An Introduction to Type-2 Fuzzy Sets and Systems Dr Simon Coupland simonc@dmu.ac.uk Centre for Computational Intelligence De Montfort University Leicester United Kingdom www.cci.dmu.ac.uk Contents My background Motivation Interval Type-2 Fuzzy Sets and Systems Generalised Type-2 Fuzzy Sets and Systems An Example Application – Mobile Robotics My Background Research Fellow from the UK Here on a collaborative grant with Prof. Keller Worked in type-2 fuzzy logic for 5 years Awarded PhD “Geometric Type-2 Fuzzy Systems” in 2006 Working on: Computational problems of generalised type-2 fuzzy logic Applications My Background Created and maintain type2fuzzylogic.org Information, experts, publications (~450), news and events ~600 members ~70 countries Type-2 Publications Type-1 Fuzzy Sets Extend crisp sets, where x A or x A Membership is a continuous grade [0,1] Describe vagueness – not uncertainty (Klir and Yuan) Why do we need type-2 fuzzy sets? Type-1 fuzzy sets do not model uncertainty: 1 Tall 0.62 0 1.8 Height (m) Why do we need type-2 fuzzy sets? So, a person x, who’s height is 1.8 metres is Tall to degree 0.62 (Tall(1.8) = 0.62) Improvement on Tall or not Tall Vagueness, but no uncertainty How do we model uncertainty? Why do we need type-2 fuzzy sets? We need, x is Tall to degree about 0.62 But how to model about 0.62? Two schools of thought: Interval type-2 fuzzy sets – about 0.62 is a crisp interval Generalised type-2 fuzzy sets – about 0.62 is a fuzzy set Run blurring example Interval Type-2 Fuzzy Sets Interval type-2 fuzzy sets - interval membership grades ~ A = {((x,u), 1) | x X, u Jx, Jx [0,1]} X is primary domain Jx is the secondary domain All secondary grades (A~(x,u)) equal 1 Fully characterised by upper and lower membership functions (Mendel and John) Interval Type-2 Fuzzy Sets Returning to Tall ~ Tall 1 Upper MF Tall Type -1 MF = FOU 0 Height (m) Lower MF Tall Interval Type-2 Fuzzy Sets Fuzzification: ~ 1 Tall 0.78 ~ (1.8) = [0.42,0.78] Tall 0.42 0 1.8 Height (m) Interval Type-2 Fuzzy Sets Defuzzification – two stages: Type-reduction Interval centroid Type-reduction (centroid): GC = 1Jx … 1Jx 1 N 1 Ni=1 xii / N i=1 i = [Cl, Cr] (Karnik and Mendel) Interval Type-2 Fuzzy Sets Only need to identify two embedded fuzzy sets Only Jx1 and JxN will belong to those sets Identify two ‘switch points’ on X Switch point against X is a convex function Mendel and Liu showed switch point = C where {l,r} Interval Type-2 Fuzzy Sets Defuzzification: ~ 1 Tall 0 Cl Cr Height (m) Centroid Interval Type-2 Fuzzy Sets Cl Cl switch point X Interval Type-2 Fuzzy Sets Centroid Cr Cr switch point X Interval Type-2 Fuzzy Sets These properties are exploited by KarnikMendel algorithm Converges in at most N steps 3-4 steps typical Widely used Hardware implementation Interval Type-2 Fuzzy Systems Rules Output processing Defuzzifier Crisp inputs Fuzzifier Type-reducer Inference Type-2 Interval FIS Crisp outputs Typereduced outputs (interval) Interval Type-2 Fuzzy Systems Mamdani or TSK systems We’ll only look at Mamdani Example rule base: 1. 