Data processing: artificial intelligence – Fuzzy logic hardware – Digital fuzzy computer
Reexamination Certificate
1998-04-08
2002-02-12
Davis, George B. (Department: 2122)
Data processing: artificial intelligence
Fuzzy logic hardware
Digital fuzzy computer
C706S003000, C706S900000
Reexamination Certificate
active
06347308
ABSTRACT:
FIELD OF THE INVENTION
The present invention is directed to the field of information processing, and, more particularly, to the field of information processing using fuzzy logic.
The invention relates to electronic systems in which the management of information is particularly complex and cannot typically be processed analytically, although the invention could also be applied in these systems. Furthermore, the invention is especially useful for systems organized in networks for it provides greater flexibility in processing.
BACKGROUND OF THE INVENTION
A method for processing of information in fuzzy logic comprises, first, the assessment of the likelihood of the membership of the value of a variable in membership functions of this variable and, second, the preparation of elementary decision signals for the application of rules of the type “if a given variable belongs to a given membership function, then the elementary decision signal will be equal to a given result”. At the end of processing, a combination of the different elementary signals is used to obtain a processed result that is applicable to an activation element or actuator of the system.
The units of such a system which send out signals representing variables are chiefly detectors that are independent or placed directly on an actuator unit, setting units such as a keyboard, or a master processor that takes control of the system to force it to act in a specified way. The actuators are generally drive units (electrical motors, remote control relays or even display screens). The actuators may nevertheless be a part of the detector itself, for example, its address register, when this register is modified as a function of an assignment that is dictated to it by a processor of the system. This is why the term “transmitter” is used herein, since a detector can be partly an actuator and an actuator can be partly a detector. In other words, a transmitter unit is a unit that transmits information to another unit, to at least one fuzzy logic processor.
One example of a system managed according to a fuzzy logic information processing method is a system of surveillance against intrusion. In such a system, there are presence detectors to detect the presence of a unauthorized individual inside a room under surveillance. Coupled with this detection, there may also be a temperature detection system to report, for example, the occurrence of a fire.
With regard to presence detectors, there are known ways of using frequency modulation detectors to recognize the presence of an object or a person moving in a space under surveillance. However, if the moving object is insignificant such as, for example, a butterfly passing before a detector or a cat going under a grating, it is important that no untimely alarm should be triggered. To resolve these problems, a local information-filtering system has already been provided. For example, if a presence detector has detected a presence at a given time and no longer detects it thereafter (because the butterfly has gone out of the field of the detector), then the local filters are adjusted so that no alarm is transmitted. The filters may include operations that are themselves complex.
Furthermore, especially when the system is in a network, there may be many information elements to be processed, and preference has to be given to certain information elements as compared with others. Thus, it is possible to conceive of a situation where it is very important to react swiftly when there is a crossing of a temperature that indicates a fire at a certain place while, at the same time, simple operations of management also have to be performed, but with a smaller degree of urgency.
In the prior art, there is no solution to this problem. Either the filtering is done locally and information elements are transmitted only if they are sufficiently serious or else all the information elements are processed one after the other with the consequences of delay that this might entail.
SUMMARY OF THE INVENTION
An object of the present invention is to overcome the problem that filtering is typically done locally and information elements are transmitted only if they are sufficiently serious or else all the information elements are processed one after the other with the consequences of delay that this might entail.
The invention is aimed at overcoming this problem by recommending the dispatch, at the same time as the information to be processed, of a force coefficient, the term “force coefficient” being a generic term that quantifies the importance of the information transmitted. For example, in the case of the butterfly that penetrates the field of the detector, on the one hand an information element could be sent reporting the fact that a detection has taken place and, on the other hand, by way of a force coefficient, a value of a variable could be sent, this value being greater as the information itself is significant. To put it simply, the force coefficient could be proportional to the time that the butterfly remains in the field of the detector.
In the other example, the force coefficient, rather than being measured prior to its transmission, may be set definitively: temperature sensors or fire detectors are provided with a maximum force coefficient, and all the other logistical operations may be provided with a smaller force coefficient.
The present invention is directed to a method of processing information by fuzzy logic. More particularly, the method preferably comprises the following steps:
data elements are sent from the transmitters to at least one fuzzy logic processor;
the fuzzy logic processor processes the data elements sent on the basis of membership functions and on the basis of conditional rules, and the fuzzy logic processor produces a processed signal, wherein:
data elements are sent to the fuzzy logic processor in common with a force coefficient of these data elements; and
the fuzzy logic processor weights the signal processed by this force coefficient.
As shall be seen hereafter, the invention is furthermore preferably put into operation in the context of a use in a network. In an installation of this kind, each detector or actuator reports itself by its address. According to one advantage of the invention, the force coefficient is incorporated into the address sent by the transmitter and corresponding address classes are organized.
REFERENCES:
patent: 4961225 (1990-10-01), Hisano
patent: 5051932 (1991-09-01), Inoue et al.
patent: 5249269 (1993-09-01), Nakao et al.
patent: 5252789 (1993-10-01), Sirag, Jr. et al.
patent: 5267348 (1993-11-01), Someya et al.
patent: 5363023 (1994-11-01), Choho
patent: 5561738 (1996-10-01), Kinerk et al.
patent: 5655056 (1997-08-01), Takaka et al.
patent: 5671138 (1997-09-01), Bessacini et al.
patent: 5687290 (1997-11-01), Lewis
patent: 5706193 (1998-01-01), Linzenkirchner
patent: 5720005 (1998-02-01), Goke et al.
patent: 5731868 (1998-03-01), Okey et al.
patent: 5758025 (1998-05-01), Wu
patent: 5758031 (1998-05-01), De Silva
patent: 5956464 (1999-09-01), Madni et al.
patent: 0 385 387 (1989-02-01), None
Dörsam et al, “Fuzzy-Based-Force-Adaptation for Multifingered Robot Hands”, IEEE Proceeding of the 3rd Conference on Fuzzy Systems, Jun. 1994.*
Lever et al, “A Fuzzy Control System for an Automated Mining Excavator”, IEEE International Conference on Robotics and Automation, May 1994.*
Nitta et al, “A High Performance Fuzzy Inference Processor and its Evaluation System”, Proceedings of 1995 IEEE Interational Conference on Fuzzy Systems, 1995, International Joint Conference of the 4th IEEE Internationl Conference on Fuzzy Systems and the 2nd Inter-Fuzzy Engineering Symposium.*
Patent Abstracts of Japan, vol. 15, No. 16, (p. 1152) Jan. 14, 1991, JP 02259906.
M. Mizumoto, “State Dependent Fuzzy Controls by Fuzzy Singleton-Type Reasoning Method”,Soft Computing in Intelligent Systems and Information Processing, Proceedings of the 1996 Asian Fuzzy Systems Symposium (Cat. No. 96TH8239).
Thomas Tilli, “Fuzzy-Logik in der Elektronik Teil 3: Ablaufsteuerungen”,Elektronik, vol. 43
Allen Dyer Doppelt Milbrath & Gilchrist, P.A.
Davis George B.
Jorgenson Lisa K.
SGS-Thomson Microelectronics S.A.
LandOfFree
Method for the processing of information by fuzzy logic does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for the processing of information by fuzzy logic, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for the processing of information by fuzzy logic will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2969377