Robot behavior learning in logical perceptual space

The paper addresses features extraction of sensor date for robot behavior learning using factor analysis. Redundancies In sensor types and quantities are common in sensing competence of robots. The redundancies cause the high dimensionality of the perceptual space. It Is impractical to incorporate all available sensor information in robot behavior learning due to the huge memory and computational requirements. This paper proposes a new approach to extract important knowledge from sensor date based on the inter-correlation of sensor data using factor analysis and construct Weal perceptual space for robot behavior learning. The logical perceptual space is constructed by the extracted hypothetical latent factors extracted, which have fewer dimensions than raw sensor data, using factor analysis. Experiments have been conducted to demonstrate the effectiveness of employing logical perceptual space In robot behavior learning with a mobile robot.