Comparison of Fuzzy Optimization and Genetic Fuzzy Methods in Solving a Modified Traveling Salesman Problem

Abstract

There are a growing number of applications demonstrating the effectiveness of emulating human decision making using fuzzy logic. Main research challenges include situational awareness and decision making in an uncertain spatio-temporal environment. In this effort, a MATLAB simulation of a surveillance environment was created that placed targets in random areas on a map, with each target having a circular area imposed around it. In the simulation, a fuzzy robot was to find the shortest path around the environment, where it touched each target area at least once (meaning the areas can be passed through) before returning to its starting position. Through fuzzy optimization of a path produced through a genetic algorithm, this task was completed and it was shown that a shorter path could be found through the fuzzy optimization. The project could then be further optimized through created straight line optimization, and was made more realistic with Dubins paths. Finally, this optimization was compared to a created genetic fuzzy algorithm to determine differences in accuracy and precision between the two solutions.