JPH0340105A - Fuzzy output sensor controller - Google Patents

Abstract

PURPOSE:To simplify a control operation when a sensor is separate from a fuzzy controller by controlling the membership function of the antecedent part of a fuzzy rule and calculating the goodness of fit of the membership function in accordance with the input signal of the sensor to output this calculated adaptation to the fuzzy controller. CONSTITUTION:A means 2 is used to store the membership function of the antecedent part of a fuzzy rule for the fuzzy arithmetic together with the means 5 - 8 which control the stored membership function, and a means which calculates the goodness of fit of the membership function in accordance with the input signal of a sensor and outputs the adaptation to a fuzzy controller 20. Thus it is possible to control the sensitivity of the sensor and also to control the membership function of the antecedent part of the fuzzy rule while controlling the distance between the sensor and a work when a fuzzy output sensor controller is set close to the sensor. As a result, a control operation is simplified when the sensor is separate from the controller 20.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a fuzzy output sensor controller that outputs the goodness of fit of a membership function in the antecedent part of a fuzzy rule to a fuzzy controller in response to a human input signal from a sensor. .

<Prior art> In general, fuzzy rules used in fuzzy controllers are, for example, If x,=A & x2==B, then
It is expressed as y=C, and the antecedent part of the fuzzy rule is ``F~'', and the consequent part is ``then...''.

In addition, Xl and X2 are fuzzy variables for the antecedent part, A and B are fuzzy labels for the antecedent part, y is a fuzzy variable for the consequent part, and C
is called the fuzzy label of the consequent.

Conventionally, in this type of fuzzy controller, when the fuzzy variables X1 and X2 in the antecedent part are human signals from the sensor, and the fuzzy variable y in the consequent part is a drive signal to the actuator, the "distance between the sensor and the workpiece (fuzzy If the variables X1, It is created by adjusting and stored in a file, and fuzzy inference is performed using this adjusted membership function.

<Problem to be solved by the invention> However, with the above conventional fuzzy controller,
The sensitivity of the sensor is adjusted in advance and the distance between the sensor and the workpiece is adjusted, so if the sensor is installed far away, the operator does not have to go back and forth between the sensor and the fuzzy controller many times. However, there is a problem that the adjustment work is troublesome.

In view of the above-mentioned conventional problems, the present invention outputs the goodness of fit of the membership function in the antecedent part of a fuzzy rule to a fuzzy controller in response to a human input signal from a sensor, and makes adjustments when the sensor and fuzzy controller are far apart. Our objective is to provide a fuzzy output sensor controller that can simplify your work.

Means for Solving the Problems> In order to achieve the above object, the fuzzy output sensor controller of the present invention includes means for storing membership functions of the antecedent part of fuzzy rules for fuzzy calculations, and a means for storing membership functions of the antecedent part of fuzzy rules for fuzzy calculations. The system is equipped with means for adjusting the membership function determined by the sensor, and means for calculating the degree of fitness of the membership function according to the human power signal from the sensor and outputting the result to the fuzzy controller.

According to the present invention, with the above configuration, by installing this fuzzy output sensor controller close to the sensor, the sensitivity of the sensor can be adjusted, and the antecedent part of the fuzzy rule can be adjusted while adjusting the distance between the sensor and the workpiece. The membership function can be adjusted, thus simplifying the adjustment task when the sensor and fuzzy controller are far apart.

<Example> Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the fuzzy output sensor controller according to the present invention, FIG. 2 is a front view showing the fuzzy output sensor controller of FIG. 1, and FIG.
Figure 4 is an explanatory diagram showing the memory contents of the RAM in Figure 4. Figure 4 is an explanatory diagram showing the relationship between the human power level of the sensor and the fitness of the membership function of the antecedent. Figures 5 and 6 are Figure 1. FIG. 7 is a flowchart for explaining the operation of the fuzzy output sensor controller, and is an explanatory diagram showing various output forms for the fuzzy controller.

