Co-advisor: Yasmina BESTAOUI
Grant from french Government
Date PhD finished: March 4th, 2014
Collaboration: Evry University
Current position: ATER at Evry University.
Today, the inspection of structures is carried out through visual assessments realized by qualied inspectors. This procedure is very expensive and can put the personal in dangerous situations. Consequently, the development of an unmanned aerial vehicle equipped with on-board vision systems is privileged nowadays in order to facilitate the access to unreachable zones.
In this context, the main focus in the thesis is developing original methods to deal with planning, reference trajectories generation and tracking issues by a hovering airborne platform. These methods should allow an automation of the flight in the presence of air disturbances and obstacles. Within this framework, we are interested in two kinds of aerial vehicles with hovering capacity: airship and quad-rotors.
Firstly, the mathematical representation of an aerial vehicle in the presence of wind has been realized using the second law of Newton.
Secondly, the question of trajectory generation in the presence of wind has been studied: the problem of minimal time was formulated, analyzed analytically and solved numerically.Then, a strategy of trajectory planning based on operational research approaches has been developed.
Thirdly, the problem of trajectory tracking was carried out. A nonlinear robust control law based on Lyapunov analysis has been proposed. In addition, an autopilot based on saturation functions for quad-rotor crafts has been developed. All methods and algorithms proposed in this thesis have been validated through simulations.