VIVA-SAX – Validation Funding

Title:
From the laboratory to industry: Validation of patented, innovative ultrashort-pulse laser technologies for industrial applications

Project Leader:
Prof. Dr. rer. nat. Steffen Weißmantel
, Chair of Physics/Physical Technologies

Project:
LVP

Abstract:
The project aims to validate patented ultra-short-pulse laser technologies with regard to their suitability for industrial use. It is based on several patents held by Mittweida University of Applied Sciences, which enable innovative applications in the surface treatment of solid-state materials. The process utilises ultrashort laser pulses in GHz pulse trains, which can be used for high-gloss polishing of metallic surfaces as well as for the creation of novel surface alloys and sponge-like porous structures. However, the process using conventional GHz pulse trains carries the risk of extremely hazardous X-ray radiation as a by-product of the irradiation process. This risk was previously confirmed in a previous ZIM project in collaboration with ACSYS Laser Technology GmbH and has so far represented a major hurdle to industrial application. To address this issue, a novel technology is being utilised in cooperation with Light Conversion from Vilnius, Lithuania – the manufacturer of the laser capable of generating these GHz pulse trains – which is to be validated within the project with regard to its potential for industrial application. A laser prototype is being used for this purpose; it was funded by the German Research Foundation (DFG) as part of a DFG-FH grant and is available at the Laser Institute at Mittweida University of Applied Sciences. Initial tests have identified parameter ranges in which high-gloss surfaces can be achieved without the generation of hazardous X-rays; however, instabilities currently occurring in the laser prototype are impairing the reproducibility of the results. The project therefore focuses on optimising the system stability, reproducibility and robustness of the entire laser process in order to achieve industrial readiness (target: TRL 5) in a pilot trial. This validation process is supported by letters of intent from Light Conversion and ACSYS Laser Technology GmbH, which have pledged to provide cost-neutral technical support as well as subsequent testing under real industrial conditions. This paves the way for a follow-up project leading to market readiness (TRL 9), which opens up applications not only for the high-gloss polishing of metal surfaces but also for precision machining, microstructuring, materials engineering and other innovative industrial and medical applications.

Title:
Integration of bidirectional charging infrastructure into communal building energy supply

Project Leader:
Prof. Dr.-Ing. Ralf Hartig
, Chair of Renewable Energies

Project:
LVP

Abstract:
The integration of bidirectional charging infrastructure into communal building energy supply in accordance with Section 42b of the German Energy Act (EnWG) opens up new possibilities for a sustainable and resilient energy supply. In the planned project, on-site photovoltaic (PV) energy is to be used to charge electric vehicles. At the same time, these vehicles are to be able to feed electrical energy back into the building, either to ensure an off-grid supply during power cuts or to specifically shift peak loads within the building.

Several key components are required for successful implementation:
1. Metering concept: Development of a precise metering infrastructure to monitor and control energy flows between the PV system, vehicles and the building. 
2. Billing concept: Definition of a legally compliant and cost-effective model for the remuneration of energy transfers between vehicle owners, the building and the PV system operator. 
3. Energy management concept: Implementation of an intelligent energy management system to optimise energy distribution, taking into account load profiles, grid requirements and user needs.

A demonstrator is being developed to validate these concepts in practice. The Telewerk facility at Mittweida University of Applied Sciences is earmarked as the demonstration site. There, the developed concepts are to be tested and evaluated in a real-world environment. The project thus makes an important contribution to the further development and scalability of bidirectional charging systems in building services engineering and enhances energy autonomy and grid stability through smart sector coupling.

Title:
Laser-based additive manufacturing of high-resolution microcomponents made of pure copper using micro-SLM

Project Leader:
Prof. Dr.-Ing. André Streek,
Chair of Laser-Assisted Additive Manufacturing/Digitalisation of Laser Processes

Project:
Laser-based additive manufacturing of high-resolution microcomponents made of pure copper using micro-SLM

