LQGNet: Hybrid Model-Based and Data-Driven Linear Quadratic Stochastic Control
Authors:
Solomon Goldgraber Casspi,
Oliver Husser,
Guy Revach,
Nir Shlezinger
Abstract:
Stochastic control deals with finding an optimal control signal for a dynamical system in a setting with uncertainty, playing a key role in numerous applications. The linear quadratic Gaussian (LQG) is a widely-used setting, where the system dynamics is represented as a linear Gaussian statespace (SS) model, and the objective function is quadratic. For this setting, the optimal controller is obtai…
▽ More
Stochastic control deals with finding an optimal control signal for a dynamical system in a setting with uncertainty, playing a key role in numerous applications. The linear quadratic Gaussian (LQG) is a widely-used setting, where the system dynamics is represented as a linear Gaussian statespace (SS) model, and the objective function is quadratic. For this setting, the optimal controller is obtained in closed form by the separation principle. However, in practice, the underlying system dynamics often cannot be faithfully captured by a fully known linear Gaussian SS model, limiting its performance. Here, we present LQGNet, a stochastic controller that leverages data to operate under partially known dynamics. LQGNet augments the state tracking module of separation-based control with a dedicated trainable algorithm. The resulting system preserves the operation of classic LQG control while learning to cope with partially known SS models without having to fully identify the dynamics. We empirically show that LQGNet outperforms classic stochastic control by overcoming mismatched SS models.
△ Less
Submitted 24 October, 2022; v1 submitted 23 October, 2022;
originally announced October 2022.
OGLE2-TR-L9: An extrasolar planet transiting a fast-rotating F3 star
Authors:
I. A. G. Snellen,
J. Koppenhoefer,
R. F. J. van der Burg,
S. Dreizler,
J. Greiner,
M. D. J. de Hoon,
T. O. Husser,
T. Kruhler,
R. P. Saglia,
F. N. Vuijsje
Abstract:
Context: Photometric observations for the OGLE-II microlens monitoring campaign have been taken in the period 1997-2000. All light curves of this campaign have recently been made public. Our analysis of these data has revealed 13 low-amplitude transiting objects among ~15700 stars in three Carina fields towards the galactic disk. One of these objects, OGLE2-TR-L9 (P~2.5 days), turned out to be a…
▽ More
Context: Photometric observations for the OGLE-II microlens monitoring campaign have been taken in the period 1997-2000. All light curves of this campaign have recently been made public. Our analysis of these data has revealed 13 low-amplitude transiting objects among ~15700 stars in three Carina fields towards the galactic disk. One of these objects, OGLE2-TR-L9 (P~2.5 days), turned out to be an excellent transiting planet candidate.
Aims: In this paper we report on our investigation of the true nature of OGLE2-TR-L9, by re-observing the photometric transit with the aim to determine the transit parameters at high precision, and by spectroscopic observations, to estimate the properties of the host star, and to determine the mass of the transiting object through radial velocity measurements.
Methods: High precision photometric observations have been obtained in g', r', i', and z' band simultaneously, using the new GROND detector, mounted on the MPI/ESO 2.2m telescope at La Silla. Eight epochs of high-dispersion spectroscopic observations were obtained using the fiber-fed FLAMES/UVES Echelle spectrograph, mounted on ESO's Very Large Telescope at Paranal.
Results: The photometric transit, now more than 7 years after the last OGLE-II observations, was re-discovered only ~8 minutes from its predicted time. The primary object is a fast rotating F3 star, with vsini=39.33+-0.38 km/s, T=6933+-58 K, log g = 4.25+-0.01, and [Fe/H] = -0.05+-0.20. The transiting object is an extrasolar planet with M_p=4.5+-1.5 M_Jup and R_p=1.61+-0.04 R_Jup. The rejection of possible blend scenarios was based on a quantitative analysis of the multi-color photometric data [abridged].
△ Less
Submitted 3 December, 2008;
originally announced December 2008.
