Search

With the digitalization of health care systems, artificial intelligence becomes more present in medicine. Especially machine learning shows great potential for complex tasks such as time series classification, usually at the cost of transparency and comprehensibility. This leads to a lack of trust by humans and thus hinders its active usage. Explainable artificial intelligence tries to close this gap by providing insight into the decision-making process, the actual usefulness of its different methods is however unclear. This paper proposes a user study based evaluation of the explanation method Grad-CAM with application to a neural network for the classification of breaths in time series neonatal ventilation data. We present the perceived usefulness of the explainability method by different stakeholders, exposing the difficulty to achieve actual transparency and the wish for more in-depth explanations by many of the participants., Comment: \c{opyright} 2024 The authors. This work has been accepted to IFAC for publication under a Creative Commons Licence CC-BY-NC-ND. Accepted for the 12th IFAC Symposium on Biological and Medical Systems. 6 pages, 7 figures

IEC 60848 GRAFCET is a standardized, graphical specification language for control functions. Because of the semiformal nature of IEC 60848, the details of specifications created with GRAFCET can be interpreted in different ways, possibly leading to faulty implementations. These ambiguities have been partially addressed in existing literature, but solved in different manners. Based on a literature review, this work aims at providing an overview of existing interpretations and, based on that, proposes a comprehensive interpretation algorithm for IEC 60848, which takes all relevant ambiguities from the literature review into account., Comment: \c{opyright} 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

The graphical modeling language GRAFCET is used as a formal specification language in industrial control design. This paper proposes a structural analysis that approximates the variable values of GRAFCET to allow verification on specification level. GRAFCET has different elements resulting in concurrent behavior, which in general results in a large state space for analyses like model checking. The proposed analysis approach approximates that state space and takes into consideration the entire set of GRAFCET elements leading to concurrent behavior. The analysis consists of two parts: We present an algorithm analyzing concurrent steps to approximate the step variables and we adapt analysis means from the field of Petri nets to approximate internal and output variables. The proposed approach is evaluated using an industrial-sized example to demonstrate that the analysis is capable of verifying behavioral errors and is not limited by the specification size of practical plants., Comment: \c{opyright} 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

The graphical modeling language GRAFCET is used as a formal specification language in industrial control design. This paper proposes a static analysis approach based on the control flow of GRAFCET using abstract interpretation to allow verification on specification level. GRAFCET has different elements leading to concurrent behavior, which in general results in a large state space. To get precise results and reduce the state space, we propose an analysis suitable for GRAFCET instances without concurrent behavior. We point out how to check for the absence of concurrency and present a flow-sensitive analysis for these GRAFCET instances. The proposed approach is evaluated on an industrial-sized example., Comment: \c{opyright} 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

This paper demonstrates a system comprised of infrared beacons and a camera equipped with an optical band-pass filter. Our system can reliably detect and identify individual beacons at 100m distance regardless of lighting conditions. We describe the camera and beacon design as well as the image processing pipeline in detail. In our experiments, we investigate and demonstrate the ability of the system to recognize our beacons in both daytime and nighttime conditions. High precision localization is a key enabler for automated vehicles but remains unsolved, despite strong recent improvements. Our low-cost, infrastructure-based approach is a potential step towards solving the localization problem. All datasets are made available here https://embedded.rwth-aachen.de/doku.php?id=forschung:mobility:infralocalization:concept., Comment: Accepted at IEEE Intelligent Vehicle 2022

This paper introduces our Cyber-Physical Mobility Lab (CPM Lab). It is an open-source development environment for networked and autonomous vehicles with focus on networked decision-making, trajectory planning, and control. The CPM Lab hosts 20 physical model-scale vehicles ({\mu}Cars) which we can seamlessly extend by unlimited simulated vehicles. The code and construction plans are publicly available to enable rebuilding the CPM Lab. Our four-layered architecture enables the seamless use of the same software in simulations and in experiments without any further adaptions. A Data Distribution Service (DDS) based middleware allows adapting the number of vehicles during experiments in a seamless manner. The middleware is also responsible for synchronizing all entities following a logical execution time approach to achieve determinism and reproducibility of experiments. This approach makes the CPM Lab a unique platform for rapid functional prototyping of networked decision-making algorithms. The CPM Lab allows researchers as well as students from different disciplines to see their ideas developing into reality. We demonstrate its capabilities using two example experiments. We are working on a remote access to the CPM Lab via a webinterface., Comment: This work has been presented on ECC21

This paper presents the $\mathrm{\mu}$Car, a 1:18 model-scale vehicle with Ackermann steering geometry developed for experiments in networked and autonomous driving in research and education. The vehicle is open source, moderately costed and highly flexible, which allows for many applications. It is equipped with an inertial measurement unit and an odometer and obtains its pose via WLAN from an indoor positioning system. The two supported operating modes for controlling the vehicle are (1) computing control inputs on external hardware, transmitting them via WLAN and applying received inputs to the actuators and (2) transmitting a reference trajectory via WLAN, which is then followed by a controller running on the onboard Raspberry Pi Zero W. The design allows identical vehicles to be used at the same time in order to conduct experiments with a large amount of networked agents., Comment: This work has been accepted to IFAC for publication under a Creative Commons Licence CC-BY-NC-ND

We propose a novel external indoor positioning system that computes the position and orientation of multiple model-scale vehicles. For this purpose, we use a camera mounted at a height of 3.3m and LEDs attached to each vehicle. We reach an accuracy of about 1.1 cm for the position and around 0.6 {\deg} for the orientation in the mean. Our system is real-time capable with a soft deadline of 20 ms. Moreover, it is robust against changing lighting conditions and reflections., Comment: Accepted in IFAC Wolrd Congress 2020

We discuss challenges to control systems engineering arising from the advent of cyber-physical systems (CPS). After discussing the terminology, general, IT-related issues are treated which need cooperation with computer science, in particular software engineering. Then we study those challenges that require specific core competencies from control systems engineering. We sketch solution approaches for the exemplary problem of dealing with changes in the physical environment of a CPS. ---- Der Beitrag befasst sich mit den methodischen Herausforderungen, die durch die Verbreitung der Cyber-Physical Systems (CPS) in der Automatisierungstechnik entstehen, und stellt L\"osungsans\"atze vor. Nach einer Behandlung des Begriffs CPS werden zun\"achst die allgemeinen, IT-bezogenen Fragestellungen angesprochen, die gemeinsam mit der Informatik gel\"ost werden m\"ussen. Danach gehen wir auf die Herausforderungen ein, deren Behandlung spezifisch automatisierungstechnische Kernkompetenzen erfordern und skizzieren f\"ur eine beispielhafte Problemstellung, den Umgang mit \"Anderungen in der physikalischen Umgebung, wie entsprechende L\"osungen aussehen k\"onnen., Comment: 12 pages, in German, 4 figures

Concurrently accessing shared data without locking is usually a subject to race conditions resulting in inconsistent or corrupted data. However, there are programs operating correctly without locking by exploiting the atomicity of certain operations on a specific hardware. In this paper, we describe how to precisely analyze lockless microcontroller C programs with interrupts by taking the hardware architecture into account. We evaluate this technique in an octagon-based value range analysis using access-based localization to increase efficiency., Comment: In Proceedings SSV 2012, arXiv:1211.5873