Stereovision Feedback and Fuzzy Control for Autonomous Robot Navigation-Edición Única

Abstract

The main objective of this thesis is the implementation of an autonomous robot that is capable of following a road w i t h the use of a stereocamera. The use of vision i n 3D is a trend i n many vision applications, and we would like to adopt it for autonomous vehicle navigation. Road following requires two crucial steps: the road recognition and the control of the speed and steering of the vehicle. To reach controllability i n these type of robots it is necessary to detect the road and make the image processing to calculate the manipulations needed to keep the robot i n the desired position and path. It can be said that vision systems are a very important factor i n the development of autonomous vehicles because they are the sensors of the robot; therefore, the image processing needs to be done on time i n order to detect the road and let the entire system to be controllable. Through this thesis, various techniques i n different areas such as road detection, image processing and fuzzy control theory will be examined i n order to achieve the objective.