Abstract: Additive
manufacturing now enables the production of complex component geometries with a high degree of design freedom. In particular, powder-bed-based selective laser melting (SLM) has become established for metal fabrication. Materials such as stainless steel or aluminium can be used to produce high-density components with resolutions of up to 300 μm and very good mechanical properties to an industrial standard. Copper is becoming increasingly important for laser-based additive manufacturing due to its excellent thermal and electrical properties. However, processing pure copper is particularly challenging due to its low absorption in the near-infrared (NIR) wavelength range of the lasers typically used and its high thermal conductivity. These factors limit the minimum feature widths in the SLM process to around 0.7–0.8 mm, meaning that components requiring finer structures cannot be manufactured using this method. Yet there is growing interest and increasing market demand, particularly for structures smaller than 100 μm. A promising solution is offered by the micro-SLM process developed at the Laser Institute of Mittweida University of Applied Sciences (LHM), which represents a further development of the established laser powder bed fusion (SLM or L-PBF) method for 3D Printing of metal components into the micro-scale. It enables 3D printing of components with higher structural resolution, lower surface roughness and greater dimensional accuracy, using very fine powders with particle sizes of less than 10 μm and a laser spot of approximately 30 μm. This makes it possible to produce precision components with structural resolutions of up to 30 μm and layer thicknesses of up to 5 μm. Compared with the conventional SLM process, this represents a four- to five-fold improvement in resolution. The surface roughness of the untreated components is already as low as R_a = 2.5 μm. Subsequent glass bead blasting can reduce the roughness to R_a = 1.0 μm, and dry electropolishing can even reduce it to R_a = 0.1 μm. In initial trials with pure copper, a high density and feature widths of 150 μm have already been achieved. Based on the available results for the materials already in use, it can be assumed that, through targeted further development of the process – from material preparation and the adjustment of process parameters through to the optimisation of plant technology and post-treatment – it will be possible in future to produce high-purity copper structures with previously unattainable structural resolutions.

Title:
ValiXBloks

Project Leader:
Prof. Dr.-Ing. Christian Roschke,
Chair of Digital Transformation and Applied Media Informatics

Project:
LVP

Abstract:
The completed research project ‘xBloks’ at Mittweida University of Applied Sciences (BMBF) has produced a comprehensive security system for digital motorsport events, comprising several essential components: a motion simulator, a driver identification system, a blockchain infrastructure, a hardware dongle and a web-based user platform. The aim of this system is to take security and authentication at digital motorsport events to a new level by enabling attempts at manipulation to be effectively detected, prevented and reliably proven. The following summary explains the key findings and the planned validation steps in more detail:

1. Motion simulator technology (TRL Level 3) A two-seater motion simulator has been successfully designed, built and tested. This technology offers drivers and spectators a particularly immersive experience by simulating realistic driving physics and motion sequences. The design drawings, bills of materials and documentation produced enable the replication of further demonstrators whilst adhering to high quality standards. This creates a solid foundation for the further development and wider application of the technology in the professional sector.

2. Driver Identification System (TRL Level 3) By implementing a driver identification system, telemetry data from the racing simulations can be used to create digital fingerprints. This system detects deviations in driving behaviour and uses AI-based methods to detect attempts at manipulation. The integration with the motion simulator enables real-time analysis and continuous updating of driver profiles. This significantly improves the authenticity and traceability of the driving experience.

3. Blockchain infrastructure (TRL Level 3) The newly developed blockchain platform ensures the immutable storage and tamper-proof retrieval of all relevant data. This includes, in particular, the digital fingerprints and user information. With the help of smart contracts, the system can also respond automatically to defined events and, for example, trigger appropriate verification mechanisms in the event of suspected tampering. Overall, this technological foundation enhances the transparency and trustworthiness of digital motorsport events.

4. Hardware dongle (TRL Level 3) To ensure end-to-end protection at both the hardware and software levels, a special hardware dongle has been developed. This is integrated into the motion simulator and continuously reads system data, which is then sent to a central security gateway. In this way, it is determined whether both the hardware and software in use are free from tampering. This provides an additional layer of security across the entire system.

5. Web-based user platform (TRL Level 3) A web-based platform, comprising a front-end and a back-end, serves as the link between the individual components of the system. It provides users with transparent access to their profiles, telemetry data, and past and upcoming racing events. Event administration is also integrated here, which significantly simplifies the organisation of and participation in digital races. At the same time, all results can be linked directly to the corresponding driver profiles and securely stored on the blockchain.

Proof of concept has been demonstrated for the five components listed. The main objective of the validation project is to confirm the functionality and security of the developed overall system in real-world application scenarios (TRL Level 5). This includes testing identity validation under real-world conditions, as well as preventing and detecting potential attempts at manipulation. Furthermore, the plan is to draw up business and commercialisation plans, in which the economic feasibility will be examined and potential revenue streams and a business model identified. A comprehensive cost-benefit analysis is intended to demonstrate how the system can be operated profitably and sustainably.

