Faites connaître cette offre !
Reference : UMR8109-YANCLE-008
Workplace : MEUDON
Date of publication : Wednesday, April 29, 2020
Scientific Responsible name : Yann Clénet
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 October 2020
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly
Description of the thesis topic
Adaptive optics compensates in real time for disturbances affecting the quality of astronomical images thanks to a deformable mirror inserted in the optical path. The mirror is controlled from measurements made on a guide star by a wave front sensor.
The proposed PhD project concerns the Single Conjugate Adaptive Optics (SCAO) system of the MICADO instrument intended to equip the European Extremely Large Telescope (ELT), the first of which light is expected by the end of 2025.
MICADO, one of the two 1st light instruments of the ELT, is a near infrared camera with spectroscopic capabilities and working at the resolution limit of the telescope with a field of view of approximately one arcminute. Its design is focused on very high sensitivity and very high astrometric precision. With its wide field, its high angular resolution, its great astrometric precision and a remarkable sensitivity, MICADO will thus have the capacities to scan a wide range of astrophysical subjects.
The MICADO instrument is coupled to a multi-conjugate AO system but also to a conventional SCAO system with wavefront sensing on a natural guide star. In its initial phase, MICADO must be able to operate with the SCAO system only. This on-axis AO system, for which LESIA is responsible, must be controlled to be able to compensate for major disturbances affecting the quality of astronomical images.
The disturbances which impact on the AO system include in particular the catch in the wind of the immense structure of the ELT which will make oscillate and tremble the telescope (windshake) with an amplitude much higher than the degradation of the image quality due to atmospheric turbulence alone. To this will be added various sources of excitation such as the motors, fans, cryo-generators present around the instrument, and which will also induce mechanical vibrations in the structures. The whole will have an optical effect mainly on the tip / tilt aberrations, but perhaps also on a limited number of other optical modes of relatively low spatial frequencies, known as “low order” modes. Perceived by the MICADO SCAO system, these low order disturbances will degrade its performance unacceptably if they are not effectively compensated.
Low-order model-based control loops have already been developed and implemented for the current generation of AO systems. These are in particular optimal tip-tilt mode commands (minimizing the variance of the residual disturbance after correction) of LQG type (Linear Quadratic Gaussian). These commands include a Kalman filter which predicts the disturbance a few milliseconds in advance. An control of this type is currently used, for example, in the tip-tilt loop of the AO of the SPHERE exoplanet detection instrument at the VLT. The design of such regulators requires a thorough knowledge of the AO system, in order to propose a relevant modeling on which the Kalman filter will be built. The proposed PhD project follows this logic, with three main objectives:
1) Develop a specific strategy for the control of the SCAO of MICADO, taking into account the fact that the optical "high orders" will be compensated by a simple control, of pure integrator type
2) Develop the corresponding algorithms, while respecting three requirements: efficiency of the correction, compatibility with the constraints of the real-time calculator, and robustness vis-à-vis the very large variety and variability of the observation conditions. This latter specification will require the implementation of strategies to identify and re-identify disturbance models during operation
3) Experimentally test the selected control algorithms on an optical bench, then if the results are conclusive conduct tests as part of a demonstration on the sky carried out at the William Herschel Telescope installed in the Canary Islands
The design of the control law will have to answer new questions that have not been addressed so far. In particular, it will have to integrate the specific features of the “pyramid” wave front sensor used by the MICAO SCAO loop, and in particular its highly non-linear response. It should also take into account the interactions with the other servo loops of the telescope. This includes for example taking into account the temporal response of the large deformable mirror of about 2.5 m in diameter.
The first step is to understand the functioning of an adaptive optics system equipped with an integrator type regulator. The study of the LQG regulators proposed so far will then allow to put ourselves in the state of the art. Taking into account the constraints of the AO system will lead to propose candidate control algorithms. After getting started with the AO COMPASS simulation code (developed and maintained by LESIA), it will involve implementing the proposed algorithms, evaluating and optimizing their performance. The selected control algorithms can then be tested and validated experimentally, first on optical benches at IOGS and LESIA then on the William Herschel telescope. The constraints imposed by the operation of the telescope itself will finally have to be studied, and complete strategies will then be developed and evaluated in simulation.
This thesis will take place in the “Adaptive Optics” teams of the Charles Fabry Laboratory (LCF, CNRS-Institut d'Optique Graduate School, Palaiseau) and the Laboratory d'Etudes Spatiales (LESIA, CNRS-Observatoire de Paris, Meudon). It will also be carried out in collaboration with the team developing the ELT telescope control at the European Southern Observatory (ESO) in Garching (Germany).
The candidate must hold a Master degree or an engineering degree in the fields of optics or astronomical instrumentation (automatic control skills will be appreciated but not required) or automatic. The candidate is interested in the application aspects of the thesis, ranging from simulations to experimentation. He is motivated by the prospect of participating in the ELT adventure and enjoys working in interaction with different teams.
Constraints and risks
We talk about it on Twitter!