Performance Comparison of Fuzzy Logic and Neural Network Design for Mobile Robot Navigation

Abstract

The mobile robot is the type of robot that emerges not only in industry but also in the domestic application, intended to substitute or assist human in a dull, dirty, or dangerous environment. The robot is designed to imitate or resemble human abilities to perform a physical task using a simple control theorem, or even sophisticated task by implementing artificial intelligent (AI) to create a smart robot. The most applied AI is Fuzzy Logic Controller (FLC) and Neural Network (NN). The main issue in the mobile robot is the navigation, defined as how to ensure the robot can finish the task safely without crushing to any obstacles. This paper investigates the application of FLC and NN in robot navigation and compares the performance in navigating the robot to the target. Sensors used in this paper is distance sensors and a camera. A robot is moved in several experimental setting, and the effectiveness of FLC and NN application is compared. The comparison is conducted in a simulation program named MobotSim, where several robots were designed in various environments. The simulation results show that NN application is more suitable confirmed by faster time in completing the task.