Reactive real time behavior for mobile robots in unknown environments

Abstract

The objective is to emulate the stimulus-response behavior for robot navigation in unknown environments. The robot's objective is to move from its current position to some predefined coordinates, namely, a predefined goal point. The proposed behavior is reactive and the robot has no prior information about the size and location of the obstacles in the environment. This control algorithm is a combination of two main behaviors: wall following and goal seeking, and is based totally on real-time data acquisition from immediate surroundings of the robot. A carefully designed fuzzy inference system (FlS) is used to determine the change in velocity and direction of the robot based on the sonar sensor and inertial cube (Gyro) sensor readings.