Sensorimotor learning for artificial body perception
This addresses the problem of enabling machines to perceive their own bodies, which is incremental as it builds on existing methods without introducing a new paradigm.
The paper reviews recent approaches for modeling artificial body self-perception, including Bayesian estimation and deep learning, highlighting the potential of unsupervised or semi-supervised crossmodal learning methods, but notes that challenges remain in achieving full artificial multisensory body perception.
Artificial self-perception is the machine ability to perceive its own body, i.e., the mastery of modal and intermodal contingencies of performing an action with a specific sensors/actuators body configuration. In other words, the spatio-temporal patterns that relate its sensors (e.g. visual, proprioceptive, tactile, etc.), its actions and its body latent variables are responsible of the distinction between its own body and the rest of the world. This paper describes some of the latest approaches for modelling artificial body self-perception: from Bayesian estimation to deep learning. Results show the potential of these free-model unsupervised or semi-supervised crossmodal/intermodal learning approaches. However, there are still challenges that should be overcome before we achieve artificial multisensory body perception.