Basic agents for visual/motor coordination of a mobile robot

Abstract

Present work addresses the guidelines that have been followed to construct Basic Behavioral Agents for visually guided navigation tasks within the framework of a Hierarchical Architecture. Visual and motor interactions are also described within this generic framework that allows for an incremental adaptation of knowledge, to be reused in tasks of ever increasing complexity. Basic Locomotion Agents as, Stop&Backward, Avoid, and Forward are implemented as Fuzzy knowledge- based systems to embody the uncertainty and imprecision inherent to real system-environments. Basic Visual Agents as, Saccadic, Contour, and Center have been developed under a space-variant representation in an anthropomorphic approach. Coordination of Basic Agents has been demonstrated with a caterpillar type mobile robot in tasks related to qualitative descriptions of visual stimuli.