Unsupervised Classiication of Sensory-motor States in a Real World Artifact Using a Temporal Kohonen Map

Classiication is a fundamental act of cognition. It underlies nearly all learning, transfer of learning, generalization and abstraction. One of the well-known hard and fundamental problems is the one of perceptual aliasing, i.e. that the sensory stimulation caused by one and the same object varies enormously depending on the distance from the object, orientation, lighting conditions, etc. In this paper an approach to the solution of this problem, perceptual aliasing, is illustrated in a set of experiments with a robot that learns to distinguish betwenn graspable and non-graspable pegs. Herewith, the robot learns to classify pegs based on information about its sensory-motor state.