Autonomous neural control in flexible space structures

Abstract

The present paper proposes the use of a hybrid connectionist system as a learning controller with reconfiguration capability. Embedded with adjustable time-delays, an adaptive radial basis function network offers an effective modeling technique to capture most of the spatiotemporal interactions among the structure members. A fuzzy-based fault diagnosis system is applied for health monitoring to provide the neural controller with various failure scenarios. Associative memory is incorporated to compensate for catastrophic changes of structural parameters by providing a continuous solution space of acceptable controller configurations. This paper addresses the theoretical foundation of architecture and demonstrates its applicability via specific examples.