Designing a Nervous System for an Adaptive Mobile Robot

摘要

Historically, the original motivation for building robots was to facilitate industrial automation. Indeed, the word robot derives from the Czech robota meaning ‘work’. However, increasing attention has been paid to the possibilities which robots—especially mobile, learning robots—offer for modelling intelligent behaviour. In parallel distributed processing or ‘connectionism’ has become the majority paradigm in cognitive modelling and artificial intelligence. The gill-withdrawal reflex in Aplysia is the classic example of neurobiology being able to explain an adaptive behaviour. This is a simple defensive reflex action in which the gill is withdrawn after a stimulus. The gill lies under the protective mantle shelf, in the mantle cavity, which terminates with a fleshy spout—the siphon. Habituation is, perhaps, the simplest form of ‘learning’ possible. It occurs when an animal learns to ignore a weak repetitive stimulus which is neither rewarding nor noxious.