Adaptation and learning for hierarchical intelligent control of robotic manipulator

Abstract

This article deals with a new strategy for hierarchical intelligent control. We propose this strategy for the neural network-based controller to be generalized with the higher level control based on the artificial intelligence technology and to acquire knowledge heuristically. Therefore, this system comprises two levels: learning and adaptation. The neural networks are employed for both levels. The learning level has a hierarchical structure for recognition. It is used for strategic planning of robotic manipulation in conjunction with the knowledge base to expand the adaptive range to the environment. Recent information from the adaptation level updates the learning level through the long-term learning process. Conversely, adaptation is used for the adjustment of the control law to the status of the dynamic process. This is one of the analogous control systems to the human cerebral control structure.