Grounding the Experience of a Visual Field through Sensorimotor Contingencies
This work addresses the challenge of autonomous perception for robotics, but it is incremental as it builds on existing sensorimotor theory with a specific application.
The paper tackles the problem of enabling autonomous robots to develop perceptive abilities through interaction, rather than hand-designed algorithms, by applying sensorimotor contingencies theory to discover a visual field as regularities in sensory inputs. The result is a formalism for capturing these regularities in a predictive model, evaluated on a simulated retina-like system.
Artificial perception is traditionally handled by hand-designing task specific algorithms. However, a truly autonomous robot should develop perceptive abilities on its own, by interacting with its environment, and adapting to new situations. The sensorimotor contingencies theory proposes to ground the development of those perceptive abilities in the way the agent can actively transform its sensory inputs. We propose a sensorimotor approach, inspired by this theory, in which the agent explores the world and discovers its properties by capturing the sensorimotor regularities they induce. This work presents an application of this approach to the discovery of a so-called visual field as the set of regularities that a visual sensor imposes on a naive agent's experience. A formalism is proposed to describe how those regularities can be captured in a sensorimotor predictive model. Finally, the approach is evaluated on a simulated system coarsely inspired from the human retina.