A simple and lightweight fuzzy logic engine written in Java 8 and published under the MIT licence, notably without utilising any third party libraries. Currently, the binary size of the JAR file is approximately 45 KB.
The engine provides a convenient fluent API that lets you model the reference system in a easy way.
Controller Type Mamdani
Membership Functions Triangle, Trapezoid
Reasoning Scheme Max-Min Composition is used
Defuzzifier Center of Mass
/**
* A simple reference system is given, which models the brake behaviour of a car driver
* depending on the car speed. The inference machine should determine the brake force for a
* given car speed. The speed is specified by the two linguistic terms 'low' and 'medium', and
* the brake force by 'moderate' and 'strong'.
*/
@Test
public void testCar() {
FuzzyModel model = model().name("car")
.vars(lv().usage("input")
.name("carSpeed")
.terms(triangle().name("low")
.start(20)
.top(60)
.end(100),
triangle().name("medium")
.start(60)
.top(100)
.end(140)),
lv().usage("output")
.name("brakeForce")
.terms(triangle().name("moderate")
.start(40)
.top(60)
.end(80),
triangle().name("strong")
.start(70)
.top(85)
.end(100)))
.rules("if carSpeed is low then brakeForce is moderate",
"if carSpeed is medium then brakeForce is strong");
FuzzyEngine engine = new FuzzyEngine(model);
OutputVariable output = engine.evaluate(new InputVariable("carSpeed", 70));
// test output value
assertEquals(65.9939, output.getValue(), 0.01);
System.out.println(output);
}If you would like to compute output values for a range of input values then do the following
for (int i = 0; i < 50; ++i) {
double speed = 20 + i * (120.0 / 50);
InputVariable input = new InputVariable("carSpeed", speed);
OutputVariable output = engine.evaluate(input);
System.out.println(engine.printResult(input, output, 6, 2));
}See also examples in fuzzy/src/test/java/ch/x01/fuzzy/api/FuzzyEngineTest.java
To build the project with Maven from the command line go to the directory fuzzy and run
mvn clean install