A fuzzy-genetic robust optimization framework for UAV conceptual design

Date of Publication


Document Type

Master's Thesis

Degree Name

Master of Science in Mechanical Engineering


Gokongwei College of Engineering


Mechanical Engineering

Thesis Adviser

Gerardo L. Augusto

Defense Panel Chair

Laurence A. Gan Lim

Defense Panel Member

Aristotle T. Ubando
Arvin H. Fernando
Laurence A. Gan Lim
Jonathan R. Dungca


A fuzzy-genetic robust optimization framework was used in the conceptual design of a fixed-wing battery-powered propeller-driven Unmanned Aerial Vehicle. This framework essentially views the design process as a multi-objective optimization problem. In particular, the design process was formulated as a 4-objective, 17-design variable, 16-constraint optimization problem arising from a model that describes certain performance and stability parameters as functions of aircraft geometry, aerodynamics, propulsion, and weight parameters. A real-coded genetic algorithm was used as a solution tool, considering the complexity of the problem and the nature of the design parameters. Fuzzy logic was used to evaluate the satisfaction of targets such as objectives and constraints, avoiding the unwarranted imposition of crisp criteria on a low-fidelity model, which is a typical element of the conceptual design phase. Fuzzy logic was also used to define fuzzy-Pareto dominance, which replaced conventional Pareto dominance in the ranking of individuals, with the intention of increasing selection pressure in the search for the fuzzy-Pareto front. Principles of robust design were also integrated into the algorithm to avert the pitfalls of a deterministic optima of an approximate model. Robustness of objectives was sought by accounting for the mean and standard deviation of variations of objectives resulting from variations in the design variables generated by a Monte Carlo simulation. The fuzzy-genetic algorithm developed can compete against NSGA-II and is much faster than it, making the integration of a Monte Carlo simulation more viable. Such integration can considerably increase run time but the use of a surrogate can resolve that problem. The robust fuzzy-genetic algorithm developed was shown to be able to produce results that closely match design parameters already embodied by an existing aircraft.

Abstract Format






Accession Number


Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

1 computer disc ; 4 3/4 in.


Drone aircraft; Vehicles; Remotely piloted

This document is currently not available here.