Fuzzy Logic in Swift
I’ve determined the reason that I cannot play video games at length anymore. Every time I play a game, it leads me into a new development project and I get sidetracked from actually playing the game. Most recently, I was tinkering around with Viva Piñata, an old favorite of mine, and I thought to myself “maybe I can recreate the logic used by the piñata animals”. Of course, I don’t really know how the developers created everything, but I assumed maybe a fuzzy logic system is being used to determine animal behavior. So, I built my own fuzzy login system in Swift. You can contribute to the project on Github.
The system is built using a few primitives:
- Point
- Line
- FuzzySet
- FuzzyVariable
Point
A struct representing a point in 2D-space.
Line
A struct representing a line segment. It is composed of a slope, minimum x-value, maximum x-value, and the line’s functional representation. The functional representation is a closure that can used to produce a y-value in the format of y = f(x).
FuzzySet
A struct that contains a collection of Line objects (line segments) which form a shape that can be represented on a 2D graph/chart. The main purpose of this struct is to generate membership values (percentages) based on a value related to a FuzzyVariable.
FuzzyVariable
A struct that contains a collection of FuzzySet objects. Given a value, one can calculate the degree of membership for all associated FuzzySet objects.