The purpose of my thesis is to investigate the application of soft computing
methods to solve the navigation problem of a mobile robot moving in an unknown
environment. In my thesis I designed a fuzzy-based controller, that is able to
guide the mobile robot from an initial position to a final. During the motion,
the robot avoids all of the obstacles that are located on the map. I used
genetic algorithms to tune the parameters of the fuzzy controller. The
simulation environment was implemented using MATLAB and Simulink.
The first chapter is a brief summary of the simulation environment, I briefly
describe each of its components. The second chapter is a theoretical overview of
the fuzzy logic. The third chapter deals with implementation details of two
crutial components: the fuzzy steering controller, and the fuzzy speed
controller. In the fourth chapter I explain how I solved the local minima
problem that I discovered while testing the controllers. In the fifth chapter,
I explain how I applied genetic algorithms to tune the fuzzy steering
controller. The thesis ends with the evaluation of the simulation results.