In Fig. 1, 1 is an A/D converter that converts the signal X from the sensor S (Fig. 2) into a digital signal, and 2 is a central processing unit (CPU 4) as shown in Figs. As shown, it controls the setting of the membership function of the antecedent part of the fuzzy rule for fuzzy calculations, calculates the fitness of the membership function according to the input signal from the sensor, and controls the output to the fuzzy controller. A read-only memory (ROM) 3 stores programs for the CPU 4 in advance, and 3 is a random access memory (RAM) having a work area for the CPU 4 and the like.

As shown in Fig. 3, RAM3 stores sensor S and workpiece (
For each label of the antecedent part of the fuzzy rule indicating that the distance of (not shown) is "far", "medium", or "close", the minimum human power level MIN of the sensor S for which the value of the membership function is positive.
, a data table for storing the maximum input level MAX of the sensor S whose membership function value is "1", and a data table for storing the maximum input level MAX of the sensor S whose membership function value is positive. A label pointer area indicating which table of the data tables for each label such as ``Nearest'' and ``Nearest'' is to be set, and an area rMINJ of each data table.

It has a level pointer area that indicates in which area of rMIDJ and rMAXJ data is to be set.

Note that the label pointer indicates the labels "far", "medium", and "near" in the case of rOJ "1" and "2", respectively, and the level pointer indicates the labels "far", "medium", and "near", respectively, and the level pointer indicates "0", "1", and "2". ” indicates the human power level MIN, MID, and MAX, respectively.

Returning to Fig. 1, 5 is determined by the human power level X of the sensor S.
a setting switch for setting the membership function of the antecedent part of the fuzzy rule for fuzzy operation in RAM3;
Reference numeral 6 denotes a setting/operation changeover switch for switching between the above setting mode or a mode in which the fitness degree y of the membership function is calculated according to the input signal from the sensor and outputted to the fuzzy controller.

7 is a label selection switch for selecting the above-mentioned "far", "medium", and "near" labels, and 8 is each adaptation point A, B, of the membership function, as shown in FIG. Input levels MIN, MID, and sensor S of sensor S corresponding to C, respectively.
This is a level selection switch for selecting MAX.

9 is the label selected by the label selection switch 7;
The label display LED 10 for displaying "far", "medium", and "near" is a level display LED for displaying the input level MIN, MID, and MAX selected by the level selection switch 8; Switches 5-8 and indicator LEDs 9.10 are arranged at the front of the housing of the device, as shown in FIG.

11 is a D/A for converting the fitness degree y of the antecedent part of the fuzzy rule calculated by the CPU 4 into a digital signal and outputting it to the fuzzy controller 20 as described later.
The fuzzy controller 20, which is a converter, performs fuzzy inference based on this fitness degree y.

Next, the operation of the above embodiment will be explained.

First, with reference to FIG. 5, we will explain the operation when setting the input levels MIN, MID, and MAX of the sensor S for each of the labels "far", "medium", and "near" in the antecedent part of the fuzzy rule. do.

In the setting mode shown in FIG. 5, in step 51,
First, both the label pointer and the level pointer of the RAM 3 are reset to "O", and in the following step 52, it is determined whether or not the switches 5 to 8 have been pressed. In this case, since both the label pointer and the level pointer are "○", the label display LED 9 is set to the label "○".
"Far" is selected, and the level display LE
D10 indicates that the human power level MIN is selected.

In step 52, when the setting switch 5 is pressed, the process proceeds to step 53, where the human power level set in the area.

Further, when the level selection switch 8 is pressed in step 52, the process proceeds to step 55, where the level pointer is incremented by "1", and in the subsequent step 56, it is determined whether the value of the level pointer is "3" or more. do. If it is "3" or more, the process returns to step 51; if it is not "3" or more, the process returns to step 52.

Therefore, the operator can adjust the sensitivity of the sensor S,
By adjusting the distance between the sensor S and the workpiece and pressing the level selection switch 8, the manual power levels MIN, MID, and MAX for the label "Far" can be set.

In step 52, when the label selection switch 7 is pressed, the process proceeds to step 57 and the label pointer is set to "1".
In the following step 58, it is determined whether the value of the label pointer is "3" or more. "3
If the number is "2" or more, the process returns to step 51, and if the number is not "3" or more, the process returns to step 52.