Alongside technical and economic considerations, the focus is on environmental sustainability. The project will evaluate the extent to which the implementation of the system – particularly with regard to blockchain technology – results in additional resource consumption or can make a positive contribution to the sustainable organisation of digital motorsport events. This involves examining the carbon footprint, energy consumption and potential for optimisation in the areas of hardware and software. The findings will be used to guide targeted further development with regard to environmental, economic and social aspects. To adequately address all requirements, particularly in the areas of sustainability and Business Administration, a team member with in-depth knowledge of blockchain technologies as well as corporate governance and sustainability management will be involved in the project. This individual may already possess the relevant expertise or may be provided with comprehensive training during the course of the project. This expertise will enable the results of the xBloks project to be continuously analysed and optimised in line with the three aspects of sustainability.

Overall, the ‘xBloks’ project has thus laid the foundations for a novel security and authentication system in digital motorsport, the performance and market potential of which are to be demonstrated through the forthcoming validation under real-world conditions. The resulting analyses of cost-effectiveness and sustainability will play a key role in successfully establishing the system in the long term and developing it further with a view to the future.

Title:
From Proof of Concept to Application: Validation of a laser-based opto-acoustic measurement method for the non-destructive condition assessment of concrete structures

Project Leader:
Prof. Dr.-Ing. Jörn Hübelt
, Chair of Technical Mechanics and Acoustics

Project:
LVP

Abstract:
The proposed long-term validation project aims to transform a laser-based opto-acoustic measurement method, the functionality of which has already been demonstrated, into a practical, robust and validated testing method for the non-destructive condition assessment of concrete structures. The method is based on the targeted optical excitation of the component surface using high-energy laser pulses, which generates acoustic waves within the material. Material properties and internal damage influence characteristic signal parameters such as the frequency spectrum, amplitudes and damping behaviour, which can be detected and evaluated non-contact.
In previous research and development work, the fundamental proof of concept for the measurement principle was successfully demonstrated. In particular, for large-area structures with few geometric constraints, it has been demonstrated that material- and damage-related changes can be reliably detected. However, there remains a specific need for validation in the case of geometrically constrained concrete components with limited degrees of measurement freedom. Here, boundary conditions such as limited component dimensions, reflective interfaces and superimposed vibration modes lead to altered signal characteristics, the systematic recording and evaluation of which have not yet been sufficiently investigated.
The aim of the project is therefore to experimentally validate the existing measurement method under these more stringent boundary conditions and to adapt the measurement configuration and evaluation strategies specifically to geometrically constrained concrete components. To this end, reproducible reference signals are to be determined on intact components and systematically compared with signatures of defined damage states. The work deliberately focuses on the validation and readiness for practical application of the existing method and does not involve any fundamental redevelopment of the underlying measurement technology.
Upon completion of the project, the aim is to establish a robust foundation for the use of the opto-acoustic measurement method in further application- and industry-oriented projects.

Topic:
Validation of a process for the manufacture of directly boiling-cooled microelectronic components using laser-based surface structuring (LasoCool)

Project leader:
Prof. Dr.-Ing. Udo Löschner
, Chair of Laser Manufacturing Technology/High-Performance Mechanical Engineering

Project:
LVP

Abstract:
High-performance microelectronic components, such as those used in high-performance computing (HPC) for AI applications, generate significant amounts of waste heat as a result of power losses occurring during computing operations. This heat must be reliably dissipated to maintain computing performance, in order to prevent local overheating and resulting malfunctions, which could even lead to the destruction of the electronic components. This is where the proposed research project comes in. The aim of LasoCool is to develop a laser-assisted process for the functionally optimised microstructural design of cooling surfaces for electronic components, which will be researched as part of the project and validated under realistic operating conditions. Boiling cooling via bubble boiling is a highly efficient method for heat transfer at surfaces, as the formation and detachment of vapour bubbles allow very large amounts of heat to be dissipated even at comparatively small temperature differences. With the aid of laser-fabricated, defined microscale structures on the cooling surfaces, the detachment behaviour of vapour bubbles can be specifically influenced, thereby stabilising the particularly efficient bubble boiling and enabling it to be initiated even at lower temperatures. This enables a faster flow of coolant to the heat-loaded surfaces, increases heat dissipation and reduces the operating temperature of the electronic components. The improved cooling performance leads not only to enhanced performance but also to improved energy efficiency and operational reliability, as well as an extended service life for the systems. This results in direct benefits for a wide range of applications in the key, future-oriented sectors of the energy and digital transitions, such as HPC, data centres, power electronics and electromobility.
The application of microscopically small functional structures to the heat sink surface is to be carried out using the laser micro-machining technology available at the Laser Institute at Mittweida University of Applied Sciences (LHM). In particular, surface processing using ultrashort-pulse (USP) laser radiation, which is gentle on the workpiece, enables high-precision structuring without any significant thermal impact on the substrate. This even enables surface structuring on the rear side of carrier substrates without causing thermal damage to the electronic components on the functional side. This property also opens up the possibility of subsequent laser processing of electronic components that have already been fully processed.
Compared with the wet chemical etching processes predominantly used to date – which offer limited structural precision and flexibility due to isotropic material removal and are in some cases considered environmentally harmful and hazardous to health – laser surface structuring, as a novel and potentially disruptive approach, offers significant advantages. Laser Technology enables the low-damage or damage-free fabrication of microscopically small functional elements, as well as the precise and locally controlled structuring or roughening of substrate surfaces. Furthermore, laser-fabricated microstructures have high aspect ratios, which significantly increases the effective heat transfer area. This provides the basis for a significant increase in the heat flux densities currently achievable, which in turn opens up new application possibilities, for example for high-power-density electronic systems and microstructured cooling concepts. In the long term, the results achieved in the project may also make it possible to integrate cooling structures directly into the backside of silicon wafers. This could, in the future, eliminate the need for additional heat sinks, thereby facilitating the miniaturisation of electronic systems and leading to savings in both material and weight. At the same time, this avoids the use of thermal interface materials, which have hitherto been used as a thermal interface layer but act as an additional thermal barrier and are subject to ageing processes.

