Digitally programmable analog membership-function circuits have been invented for use in fuzzy-logic systems. Heretofore, fuzzy membership functions have been implemented, variously, by use of purely analog or purely digital circuits. Purely digital circuits afford flexibility at the cost of less speed and greater circuit area. Purely analog circuits offer greater speed and occupy less area but tend to be limited in flexibility and programmability. The present invention takes advantage of both the programmability of digital circuitry and the speed and compactness of analog circuitry. The analog aspect of the invention creates the potential for high-speed parallel processing with relatively low power consumption.

A Combination of Analog and Digital Circuits implements a trapezoidal membership function of an analog input signal. The digital inputs establish the parameters of the trapezoid.
In designing a circuit according to the invention, one uses current-mode circuitry to implement fuzzification in a fully parallel architecture while using a digital interface for programmability of attributes of membership functions. In a typical case, a membership-function circuit accepts an analog input potential and implements a trapezoidal membership function (see figure). First, the input potential signal (X) is converted to a current-mode signal (Iin) for further processing. Two current-mode digital-to-analog converters (DACs) provide currents IL and IR that represent zero-crossing locations of the left and right legs, respectively, of the trapezoidal membership function.

Current subtractors generate Iin — IL and IR — Iin, and feed these difference currents as analog inputs to dividing digital-to-analog converters (DIV-DACs). [Alternatively, one could use multiplying digital-to-analog converters (M-DACs); the advantage of DIV-DACs is that they provide more evenly spaced increments, which makes the range of programmable slopes more useful.] The digital input to each DIV-DAC is a number inversely proportional to the slope of the corresponding leg of the trapezoidal membership function. In each DIV-DAC, the analog difference-current input is divided by the digital inverse-slope input to obtain the ordinate on the affected leg of the trapezoid. A comparator circuit selects the output of either the left- or the right-leg DIV-DAC, and a limiting circuit clips the output at a level corresponding to the top of the trapezoid. The output current of the limiting circuit is converted to an output voltage.

This work was done by Tyson Thomas and David Weldon of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp  under the Electronics & Computers category.

NPO-20810

Topics:
Analysis methodologies Anatomy Architecture Body regions Design processes Electronics & Computers Fuzzy logic Leg Logistics Medical, health and wellness Medical, health, and wellness
This Brief includes a Technical Support Package (TSP).
Document cover
Digitally Programmable Analog Membership-Function Circuits

(reference NPO-20810) is currently available for download from the TSP library.

Don't have an account?

Magazine cover
NASA Tech Briefs Magazine

This article first appeared in the June, 2001 issue of NASA Tech Briefs Magazine (Vol. 25 No. 6).

Read more articles from the archives here.

Overview

The document is a technical support package prepared under the sponsorship of the National Aeronautics and Space Administration (NASA) and details the development of Digitally Programmable Analog Membership-Function Circuits. Authored by inventors David Weldon and Tyson Thomas, the report is associated with NASA Technical Brief Vol. 25, No. 6 and is part of the Jet Propulsion Laboratory (JPL) New Technology Report NPO-20810.

The primary motivation behind this development is to create a high-speed fuzzy logic system suitable for real-time data fusion applications. Traditional fuzzy logic systems often face limitations in speed and reconfigurability, which can hinder their effectiveness in dynamic environments. The proposed solution leverages analog current-mode circuitry to implement fuzzification in a fully parallel architecture, allowing for rapid processing while maintaining a digital interface for programmability of membership function attributes.

The document emphasizes the innovative approach of using analog circuitry to achieve high-speed performance, which is critical for applications requiring immediate data processing and decision-making. By enabling a digital interface, the system allows users to easily modify and configure membership functions, enhancing the adaptability of the fuzzy logic system to various operational scenarios.

Additionally, the report includes a disclaimer stating that references to specific commercial products or manufacturers do not imply endorsement by the U.S. Government or JPL. It also clarifies that the work was conducted at JPL under a contract with NASA, ensuring that the research and development are aligned with governmental standards and objectives.

Overall, this technical support package outlines a significant advancement in fuzzy logic technology, highlighting its potential applications in fields that require fast and flexible data processing capabilities. The combination of analog and digital technologies in this system represents a promising direction for future developments in real-time data fusion and intelligent systems.