize their own body for improving interaction.
• Investigate the self construction in humans using synthetic models.

Providing this kind of learning to robots, this research will aid the transition from cylindrical robotic platforms that avoid obstacles to a new generation of multisensory humanoids robots, with the sufficient adaptation ability for achieving correct interaction between machines and humans.

Scientific goal

To develop a synthetic model of the sensorimotor self that captures the relationship between what the robot perceives (sensory response) and the actions that it exerts, in order to enable the machine to recognize its own body from other elements in the environment.