@misc{13879,
  abstract     = {{Model Predictive Control evolved as the state of the art paradigm for safety critical control tasks. Control-as-Inference approaches thereof model the constrained optimization problem as a probabilistic inference problem. The constraints have to be implemented into the inference model. A recently introduced physics-informed Gaussian Process method uses Control-as-Inference with a Gaussian likelihood for state constraint modeling, but lacks guarantees of open-loop constraint satisfaction. We mitigate the lack of guarantees via an additional sampling step using Hamiltonian Monte Carlo sampling in order to obtain safe rollouts of the open-loop dynamics which are then used to obtain an approximation of the truncated normal distribution which has full probability mass in the safe area. We provide formal guarantees of constraint satisfaction while maintaining the ODE structure of the Gaussian Process on a discretized grid. Moreover, we show that we are able to perform optimization of a quadratic cost function by closed form Gaussian Process computations only and introduce the Matérn kernel into the inference model.}},
  author       = {{Tebbe, Jörn and Besginow, Andreas and Lange-Hegermann, Markus}},
  booktitle    = {{European Journal of Control}},
  issn         = {{1435-5671}},
  publisher    = {{Elsevier BV}},
  title        = {{{Physics-informed Gaussian Processes as Linear Model Predictive Controller with Constraint Satisfaction}}},
  doi          = {{10.1016/j.ejcon.2026.101600}},
  year         = {{2026}},
}

@misc{12804,
  abstract     = {{Data in many applications follows systems of Ordinary Differential Equations (ODEs). This paper presents a novel algorithmic and symbolic construction for covariance functions of Gaussian Processes (GPs) with realizations strictly following a system of linear homogeneous ODEs with constant coefficients, which we call LODE-GPs. Introducing this strong inductive bias into a GP improves modelling of such data. Using smith normal form algorithms, a symbolic technique, we overcome two current restrictions in the state of the art: (1) the need for certain uniqueness conditions in the set of solutions, typically assumed in classical ODE solvers and their probabilistic counterparts, and (2) the restriction to controllable systems, typically assumed when encoding differential equations in covariance functions. We show the effectiveness of LODE-GPs in a number of experiments, for example learning physically interpretable parameters by maximizing the likelihood.}},
  author       = {{Besginow, Andreas and Lange-Hegermann, Markus}},
  booktitle    = {{36th Conference on Neural Information Processing Systems (NeurIPS 2022) }},
  editor       = {{Koyejo, S. and Mohamed, S. and Agarwal, A. and Belgrave, D. and Cho, K. and Oh, A.}},
  isbn         = {{978-1-7138-7108-8 }},
  issn         = {{1049-5258}},
  keywords     = {{SMITH NORMAL-FORM, ALGORITHMS, REDUCTION}},
  location     = {{New Orleans, La.; Online}},
  pages        = {{29386 -- 29399}},
  publisher    = {{Curran Associates, Inc.}},
  title        = {{{Constraining Gaussian Processes to Systems of Linear Ordinary Differential Equations}}},
  volume       = {{35}},
  year         = {{2022}},
}

@inproceedings{4097,
  abstract     = {{The capabilities of object detection are well known, but many projects don’t use them, despite potential benefit. Even though the use of object detection algorithms is facilitated through frameworks and publications, a big issue is the creation of the necessary training data. To tackle this issue, this work shows the design and evaluation of a prototype, which allows users to create synthetic datasets for object detection in images. The prototype is evaluated using YOLOv3 as the underlying detector and shows that the generated datasets are equally good in quality as manually created data. This encourages a wide adoption of object detection algorithms in different areas, since image creation and labeling is often the most time consuming step.}},
  author       = {{Besginow, Andreas and Büttner, Sebastian and Röcker, Carsten}},
  booktitle    = {{22nd International Conference on Human-Computer Interaction}},
  isbn         = {{978-3-030-50343-7}},
  keywords     = {{Object detection, Synthetic datasets, Machine learning, Deep learning}},
  location     = {{Copenhagen, Denmark}},
  pages        = {{178--192}},
  publisher    = {{Springer}},
  title        = {{{Making Object Detection Available to Everyone - A Hardware Prototype for Semi-automatic Synthetic Data Generation}}},
  doi          = {{https://doi.org/10.1007/978-3-030-50344-4_14}},
  volume       = {{12203}},
  year         = {{2020}},
}

@inproceedings{4180,
  abstract     = {{In this paper we give an overview of features and use cases that Intelligent Adaptive Assistance Systems (IAAS) in the literature commonly provide. For this, a literature research has been executed where 29 papers were selected for inspection. In the course of this inspection, most common features are noted, compared and assessed against the definitions we gave for an IAAS. It showed that the development of IAAS can benefit from an intensified research in cooperation with machine learning experts to further develop the intelligence and adaptivity of future IAAS.}},
  author       = {{Besginow, Andreas and Büttner, Sebastian and Röcker, Carsten}},
  booktitle    = {{5. Workshop zu Smart Factories: Mitarbeiter-zentrierte Informationssysteme für die Zusammenarbeit der Zukunft, Mensch und Computer 2018}},
  location     = {{Dresden}},
  publisher    = {{Gesellschaft für Informatik e.V.}},
  title        = {{{Intelligent Adaptive Assistance Systems in an Industrial Context – Overview of Use Cases and Features}}},
  doi          = {{10.18420/muc2018-ws18-0533}},
  year         = {{2018}},
}

@inproceedings{4181,
  abstract     = {{Projection-based Augmented Reality (AR) might change the interactions with digital systems in future work environments. A lot of stationary projection-based AR assistive systems have been presented that might support future work processes. However, not much research has been done beyond stationary settings. With moving towards mobile settings, fast and robust object recognition algorithms are required that allow real-time tracking of physical objects as targets for the projected digital overlay. With this work, we present a portable projection-based AR platform that recognizes objects in real time and overlays physical objects with in-situ projections of digital content. We consider our system as a precursor to a future mobile projection-based assistive system. By presenting the system, we want to start a discussion in the HCI community about the potential of mobile projection-based AR in future work environments.}},
  author       = {{Büttner, Sebastian and Besginow, Andreas and Prilla, Michael and Röcker, Carsten}},
  booktitle    = {{Workshop on Virtual and Augmented Reality in Everyday Context (VARECo), Mensch und Computer 2018}},
  location     = {{Dresden}},
  publisher    = {{Gesellschaft für Informatik e.V.}},
  title        = {{{Mobile Projection-based Augmented Reality in Work Environments – an Exploratory Approach}}},
  doi          = {{10.18420/muc2018-ws07-0364}},
  year         = {{2018}},
}