Development of a Machine Vision based Localization and Mapping System for Mobile Robots

OData support
Supervisor:
Kovács Viktor
Department of Automation and Applied Informatics

My Bachelor thesis is a software designing process, which describes the steps I had to go through from the idea to the realization, creating the final measuring and visu-alization system.

The system is a 5G mobile communication - robotic sample application develo-ped at the BME Department of Automation and Applied Informatics. The base system was convoy of car-like robots moving on a fixed course placed on a table. The convoy members communicate with the central PC, where the image processing of the camera module and the high-level controling algorithm run. My tasks are to maintain the system, and the camera measuring system to recognize 3D obstacles and be compatible with ROS enviroment.

The previous robot-level collision-avoidance extended with height-based map-ping, which model was sent to the central unit to calculate the obstacle-avoidance cont-rol. The camera-system provides the depth images to produce the map model.

To process the map-data, instead of the modification of the old visualisation system, it should be replaced with a new visualisation software in addititon for future upgrades.

Downloads

Please sign in to download the files of this thesis.