Development of a Grid Based Path-planning Algorithm for Mobile Robots

OData support
Csorvási Gábor
Department of Automation and Applied Informatics

Robot arms have been used in the industry for various purposes for a relatively long time. These are automated and programmable but fixed to one place. Mobil robots, however, are capable of relocation. In the recent few decades, their application became common not just in the industry, but also in the everyday use. They can be found in hospitals, warehouses and in military application. Their moving can be guided with the aid of some kind of external devices (e.g.: following a line with predefined properties in a warehouse), or they can be autonomous. The latter means that they are capable of navigating in an uncontrolled environment without external aid. One subtask of navigation is path planning, which deals with the task of planning a path between A and B points, that is traversable by the robot without colliding with obstacles.

In my thesis, I focus on the subject of path planning. Firstly, I describe a few frequently used path planning algorithms. Then I present an operating system called ROS (Robot Operating System), which is commonly used in robotics. After describing the path planning system developed at the Department of Automation and Applied Informatics of Budapest University of Technology and Economics, I present the Hybrid-state A* path planner algorithm, which is included in the mentioned system. My main task focused on this planner. I describe in the 5th chapter the usage of this planner with grid-based maps, and my attempts on the development of the algorithm in the following chapter. In the 7th chapter, I present a simulation system, which provides the possibility to investigate path following with a simulated robot. At the end of my thesis, I formulate future development possibilities to the algorithm.


Please sign in to download the files of this thesis.