Projects and products of the department MRD

A project of the same name aims to harness our experience across a whole host of AI application domains to realize MLOps with maximum effectiveness and automate it as far as possible. To that end a software suite was developed that incorporates a wide range of tools for creating machine learning applications and training the ML models.

Digitally supported and modular refurbishment of apartment buildings in neighborhoods with PVT-PCM heat pump systems and self-learning energy management

Economic efficiency, sustainability, ecology and individuality: Intermodal mobility - the use of different modes of transport on one route - has great potential. But the complexity must remain manageable. In other words: Getting to your destination intermodally, for example with a combination of (rental) bike, public transport and e-scooter, must be as easy and pleasant as reaching for your own car key.

Digital technologies and platform for the forecasting and operational optimization of drinking water supply systems.

A holistic optimization of district heating supply areas offers great potential, for example, through the feed-in of renewable energies, network expansion, energy savings and cost reduction, as well as a reduction in CO2 emissions. The aim is to initially make synergies visible by exchanging and combining data/information that has not been exchanged before, and ultimately to generate added value for as many participants as possible.

The NaviBox offers advanced autonomy functions and is ideal for anyone who wants to retrofit their vehicles or robots with autonomy functions.

Fraunhofer IOSB has developed an autonomous surface vehicle that autonomously takes measurements above and below the water to generate associated 3D maps.

The RLT-Opt project is developing new methods and tools to enable the inventory and savings potential to be quickly recorded in the first step, so that corresponding optimization measures can be implemented in a targeted manner in the second step. The first step is to analyze the tightness and operating mode of the ventilation and air conditioning systems. The operation of the ventilation and air conditioning systems is successively optimized by sensor-based recording and evaluation of essential operating parameters.

The Competence center “Robot systems for decontamination in hazardous environments” is the research center for autonomous or semi-autonomous robot systems, which are to be used in practical applications where work has so far been associated with a considerable risk potential for humans.

The global objective of the ML4Heat project is the development of methods and software tools for the optimization of the operation of existing district heating networks under energetic and economic aspects. For this purpose, sensor and operating data of the district heating transfer stations as well as the heat supply shall be collected to a large extent and evaluated by means of machine learning methods.

Das Projekt CC-King hat sich zum Ziel gesetzt, KI-Spitzenforschung mit etablierten Ingenieurdisziplinen zu verbinden. Hierdurch sollen KI-Methoden und maschinelle Lernverfahren (ML) entsprechend den typischen Anforderungen und Vorgehensweisen von Ingenieuren nutzbar gemacht werden.

Multisensory system for AI-based drone detection and classification with situation visualization.

The Karlsruher Forschungsfabrik for AI-integrated production offers the special opportunity to research and develop the topic of “Intelligent Production” in a practical way on real processes.

“AutoInspect” is a system for continuous object assessment (in real time) by multimodal inspection along the production cycle. Both a direct reaction to the results and their long-term observation are possible. The system is supported by a digital quality twin with open I4.0 standards.

The dynOpt energy manager is primarily intended for commercial properties such as apartment buildings and neighbourhoods where several components such as heat pumps, PV systems, solar heating systems, gas boilers or CHP units are operated in a coupled manner. Furthermore, dynOpt-En monitors energy consumption and costs and compares the real values with the expected values. In this way, malfunctions can be detected promptly and signalled via a clear visualisation or reported to the operator.

For the early detection of health-threatening contaminations in drinking water, therefore, an online-capable broadband test procedure for drinking water conformity is needed that reacts very quickly and reliably, is robust against false alarms, can be operated by persons without scientific qualifications and whose acquisition and maintenance costs are economically justifiable. Currently, such online-capable and broadband sensor systems are not commercially available.

The production of bulk and fine chemicals on the basis of renewable raw materials has become increasingly important in recent years as "white biotechnology". Surfactants, which are currently produced industrially to a large extent from petrochemical starting materials, are a potential product for the use of biotechnological production processes.

In order to keep the total operating costs of wind turbines (WTGs) competitive, the risk of failure must be minimized, maintenance costs reduced, and system availability and energy efficiency increased. This goal is achieved by introducing the most efficient, automated multi-sensory online monitoring and diagnostic systems - so-called Condition Monitoring Systems (CMS) - whose economic importance is increasingly recognized by wind farm operators, manufacturers and insurers.

Process engineering processes usually have a large number of process variables which generate an alarm when a previously defined threshold value is exceeded in the event of abnormal process behavior. If a fault occurs in a central plant section, this leads to a flood of alarms, since the individual process variables are interdependent and the fault propagates through the plant.

By means of the developed and integrated data management and above all thanks to the advanced data evaluation, it is possible to carry out systematic data evaluation as well as plant monitoring, thus increasing the efficiency of the production process.

In the Fraunhofer ML4P lead project, several Fraunhofer Institutes are pooling their application experience and machine learning skills to develop solutions for industry. The aim is to develop a tool-supported process model that paves the way for flexible, fast-learning machines and systems.

The HeatSteel project is developing innovative and energy-efficient heat treatment lines for ultra-thin precision strip.

Water supply networks are exposed to a constant danger of deliberate or accidental contamination. For this reason, the existing Drinking Water Ordinance requires regular sampling for certain bacteria and chemical substances. However, the existing procedures are very time consuming due to testing in special laboratories.

Das europäische FP7-Projekt SAFEWATER zielt darauf ab, eine umfassende Lösung zur Ereigniserkennung und Ereignisverwaltung für das Management der Trinkwassersicherheit und die Abschwächung gegen große absichtliche, zufällige oder natürliche CBRN-bezogene Kontaminationen zu entwickeln und insbesondere um

When operating heating, ventilation and air-conditioning systems in existing commercial buildings, it is often found that both the heat and cold supply and the individual room controls are very conservatively parameterised and the rooms are not heated or cooled according to demand. Permanent monitoring and optimisation by specialised personnel is usually not profitable. Consistent and cost-effective solutions for existing buildings are not available in the desired form. The project therefore aims to realise an integrated value chain for efficient monitoring and optimised operational management of properties from the building stock with low-investment measures.

The W-Net 4.0 project aims to develop a modular and scalable platform that combines GIS system, simulation software and data analysis tools and meets high IT security standards. Combined with novel service concepts, value-added networks and training concepts, small and medium-sized water supply companies will be enabled to use these technologies for the first time. For large utilities, novel and easy-to-use data analysis and optimization tools as well as corresponding service concepts will be made available.

The ResiWater project pursues the goal of enabling utilities to optimally secure the drinking water supply even in exceptional situations and to facilitate the restart of the system to normal operation after a crisis situation by choosing appropriate measures.

The overarching project objectives are to increase the use of renewable energies in district heating and to make the operation of district heating networks more efficient.

In this cooperation project, an intelligent system is to be designed and implemented as a prototype that enables the rapid detection and precise localization of leaks in district heating networks based on data and, based on this, calculates response strategies for shutting off network sections while ensuring an optimal supply to end customers.

The idea of Team ARGGONAUTS, with which Fraunhofer IOSB competed in the Shell Ocean Discovery XPRIZE technology competition and finished among the top 5: autonomous catamarans tow lightweight diving drones into the field, where they single-handedly map the seabed at depths of up to 4000 metres.