Whether dealing with an orbital system or a balloon-borne pointing system, the presence of complex dynamics and multiple uncertainties makes the design and validation of attitude control laws all the more challenging, especially as the required performance levels — aimed at improving system accuracy, availability, and autonomy — are increasingly demanding. In the first case, such complex dynamics are often induced by propellant sloshing in the tanks or by the presence of deployment mechanisms on board the satellite. In the second case, they arise from the interactions between the uncontrolled or poorly controlled motions of the balloon and the gondola carrying the pointing system, both of which are subject to atmospheric disturbances. 

More specifically, azimuth pointing is constrained by the flexibility of the flight chain on which the pivot motor acts, by the nonlinearities of the motor at low amplitude, by the inertia of the balloon that limits feasible motions, and by the difficulty of predicting disturbances, particularly those caused by an auxiliary balloon in the flight chain. 

These difficulties combine with various other effects (in particular, the flexibility of the actuation system), leading to control issues on the other axes as well, and rendering overly simple attitude control architectures ineffective. Furthermore, in all the applications considered here, accurate modeling of the phenomena remains out of reach, which necessitates the use of simplified models that introduce multiple uncertainties.

In this context, following an extensive modeling phase of the space systems under study, this doctoral project will focus on the design of attitude control laws by combining different techniques (robust, LPV, adaptive control, etc.) that explicitly account for system properties. The handling of actuator saturation and complementarity (reaction wheels, gyros, pivot motors, fast steering mirrors, etc.) will rely on online adaptation strategies of the reference signals (Reference Governors) and control allocation schemes, which will be the subject of dedicated investigations.

Attention will then turn to the validation of the developed control laws, particularly in the presence of uncertainties not accounted for during the design phase. To this end, robustness analysis techniques will be implemented and adapted, with specific developments to address the non-stationary nature of the closed loop. During this analysis phase, optimization-based searches for worst-case configurations may also be employed to improve the design.

An experimental component will complement this doctoral work. The main application foreseen is a test-bench demonstration at CNES, aimed at illustrating the management of interactions between a balloon and a fine pointing system, with special emphasis on azimuth pointing.

=================

For more Information about the topics and the co-financial partner (found by the lab!); contact Directeur de thèse - clement.roos@onera.fr

Then, prepare a resume, a recent transcript and a reference letter from your M2 supervisor/ engineering school director and you will be ready to apply online  before March 13th, 2026 Midnight Paris time!