Transfer Learning for Hyperspectral Images01 Feb 2019 The conventional transfer learning approach to solve computer vision problems with Deep Learning relies on large generic supervised datasets (e.g. ImageNet), which are not necessarily available for non-RGB imagery (i.e. HSI, single channel medical, radar, etc. imagery). We are investigating ways to allow transferring RGB pre-trained networks ...

Basalt fibre reinforced geopolymer cement28 Jun 2018 Both geopolymeric cement and basalt fibres present pomising in-situ resources for construction on the lunar surface. Currently we investigate the combination of materials and its compatibility with additive manufacturing towards resource efficiency and applicability for lunar surface missions.

Computation in out-of-equilibrium quantum systems01 Apr 2018 The idea of using quantum mechanics to perform some computation much faster than with known classical algorithms originates from the eighties. The most straightforward method for that would be to build a general-purpose programmable quantum computer, which will be a challenging task of the 21st century.

Super-resolution image reconstruction for PROBA-V satellite01 Mar 2018 Due to theoretical, practical, and cost constraints, every satellite imaging system produces images with limited spatial resolution. Nevertheless, it is possible to enhance the resolution of the images acquired by means of image post-processing algorithms. The ACT is studying the possibility to apply such techniques to the PROBA-V satellite.

Melanin radiation-protection01 Mar 2018 Exposure to ionizing radiation is a show-stopper for human space missions. Melanin may provide a way to protect from ionizing radiation.

Near term application of solar power satellites for robotic Lunar exploration15 Jan 2018 Solar power satellites (SPS) have been studied extensively over the last 50 years as a potential solution for meeting terrestrial energy demand sustainably. Can they be applied to nearer term applications enabling lunar robotic exploration?

Symmetries, decoherence and the quantum to classical transition01 Jan 2018 Decoherence is our best physical mechanism to understand the quantum to classical transition. The goal of this project is to have a better understanding of the relationship between symmetries and which pointer states will emerge after a decoherence process.

Inducing Sleep Spindles with Auditory Closed-Loop Stimulation15 Dec 2017 We investigate auditory stimulation as a method to induce a brain oscillation that has been hypothesized to have a sleep-protective function. This might be an effective countermeasure for observed sleep disturbances in space.

Super-resolution for astronomical applications01 Jun 2017 Every telescope system has a limited resolution - if two or more celestial objects lie too close to each other in the sky, they will form a single dot in the image, making them impossible to distinguish from one another. The aim of this project is to look for novel aproaches to improve the resolution of telescopes beyond their classical limit.

G&C Networks - Deep architectures for real time optimal actions01 Jan 2017 Recent research on deep learning, a set of machine learning techniques able to learn deep architectures, has shown how robotic perception and action greatly benefits from these techniques. In terms of spacecraft navigation and control system, this suggests that deep architectures may be considered now to drive all or part of ...

Differential Intelligence01 Oct 2016 Many of the most celebrated techniques in Artificial Intelligence would not be as successful if, at some level, they were not exploiting differentiable quantities. The back-propagation algorithm, at the center of learning in artificial neural networks, leverages on the first and (sometimes) second order differentials of the error to update the n...

Radiation-Hardness Components at scaled technology nodes01 Feb 2016 The space radiation environment is a key challenge for all types of space missions particularly because it can affect the operation of spacecraft electronics leading to performance degradation or even catastrophic failure.