Being able to fly has always been one of humanities greatest dreams, and today anyone can purchase a Radio-Controlled (RC) airplane or helicopter and learn how to fly. Experienced RC pilots perform stunts such as “prop hanging”. This is when an airplane flies vertically while maintaining its position with the propeller being the only motive force. In this thesis, the first steps towards converting a manually controlled hovering airplane to an autonomous one are taken by using one single accelerometer to measure differences in acceleration as input data for a height regulator. A built prototype with the height regulator implemented is tested in a test rig. The finished regulator is able to adjust and keep the airplane stable when exposed to small to medium disturbances. The regulator’s biggest weakness is the lack of input data regarding the velocity. Big disturbances result in a constant velocity, which gives zero acceleration input data, and an airplane flying away from its equilibrium position. / Att kunna flyga har alltid varit en av mänsklighetens största drömmar och idag kan vem som helst köpa ett radiostyrt flygplan eller helikopter och lära sig att flyga. Erfarna piloter som flyger radiostyrda flygplan utför konster som att ”hänga i propellern”, vilket innebär att flygplanet flyger vertikalt samtidigt som den behåller sin position där propellern är den enda drivkraften. I den här avhandlingen tas det första steget att konvertera ett manuellt styrt hovrande flygplan till ett autonomt genom att använda en enda accelerometer för att mäta skillnaden i acceleration som indata för en höjdregulator. Ett byggt prototypflygplan med höjdregulatorn implementerad testas i en testställning. Den färdiga regulatorn fungerar och kan justera och hålla flygplanet stabilt när den utsätts för små till medelmåttiga störningar. Regulatorns största svaghet är bristen på indata av hastigheten. Stora störningar resulterar i konstant hastighet, vilket ger noll acceleration som indata och ett flygplan som flyger ifrån sitt jämviktsläge.

Being able to fly has always been one of humanities greatestdreams, and today anyone can purchase a Radio-Controlled(RC) airplane or helicopter and learn how to fly. ExperiencedRC pilots perform stunts such as “prop hanging”.This is when an airplane flies vertically while maintainingits position with the propeller being the only motive force.In this thesis, the first steps towards converting a manuallycontrolled hovering airplane to an autonomous one aretaken by using one single accelerometer to measure differencesin acceleration as input data for a height regulator.A built prototype with the height regulator implementedis tested in a test rig. The finished regulator is able toadjust and keep the airplane stable when exposed to smallto medium disturbances. The regulator’s biggest weaknessis the lack of input data regarding the velocity. Big disturbancesresult in a constant velocity, which gives zeroacceleration input data, and an airplane flying away fromits equilibrium position. / Att kunna flyga har alltid varit en av mänsklighetens störstadrömmar och idag kan vem som helst köpa ett radiostyrtflygplan eller helikopter och lära sig att flyga. Erfarna pilotersom flyger radiostyrda flygplan utför konster som att”hänga i propellern”, vilket innebär att flygplanet flygervertikalt samtidigt som den behåller sin position där propellernär den enda drivkraften. I den här avhandlingen tasdet första steget att konvertera ett manuellt styrt hovrandeflygplan till ett autonomt genom att använda en endaaccelerometer för att mäta skillnaden i acceleration som indataför en höjdregulator. Ett byggt prototypflygplan medhöjdregulatorn implementerad testas i en testställning. Denfärdiga regulatorn fungerar och kan justera och hålla flygplanetstabilt när den utsätts för små till medelmåttigastörningar. Regulatorns största svaghet är bristen på indataav hastigheten. Stora störningar resulterar i konstanthastighet, vilket ger noll acceleration som indata och ettflygplan som flyger ifrån sitt jämviktsläge.

<p> Within recent years, there has been a renewed interest in developing data-driven learning based algorithms for solving longstanding challenging control problems. This interest is primarily motivated by the availability of ubiquitous data and an increase in computational resources of modern machines.  However, there is a prevailing concern on the lack of provable performance guarantees on data-driven/model-free learning based control algorithms. This dissertation focuses the following key aspects: i) with what facility can state-of-the-art learning-based control methods eke out successful performance for challenging flight control applications such as aerobatic maneuvering?; and ii) can we leverage well-established tools and techniques in control theory to provide some provable guarantees for different types of learning-based algorithms?  </p> <p>To these ends, a deep RL-based controller is implemented, via high-fidelity simulations, for Fixed-Wing aerobatic maneuvering. which shows the facility with which learning-control methods can eke out successful performances and further encourages the development of learning-based control algorithms with an eye towards providing provable guarantees.<br> </p> <p>Two learning-based algorithms are also developed: i) a model-free algorithm which learns a stabilizing optimal control policy for the bilinear biquadratic regulator (BBR) which solves the regulator problem with a biquadratic performance index given an unknown bilinear system; and ii) a model-free inverse reinforcement learning algorithm, called the Model-Free Stochastic inverse LQR (iLQR) algorithm, which solves a well-posed semidefinite programming optimization problem to obtain unique solutions on the linear control gain and the parameters of the quadratic performance index given zero-mean noisy optimal trajectories generated by a linear time-invariant dynamical system. Theoretical analysis and numerical results are provided to validate the effectiveness of all proposed algorithms.</p>

This master thesis solves conceptual design of aerotow airplane and airplane for basic aerobatics. Airplane is designed by using of specification CS-23. The first part of the thesis solves conceptual design. Especially possible concepts and choice the best one according to requests. Next part deals with flight performance of aerotow airplane in normal flight and also in aerobatics. The third part is engaged in main flight performance in aerotow. 3D system model of whole aircraft and compact disk with mentioned 3D model and electronic version of thesis is also added.

This master's thesis discusses the possibilities of safe-life extension program for ZLIN Z 126 and 226 airplanes, which despite their age and long-end production are still widespread and popular not only in the Czech Republic. The age of these aircraft is also related to the fact, that some aircraft are already approaching its border life, so there is an interest for the longest possible technical and legislative airworthiness. As mentioned in my bachelor's thesis, during the existence of this model series, there was no major damage occurred, except for sporadic exfoliation corrosion of wing flanges. Experience with maintenance of this aircraft points to certain underestimation of the current 5500hrs lifetime and it is obvious to suggest possible steps to increase the current limit. The thesis is therefore considered in this context the possibility of producing new wing flanges made of another material, performed the calculation for wing fatigue exchanged with flanges and designed technological process of this exchange, so that ZLIN Z 126 and 226 airplanes can be safely operated in the future.