Fuzzy preferences in multi-objective optimization (MOO)
AssigneeHonda Research Institute Europe GmbH
InventorsYaochu Jin, Bernhard Sendhoff
Life of this patent
Estimating the term…
Abstract
A method to obtain the Pareto solutions that are specified by human preferences is suggested. The main idea is to convert the fuzzy preferences into interval-based weights. With the help of the dynamically-weighted aggregation method, it is shown to be successful to find the preferred solutions on two test functions with a convex Pareto front. Compared to the method described in “Use of Preferences for GA-based Multi-Objective Optimization” (Proceedings of 1999 Genetic and Evolutionary Computation Conference, pp. 1504-1510, 1999) by Cvetkovic et al., the method according to the invention is able to find a number of solutions instead of only one, given a set of fuzzy preferences over different objectives. This is consistent with the motivation of fuzzy logic.
Claims9
A method for multi-objective optimization of a mechanical, aerodynamic or hydrodynamic body using evolutionary algorithms, the method comprising the steps of:
- (a) encoding object parameters of the mechanical, aerodynamic or hydrodynamic body to be optimized as individuals;
- (b) setting up an initial population of the individuals as parents;
- (c) reproducing a plurality of offspring individuals from the parents, the individuals representing object parameters to be optimized;
- (d) evaluating quality of the offspring individuals using a fitness function comprising a sum of weighted sub-functions, each weighted sub-function corresponding to an objective of the multiple objective optimization;
- (e) selecting the one or more offspring individuals having the highest evaluated quality value as parents for a next evolution cycle;
- (f) changing weights of the weighted sub-functions for the next cycle within predetermined ranges, a first weight of the weighted sub-functions changing within a first predetermined range, a second weight of the weighted sub-functions changing within a second predetermined range different from the first predetermined range, the first predetermined range and the second predetermined range representing preferences given to objectives of the multiple objective optimization;
- (g) repeating steps (c) to (f) until a termination criterion is met;
- and
- (h) outputting one or more offspring individuals after the termination criterion is met as the optimized object parameters of the mechanical, aerodynamic or hydrodynamic body.
A computer product for multi-objective optimization of a mechanical, aerodynamic or hydrodynamic body, the computer program product comprising a computer readable storage medium structured to store instructions executable by a processor, the instructions, when executed cause the processor to:
- (a) encode object parameters of the mechanical, aerodynamic or hydrodynamic body to be optimized as individuals;
- (b) set up an initial population of the individual as parents;
- (c) reproduce a plurality of offspring individuals from the parents, the individuals representing object parameters to be optimized;
- (d) evaluate quality of the offspring individuals using a fitness function comprising a sum of weighted sub-functions, each weighted sub-function corresponding to an objective of the multiple objective optimization;
- (e) select the one or more offspring individuals having the highest evaluated quality value as parents for a next evolution cycle;
- (f) change weights of the weighted sub-functions for the next cycle within predetermined ranges, a first weight of the weighted sub-functions changing within a first predetermined range, a second weight of the weighted sub-functions changing within a second predetermined range different from the first predetermined range, the first predetermined range and the second predetermined range representing preferences given to objectives of the multiple objective optimization;
- (g) repeat steps (a) to (f) until a termination criterion is met;
- and
- (h) output one or more offspring individuals after the termination criterion is met as the optimized object parameters of the mechanical, aerodynamic or hydrodynamic body.
Dependencies are inferred from the claim text.