2. ~ ~ ~ If x is A and y is B then z is G1 ~ ~ ~ If x is C and y is D then z is G2 Interval Type-2 Fuzzy Systems Antecedent calculation: Rule 1: RA1 = [A~ (x) B~ (y), ~A(x) B~ (y)] Rule 2: RA2 = [C~ (x) D~ (y), ~C(x) ~D(y)] where is a t-norm, generally min or prod Interval Type-2 Fuzzy Systems Consequent calculation: Rule 1: G 1 = i..n[G~ 1(zi) RA1, G~ 1(zi) RA1] ~’ Rule 2: G 2 = i..n[G~2(zi) RA2, G~ 2(zi) RA1] ~’ Interval Type-2 Fuzzy Systems Consequent combination: Gc = i..n [G~1’ (gi) V G~2’ (gi) , G~1’ (gi) V G~2’ (gi) ] ~ Where V is a t-conorm, generally max Interval Type-2 Fuzzy Systems A~ 1 0 1 0 ~ C B~ 1 0 1 0 ~ D 1 0 1 0 ~ G 1 ~ G 2 Interval Type-2 Fuzzy Systems A~ 1 0 ~ C 1 0 B~ 1 1 0 1 ~ D 1 0 x 0 0 y ~ G 1 ~ G 2 Interval Type-2 Fuzzy Systems ~ ~ (min) ~ A B G 1 1 1 0 ~ C 1 0 0 1 1 ~ D 1 0 x 0 0 y ~ G 2 Interval Type-2 Fuzzy Systems ~ ~ (min) ~ A B G 1 1 1 0 ~ C 1 0 0 1 1 ~ D 1 0 x 0 0 y ~ G 2 Interval Type-2 Fuzzy Systems ~ ~ (min) ~ A B G 1 1 1 0 ~ C 1 0 0 1 1 ~ D 1 0 x 0 0 1 y ~ G 2 ~ GC max 0 Cl Cr Interval Type-2 Fuzzy Systems ~ ~ (prod) ~ A B G 1 1 1 0 ~ C 1 0 0 1 1 ~ D ~ G 2 1 0 x 0 0 ~ GC 1 y max 0 Cl Cr Interval Type-2 Fuzzy Systems Summary: Membership grades are crisp intervals Two parallel type-1 systems (up to defuzzification) Defuzzification in two stages: Type-reduction (KM) Defuzzification Generalised Type-2 Fuzzy Sets Generalised type-2 fuzzy sets – type-1 fuzzy numbers for membership grades ~ A = {((x,u), A~ (x,u)) | x X, u Jx, Jx [0,1]} X is primary domain Jx is the secondary domain A~ (x) is the secondary membership function at x (vertical slice representation) All secondary grades (~A(x,u)) [0,1] Generalised Type-2 Fuzzy Sets Representation theorem (Mendel and John) Represent generalised type-2 fuzzy sets and operations as collection of embedded fuzzy sets ~ A= n Ae ~j j=1 ~ Ae = {(x, (u, A~ (x,u)) | x X, u Jx, Jx [0,1]} Only used for theoretical working (to date) Generalised Type-2 Fuzzy Sets Fuzzification (x) (x,u) 1 1 ~ A X Generalised Type-2 Fuzzy Sets Fuzzification (x) (x,u) 1 1 ~ A x X Generalised Type-2 Fuzzy Sets Fuzzification (x) (x,u) 1 1 ~ A x X Generalised Type-2 Fuzzy Sets Fuzzification (x) (x,u) 1 1 ~ A x (x,u) 1 ~A(x) (x) X 1 Generalised Type-2 Fuzzy Sets Antecedent ‘and’ – the meet Two SMF’s: f = i / vi and g = j / wj The meet: f g= i j / vi wj (Zadeh) Generalised Type-2 Fuzzy Sets Antecedent ‘or’ – the join Two SMF’s: f = i / vi and g = j / wj The join: f g= i j / vi V wj (Zadeh) Generalised Type-2 Fuzzy Sets (x,u) 1 Join and meet under min: f (x,u) 1 g meet join (x) 1 (x) 1 Generalised Type-2 Fuzzy Sets (x,u) 1 Join and meet under prod: f (x,u) 1 g meet join (x) 1 (x) 1 Generalised Type-2 Fuzzy Sets More efficient join and meet operations: Apex points 1 and 2 (x,u) 1 f 1 g 2 1 (x) Generalised Type-2 Fuzzy Sets More efficient join and meet operations: f g (u) = f g (u) = { { f(u) Λ g(u), u<1 f(1) Λ g(u), 1u<2 (f(u) V