Intrinsic and extrinsic implementation of a bio-inspired hardware system

Due to the complexity of modern day systems and the fact that most real world problems are not fixed but rather change with time, it is necessary to look for new methodologies and strategies to deal with reliability problems. Practical experience has demonstrated that the goal of building totally fault-free systems, although desirable, is impossible to achieve. Considering these observations and taking inspiration from biology, this project attempts to create an adaptive hardware system that is capable of modifying its architecture and thus its behaviour autonomously, to cope with variations in its external environment. The adaptive hardware system takes the form of a neurofuzzy network implemented on an FPGA device. Two alternative approaches are implemented, extrinsic and intrinsic, and results detailing the performance of the two are presented and compared. 2003 Elsevier Inc. All rights reserved.