Title: Online context switching for real time control applications using reconfigurable fuzzy inference chip

Authors: Qi Cao, Meng Hiot Lim, Keck Voon Ling, Ju Hui Li, Ching Wern Chan

Addresses: School of Electrical and Electronic Engineering, Block S1, Nanyang Technological University, Nanyang Avenue, 639798, Singapore. ' School of Electrical and Electronic Engineering, Block S1, Nanyang Technological University, Nanyang Avenue, 639798, Singapore. ' School of Electrical and Electronic Engineering, Block S1, Nanyang Technological University, Nanyang Avenue, 639798, Singapore. ' School of Electrical and Electronic Engineering, Block S1, Nanyang Technological University, Nanyang Avenue, 639798, Singapore. ' School of Electrical and Electronic Engineering, Block S1, Nanyang Technological University, Nanyang Avenue, 639798, Singapore

Abstract: The reconfigurable fuzzy inference chip (RFIC) is a novel concept of a hardware fuzzy inference processor. It supports online realtime context switching. Such a capability is significant, particularly for situations where the application scenarios are expected to change dynamically. Furthermore, the RFIC can be extended to support hardware that reconfigures itself, a notion termed as evolvable fuzzy hardware. We outline in this paper a prototype implementation of the RFIC and demonstrate its capability in handling online realtime context switching for the control of a rotary inverted pendulum. Our results confirm the applicability of the RFIC in supporting the implementation of intrinsic evolvable fuzzy hardware.

Keywords: reconfigurable fuzzy inference chip; RFIC; online context switching; intrinsic evolution; evolvable fuzzy hardware; RFIC; real time control; fuzzy control; fuzzy logic; hardware fuzzy inference processor; rotary inverted pendulum; pendulum control.

DOI: 10.1504/IJCAT.2006.011141

International Journal of Computer Applications in Technology, 2006 Vol.27 No.2/3, pp.153 - 165

Published online: 19 Oct 2006 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article