g(u)) Λ (f(1) Λ g(2)), u2 (f(u) V g(u)) Λ (f(1) Λ g(2)), u<1 f(u) Λ g(2), 1u<2 f(u) Λ g(u), u2 (Karnik and Mendel), (Coupland and John) Generalised Type-2 Fuzzy Sets Implication: Meet every point in consequent with antecedent value: ~A(x) ~B(y) G = (~A(x) ~B(y)) ~ zZ ~ (z) G Generalised Type-2 Fuzzy Sets Combination of Consequents: Join all consequent sets at every point in the in the consequent domain: G = G~1(z) G~2(z) ~ zZ … G~n(z) Generalised Type-2 Fuzzy Sets Type-reduction (centroid) Gives a type-1 fuzzy set: GC = 1Jz … 1Jz 1 N N i=1 G(zii) ~ / N i=1 zii Ni=1 i (Karnik and Mendel) Generalised Type-2 Fuzzy Sets Type-reduction (z) (z,u) 1 1 Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 1 (z,u) 1 Z Z Generalised Type-2 Fuzzy Sets Type-reduction (z) 1 CZ 1 (z,u) 1 Z Z CZ Show again Generalised Type-2 Fuzzy Sets Type-reduction – number embedded sets: Generalised Type-2 Fuzzy Sets Computational complexity is a huge problem Inference complexity relates to join and meet Type-reduction is not a sensible approach Generalised Type-2 Fuzzy Sets Geometric approach (Coupland and John): Model membership functions as geometric objects Operations become geometric Run geometric model Generalised Type-2 Fuzzy Sets Let the generalised type-2 fuzzy set A consist of n triangles: Generalised Type-2 Fuzzy Sets The centroid of A is the weighted average of the area and centroid of each triangle: Generalised Type-2 Fuzzy Sets The centroid of a triangle is the mean of the x component of the three vertices The area of a triangle is half the cross product of any two edge vectors Generalised Type-2 Fuzzy Sets Generalised Type-2 Fuzzy Sets Generalised Type-2 Fuzzy Sets Generalised Type-2 Fuzzy Sets Criticisms: No ‘measure of uncertainty’ Problems with rotational symmetry On the plus side: Computes in a reasonable time Interesting potential implementations Generalised Type-2 Fuzzy Sets Summary: Rich model – membership grades are fuzzy numbers High computational complexity Inference problems solved Type-reduction partly solved (geometric approach) Generalised Type-2 Fuzzy Sets Applications: Control: Signal Processing: Robot navigation (Hagras, Coupland, Castillo) Plant (Castillo, Chaoui, Hsiao) Classification (Mendel, John, Liang) Prediction (Rhee, Mendez, Castillo) Perceptual reasoning: Perceptual computing (Mendel) Modelling perceptions (John) Generalised Type-2 Fuzzy Sets Example Application: Robot control and navigation: Generalised Type-2 Fuzzy Sets Example Application: Robot control and navigation: Summary Type-1 fuzzy sets can’t model uncertainty Interval type-2 fuzzy sets – crisp interval Generalised type-2 fuzzy sets – fuzzy set Interval systems fast, simple computation Generalised – high computational complexity Outperformed type-1 – growing applications Further Reading http://www.type2fuzzylogic.org/ Uncertain Rule-Based Fuzzy Logic Systems: Introduction and New Directions Mendel, J.M. http://www.cse.dmu.ac.uk/~simonc/eldertech/ http://www.cci.dmu.ac.uk/