Vision based robot navigation

Abstract

In this thesis we propose a vision-based robot navigation system that constructs a high level topological representation of the world. A robot using this system learns to recognize rooms and spaces by building a hidden Markov model of the environment. Motion planning is performed by doing bidirectional heuristic search with a discrete set of actions that account for the robot's nonholonomic constraints. The intent of this project is to create a system that allows a robot to be able to explore and to navigate in a wide variety of environments in a way that facilitates goal-oriented tasks.