Optics and Sensing Laboratory

The Optics and Sensing Laboratory supports research and development of new algorithms for image processing, enhancement and denoising, the integration of sensing systems of different types (stereo, video, range, etc.), new sensors and algorithms for computational photography and astronomy, and the development of real-time sensing systems for robotic platforms.

What we do

As a service group, we provide cross institute research support for capturing systems and optical equipment.

We also work and collaborate with diverse research projects in the field of psychophysiology and optics.

Sensing -

We are focused on research for the development of multi-sensor data capturing and analysis methods for generating large scale data-sets. Today, we have established a large-scale capturing facility with hundreds of cameras and further expanding to portable multi-camera systems to capture animals in their natural habitat. We also develop new algorithms for the development of computer vision solutions in psychology and psychotherapy to quantitatively evaluate the effectiveness of psychotherapy methods.

Optics -

The group also specializes in the assembly and operation of advanced optical systems, supporting cutting-edge research. Furthermore, we develop and integrate sophisticated light sources tailored for specific research applications. In collaboration with other groups we conduct original research focused on the design of novel sensors that harness optical principles, enhanced with machine learning algorithms. Additionally, we actively conduct research in the field of optical computing, pushing the boundaries of computational capabilities through light-based technologies.

Members:

Senya Polikovsky

Optics & Sensing Laboratory

	Tübingen

	+49 7071 601 1799

Tathagata Bandyopadhyay

Research Intern

	Tübingen

Karin Bierig

Research Technician

	Tübingen

	+49 7071 601 597

Cristóbal Curio

Guest Scientist

	Tübingen

Maike Eisen

	Tübingen

Gökce Ergün

	Tübingen

	1713

Ashutosh Jha

	Tübingen

Keiko Kitagawa

	Tübingen

	+49 7071 6011709

Johannes Kopf-Beck

	Tübingen

Timo Lübbing

	Tübingen

Tonmoy Rakshit

Hiwi

	Tübingen

	+49 7071 601

Carolin Schmitt


	+49 7071 601 1770

Valentin Volchkov

Optics & Sensing Laboratory


	+49 7071 601 1772

Xucong Zhang

	Tübingen

Alumni:

Fiber-based shape sensor

2021-09-29

Grassroots initiative into developing a novel shape probe made out of a multimode optical fiber.

26 results

2024

Reinforcement learning in cold atom experiments

Reinschmidt, M., Fortágh, J., Günther, A., Volchkov, V.

nature communications, 15:8532, October 2024 (article)

Abstract

Cold atom traps are at the heart of many quantum applications in science and technology. The preparation and control of atomic clouds involves complex optimization processes, that could be supported and accelerated by machine learning. In this work, we introduce reinforcement learning to cold atom experiments and demonstrate a flexible and adaptive approach to control a magneto-optical trap. Instead of following a set of predetermined rules to accomplish a specific task, the objectives are defined by a reward function. This approach not only optimizes the cooling of atoms just as an experi- mentalist would do, but also enables new operational modes such as the preparation of pre-defined numbers of atoms in a cloud. The machine control is trained to be robust against external perturbations and able to react to situations not seen during the training. Finally, we show that the time con- suming training can be performed in-silico using a generic simulation and demonstrate successful transfer to the real world experiment.

link (url) DOI [BibTex]

2024

Reinschmidt, M., Fortágh, J., Günther, A., Volchkov, V. Reinforcement learning in cold atom experiments nature communications, 15:8532, October 2024 (article)

link (url) DOI [BibTex]

Fiber-Optic Shape Sensing Using Neural Networks Operating on Multispecklegrams

Cao, C. G. L., Javot, B., Bhattarai, S., Bierig, K., Oreshnikov, I., Volchkov, V. V.

IEEE Sensors Journal, 24(17):27532-27540, September 2024 (article)

Abstract

Application of machine learning techniques on fiber speckle images to infer fiber deformation allows the use of an unmodified multimode fiber to act as a shape sensor. This approach eliminates the need for complex fiber design or construction (e.g., Bragg gratings and time-of-flight). Prior work in shape determination using neural networks trained on a finite number of possible fiber shapes (formulated as a classification task), or trained on a few continuous degrees of freedom, has been limited to reconstruction of fiber shapes only one bend at a time. Furthermore, generalization to shapes that were not used in training is challenging. Our innovative approach improves generalization capabilities, using computer vision-assisted parameterization of the actual fiber shape to provide a ground truth, and multiple specklegrams per fiber shape obtained by controlling the input field. Results from experimenting with several neural network architectures, shape parameterization, number of inputs, and specklegram resolution show that fiber shapes with multiple bends can be accurately predicted. Our approach is able to generalize to new shapes that were not in the training set. This approach of end-to-end training on parameterized ground truth opens new avenues for fiber-optic sensor applications. We publish the datasets used for training and validation, as well as an out-of-distribution (OOD) test set, and encourage interested readers to access these datasets for their own model development.

Optics and Sensing Laboratory

Members:

Alumni:

2024

2024

2023

2023

2022

2022

2021

2021

2020

2020

2019

2019

2018

2018

2017

2017

2016

2016

2015

2015

2014

2014

2013

2013

2012

2012

2011

2011

2010

2010