Attitude and Altitude Control of a Quadrotor UAV

OData support
Dr. Stumpf Péter Pál
Department of Automation and Applied Informatics

Over the past few decades research and development activity in the area of aircraft control has gone through a significant alteration. The main principle - formulated during World War II i.e. assisting the pilot via advanced electronics implementing stability control and navigation assistance - has remained the same, except that in the meantime the pilot disappeared from the system, or more precisely: the pilot was taken outside of the system. Thus a new military field has born and the development of UAVs (Unmanned Aerial Vehicles) has begun. The role of the pilot still remained important, but the task was more like guidance and supervision rather than full-time human control.

This phenomenon has become even more significant over the past few years, and the center of gravity is noticeably shifting towards total automation, decreasing the pilot’s overall influence. Another driving reason is the continuously growing public interest about RC helicopters and aircrafts, indicating the fact, that the originally military purpose UAVs are slowly emerging into the civil sector.

These remote-controlled machines can be utilized in several - peaceful - ways, some of which are still being recognized, thus often being under market-research phases, but the proposed potential uses are very promising. The presented ideas are often in connection with smart-home, or life-saving emergency services in hard-to-approach environments e.g. in mountains. For this very reason, continuously increasing researcher activity can be observed in the field of UAV and MAV (Micro Aerial Vehicle) controls.

This final-project has a shared purpose. On the one hand I would like to demonstrate, that in this era – thanks to the advancement of on-board computation and sensor technology - the results of modern control theory can be effectively applied to realize attitude and altitude control of a quadrotor MAV helicopter by means of recently published nonlinear control algorithms. On the other hand I would like to give guidance for those who are interested in such applications, and seek a fast and effective way to implement and test controllers, which are non-conventional and thus require further experiments.


Please sign in to download the files of this thesis.