PhD scholarship: Fully automated controller synthesis for UAV missions (Munich, Germany)

Abstraction-based controller synthesis is a relatively recent approach whose key advantage over classical synthesis methods is that it permits to solve control problems for nonlinear continuous-state plants described by ordinary differential equations in a fully automated, correct-by-construction fashion, even for rather complex control objectives and in the presence of uncertainties and disturbances. The approach relies on finite-state approximations (''abstractions'') of continuous-state plants described by ordinary differential equations.
These abstractions are used as substitutes of the original plants in the synthesis process, to reduce the synthesis problem to the solution of an auxiliary, purely discrete synthesis problem. The  stipendiary is expected to advance both theory and computational methods to facilitate the practical applicability of the approach to the synthesis of controllers enforcing temporal logic specifications on UAVs. The focus is on synthesis algorithms that are both efficient and formally correct, and on the reduction of the complexity of the resulting controllers. The project involves theoretical work, algorithm and software development, and, on a small scale, experimental work.


See also:

University of the Federal Armed Forces Munich
Department of Aerospace Engineering
LRT-15, Institute of Control Engineering
D-85577 Neubiberg (Munich)