Implementation of an Adaptive Visually-Guided Neural Controller for Single Postures

A neural network model has been developed that achieves adaptive visual-motor coordination of a multijoint arm, without a teacher. The model has been computer simulated to adaptively position an arm so that it reaches a cylinder arbitrarily positioned in space. The model uses a new neural architecture and a new algorithm for modifying neural-connection strengths. A simpler version of the model has also been implemented with an industrial image processing system and an industrial robot arm to adaptively reach in space. The neural model called INFANT is designed to be generalized for coordinating any number of topographic sensory inputs with limbs of any number of joints. The general scheme of the neural model is also proposed.