Title:
Method for ML-based component detection and data validation for incoming goods inspection and assembly (VeMDetect)

Project Leader:
Prof. Dr.-Ing. Leif Goldhahn
, Chair of Production Informatics

Project:
LVP

Abstract:
The project addresses a specific practical shortcoming: automated component recognition has so far failed to achieve sufficient reliability under real production conditions. In incoming goods and assembly processes, this leads to incorrect allocations, additional inspection work and disruptions to the workflow (Bock, Goldhahn, Schubert, 2026). The aim of this validation project is to develop and test an image-based method that automatically recognises components, links them to associated process and order data, and checks them in context. This ensures that components are correctly identified and that subsequent work steps are carried out using the correct components. This reduces errors, prevents incorrect assembly and customer complaints, and saves time in logistics and assembly.
The project builds on preliminary work from the visKIMa and PerspektiveArbeit Lausitz projects, as well as research carried out in the ‘Adaptive Assembly’ and ‘Training Factory 4.0’ laboratory environments. Initial applications with industrial partners such as Festo Didactic SE, EMIS Electrics GmbH and CIMPCS GmbH show that recognition rates of up to 90 per cent can be achieved under laboratory conditions. However, there are clear limitations for industrial use: Visually similar or mirror-image components are frequently confused, changing lighting conditions and perspectives lead to errors, and the transferability to new components is limited. The current stage of development therefore corresponds to TRL 4 (technology validated in the laboratory).
Building on this, the ‘VeMDetekt’ method for ML-based component detection and data validation is being further developed. To this end, machine learning methods – in particular YOLO-based approaches – are being combined with more robust image features and additional context-based verification mechanisms. The aim is to stabilise detection performance under real-world conditions and to validate results directly within the process. Continuous retraining also ensures adaptability to new components and changing operating conditions.
Development is taking place in close collaboration with the associated partners: ICM GmbH Innovation + Cooperation for Mechanical Engineering, CIMPCS GmbH and bsw Bildungswerk der Sächsischen Wirtschaft gGmbH. The relevant letters of intent (LOIs) are attached to the application. Implementation will take place in the goods-in and assembly areas in collaboration with the end-user, ICM GmbH. Through structured workshops, they will contribute their requirements, experiences and typical challenges directly to the development process. The solution will be tested iteratively under real-world conditions and continuously optimised. At the same time, organisational framework conditions such as existing processes, training requirements and implementation strategies are taken into account at an early stage in cooperation with bsw and CIMPCS GmbH to ensure sustainable use.
For practical application, the process is integrated into an assistance system. This system evaluates identified components in real time, supports quality assurance and documents results in a traceable manner. The aim is to further develop the existing laboratory prototype (TRL 4) into a validated solution in a realistic environment (TRL 5) as the basis for subsequent industrial deployment.

Title:
Validation of the automated generation of synthetic human activity data for training AI models

Project Leader:
Prof. Dr.-Ing. Matthias Baumgart,
Chair of Digital Integrated Process Management

Project:
KVP

Abstract:
The automatic detection of human activities in image and video data has become increasingly important in recent years. Technological advances in the fields of pose estimation and machine learning have made it possible to extract complex information about human behaviour from images and video clips. In security-related contexts, in healthcare, in human-machine interaction and also in media analysis, there is a growing demand for robust, high-performance AI-based systems that can learn to distinguish between different activities and interactions on the basis of large amounts of data.

