Data processing: artificial intelligence – Fuzzy logic hardware
Reexamination Certificate
1996-03-29
2002-05-07
Davis, George B. (Department: 2122)
Data processing: artificial intelligence
Fuzzy logic hardware
C706S004000, C706S005000, C706S045000, C706S046000, C706S050000, C706S052000
Reexamination Certificate
active
06385598
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to fuzzy processors, and more particularly to a fuzzy processor which can also perform non-fuzzy processes.
2. Discussion of the Related Art
It is known that there are various kinds of fuzzy processors. An example of a classic fuzzy processor is the MB94110 processor by Fujitsu, comprising a block that performs fuzzy instructions. Another example is the SAE81 C99 processor by Siemens, which can also load various knowledge bases from an external memory, although this is not possible conditionally or concurrently with normal fuzzy processing.
Accordingly, known fuzzy processors only process fuzzy instructions and cannot integrate fuzzy control with the processing of classic instructions such as shift, rotate, compare, or arithmetic/logic instructions, and also have no signals such as interrupt, stack, etcetera. Accordingly, it is not possible to perform any kind of arithmetic/logic processing on the inputs and on the outputs of the fuzzy controller.
All known fuzzy control units furthermore perform control by processing the fuzzy instructions sequentially, with no possibility of jumping from one instruction to another.
Another characteristic of known fuzzy processors is that they process a fixed number of inputs and a fixed number of fuzzy rules. A certain number of membership functions are associated with each fuzzy input, and the set of these functions is termed a knowledge base. Both the set of fuzzy rules and the knowledge base remain unchanged throughout processing.
SUMMARY OF THE INVENTION
Accordingly, an aim of the present invention is to provide a fuzzy processor with an improved architecture that is more versatile than known fuzzy processors.
Within the scope of the above aim, an object of the present invention is to provide a fuzzy processor with an improved architecture that is capable of handling inputs and outputs.
Another object of the present invention is to provide a fuzzy processor with improved architecture that is capable of loading various knowledge bases or sets of fuzzy rules from outside during normal operation.
Another object of the present invention is to provide a fuzzy processor with an improved architecture capable of changing the knowledge base or the set of fuzzy rules to be processed during normal processing, conditionally or unconditionally, the conditions occurring on the inputs or on the outputs.
Another object of the present invention is to provide a fuzzy processor with an improved architecture capable of performing arithmetic/logic processing on the inputs and on the outputs of the fuzzy controller, processing non-fuzzy instructions, and handling interrupt signals.
Another object of the present invention is to provide a fuzzy processor with an improved architecture capable of performing conditional or unconditional jumps within fuzzy instructions.
Another object of the present invention is to provide a processor that is highly reliable and relatively easy to manufacture at a competitive cost.
This aim, these objects, and others which will become apparent hereinafter are achieved by a fuzzy processor with an improved architecture, comprising: a fuzzy rule processor, an internal fuzzy instruction memory, and an internal knowledge base memory, and characterized in that it also includes an arithmetic-logic unit, a control unit capable of performing non-fuzzy instructions typical of conventional microprocessors, and an internal memory comprising non-fuzzy instructions.
REFERENCES:
patent: 5261036 (1993-11-01), Nakano
patent: 5263125 (1993-11-01), Viot
patent: 5285376 (1994-02-01), Struger
patent: 5305424 (1994-04-01), Ma et al.
patent: 5430828 (1995-07-01), Yamamoto
patent: 5524174 (1996-06-01), Eichfeld et al.
patent: 5600757 (1997-02-01), Yamamoto et al.
patent: 5790755 (1998-08-01), Pagni et al.
H. Watanabe et al., “VLSI Fuzzy Chip and Inference Accelerator Board Systems,” Proceedings of the 21st International Symposium on Multiple-Valued Logic, May 1991, pp. 120-127.*
H. P. Eichfeld et al., “An 8 B Fuzzy Coprocessor for Fuzzy Control,” Feb. 1993 IEEE International Solid-State Circuits Conference, pp. 180-181, 286.*
Partial European Search Report from European Patent Application 95830117.8, filed Mar. 30, 1995.
IEEE Micro, vol. 13, No. 5, Oct. 1993, pp 37-48, Kazuo Nakamura, et al., “Fuzzy Interference and Fuzzy Interference Processor”.
Elektronik, vol. 41, No. 17, Aug. 18, 1992, pp 40-46, Rossman, J. “Der Realismus Kehrt Ein”.
Technische Rundschau, vol. 85, No. 48, Dec. 3, 1993, Bern CH, pp 30-32, J. Zhang, et al., “Mit Fuzzy: Hohe Regelgeschwindigkeit”.
Catania Vincenzo
Giacalone Biagio
Pappalardo Francesco
Pelos Enrico
Consorzio per la Ricerca Sulla Microeleti Nel Mezzogiorno
Davis George B.
Morris James H.
Wolf Greenfield & Sacks P.C.
LandOfFree
Fuzzy processor with architecture for non-fuzzy processing does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Fuzzy processor with architecture for non-fuzzy processing, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fuzzy processor with architecture for non-fuzzy processing will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2830565