Further, in step 52, when an operation mode is selected by the setting/operation changeover switch 6, the setting mode ends and the operation mode shifts to the operation mode shown in FIG.

Therefore, according to the above embodiment, the operator can connect the sensor S and the workpiece (
The minimum human power level MIN of the sensor S for which the value of the membership function is positive for each label of the antecedent part of the fuzzy rule indicating that the distance (not shown) is "far", "medium", or "close";
The input level MID of the sensor S at which the value of the membership function is "1" and the maximum human power level MAX of the sensor S at which the value of the membership function is positive can be adjusted and set as appropriate.

Next, referring to FIG. 6, an explanation will be given of the operation when calculating the degree of conformity y of the antecedent part of the fuzzy rule and outputting it to the fuzzy controller 20.

0 In step 61, setting/operation changeover switch 6
When the operation mode is selected by step 62, the input level X of sensor S is taken in, and step 6 follows.
3, the input levels MIN, MID, M set in RAM3 for each label "far", "medium", "near"
Compare with AX.

If the comparison result is x<MIH, proceed to step 64 and set the degree of conformity y to rOJ; if MIN≦X-MID, proceed to step 65 to calculate the degree of conformity y using the following formula, and set MID≦ In the case of X≦MAX, the process proceeds to step 66 to calculate the degree of conformity y using the following equation, and in the case of MAX<x, the degree of conformity y is set to rOJ. Next, in step 68, this fitness degree y is outputted to the fuzzy controller 20 via the D/A converter 11.

■ Therefore, according to Example 1, F A (Fac
When the sensor S and the fuzzy controller 20 are installed apart from each other, as in the case of the fuzzy output sensor controller 20, by installing this fuzzy output sensor controller close to the sensor S, you can perform fuzzy calculations while checking the output signal of the sensor S. The goodness of fit y of the antecedent part for can be adjusted.

Furthermore, according to the above embodiment, the optimal membership function corresponding to the sensor S can be set in this fuzzy output sensor controller. Membership functions can be set more easily than when set in a fuzzy controller.

Incidentally, as a method of outputting the fitness degree y of the antecedent part for the fuzzy calculation to the fuzzy controller 20, the method shown in FIG. 7 (a
), the suitability y for each label "distant", "medium", and "near" is output via each data line, or as shown in Fig. 7 (
As shown in b), the label selection from the fuzzy controller 20 is performed via the common output data line according to the 2nd signal or the 7th
As shown in FIG. 7(c), the degree of suitability y of each label can be frequency modulated and output, or as shown in FIG. 7(d), the label information and the degree of suitability y can be output in parallel.

<Effects of the Invention> As explained above, the present invention provides a means for storing a membership function of an antecedent part of a fuzzy rule for a fuzzy operation, a means for adjusting the stored membership function, and a sensor. The sensor sensitivity can be increased by installing this fuzzy output sensor controller close to the sensor. The membership function of the antecedent part of the fuzzy rule can be adjusted while making the adjustment, and therefore the adjustment work can be simplified when the sensor and the fuzzy controller are separated.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the fuzzy output sensor controller according to the present invention, and FIG. 2 is a front view showing the fuzzy output sensor controller of FIG. 1.
FIG. 3 is an explanatory diagram showing the memory contents of the RAM in FIG. FIG. 6 is a flowchart for explaining the operation of the fuzzy output sensor controller in FIG. 1, and FIG. 7 is an explanatory diagram showing various output forms for the fuzzy controller. 1...A/D converter, 2...Read only memory (
ROM), 3...Random access memory, 4...Central processing unit (CPU), 5...Setting switch, 6...Setting/operation changeover switch, 7...Label selection switch,
8...Level selection switch, 9...Label display LED
, 10...Level display LED, 11...D/A
Converter, 20...Fuzzy controller, S sensor.

Claims (1)

[Claims] Means for storing a membership function of an antecedent part of a fuzzy rule for a fuzzy operation, means for adjusting the membership function stored in the storage means, and according to an input signal from a sensor. a fuzzy output sensor controller, comprising means for calculating the fitness of the membership function and outputting it to a fuzzy controller.