However, there is a key problem: access to high-quality, annotated video data that can be used without legal concerns is severely limited. In many cases, data protection laws – in particular the General Data Protection Regulation (GDPR) – prevent the unrestricted use of real-world data. Furthermore, many freely available datasets are severely restricted in terms of content, outdated, or represent a population that is too small or skewed. The manual annotation of videos is not only time-consuming and error-prone, but also presents an economically insurmountable obstacle on a large scale.

Against this background, the SMWK project ‘Automated Generation of Synthetic Human Activity Data for Training AI Models’, led by Prof. Dr.-Ing. Roschke at Mittweida University of Applied Sciences, was launched. The aim of the project was to develop a demonstrator for the synthetic generation of realistic video scenes in which people perform micro-activities. By combining tools such as MakeHuman, Mixamo and modern rendering techniques, it was possible to generate large quantities of activity data via scripting; this data is equipped with complete metadata, is non-discriminatory, anonymised and can be freely parameterised.

The planned validation project will build on these results. Whilst the original project focused on the generative component, the validation project concentrates on the scientific and technical evaluation of this generated data: Is it theoretically suitable for training AI models? How representative is the synthetic data in terms of ethnic diversity, body silhouettes, movement complexity and real-world behaviour?

The aim is to develop and implement a systematic validation procedure that assesses the quality and validity of synthetic video data in the field of AI-based activity recognition. The findings are intended to aid in the evaluation of the generation tool and to serve as the basis for the publication of an open validation methodology that can also be made available to other research institutions. The validation project focuses on the efficient generation of datasets.

Title:
Validation and integration of methods for analysing multimodal biometric systems in security-critical scenarios AMBOSS – Analysis of multimodal biometric systems in security-critical scenarios

Project Leader:
Prof. Dr. rer. nat. Dirk Labudde
, Chair of Forensics/Bioinformatics

Project:
KVP

Abstract:
The AMBOSS project aims to create a multimodal biometric dataset that can be used to validate and optimise biometric systems. Companies in the field of physical security technology and research will have the opportunity to test their biometric recognition systems using real, diverse data. In addition, recommendations for action will be developed for security authorities in various high-risk situations and threat scenarios, with a view to improving the analysis of biometric features under realistic conditions. Furthermore, the data will be made available for the validation of software systems. Future users will thus be able to draw on validated data to develop new systems or adapt existing ones.

Recent events in Germany demonstrate that more investment is needed in the protection of citizens. This also requires the continuous development of techniques and procedures in the field of biometrics. The figure illustrates various threat scenarios in which personal identification may be required. To this end, biometric features, which are available as templates, are compared with a reference trace using templates. Depending on the specific threat scenario and the relevant circumstances (e.g. time available, quality of the characteristics), different procedures must be devised.

The collection, processing and analysis of biometric data take place exclusively on the basis of the voluntary consent of the data subjects in accordance with Article 6(1)(a) of the GDPR. Only data necessary for the purposes of analysis is collected (data minimisation) and processed in accordance with the rights of data subjects (Articles 15–21 of the GDPR). Processing takes place exclusively on encrypted systems with access controls; data is only disclosed to third parties within the framework of a GDPR-compliant data processing agreement.

Title:
Validation of a kHz laser control system for high-throughput fluorescence microscopy

Project Leader:
Prof. Dr. rer. nat. Richard Börner,
Chair of Biophotonics/Physical Engineering

Project:
KVP

Abstract: The
aim of the project is to validate a newly developed laser control system (funded under Innovation Cluster 2023, AP3), which enables modular intensity and frequency control of laser sources in the kHz range – and potentially also in the MHz range in the future – for high-throughput fluorescence microscopy. This control system allows modern excitation schemes, such as stroboscopic, alternating laser excitation (sALEX for short), to be implemented efficiently and cost-effectively in the millisecond range. The innovative core lies in the combination of a cost-effective microcontroller platform with a precisely clocked high-frequency control system based on programmable delay elements. The control system is designed as an upgrade for existing microscope systems and, for the first time, enables manufacturer-independent access to time-resolved single-molecule spectroscopy at a high technical standard. Its relevance stems from the growing need for sensitive and resource-efficient measurement methods in medical and pharmaceutical diagnostics. In this context, microscopes are generally used (both from commercial manufacturers such as Olympus, Leika, Zeiss, Nikon and others, and custom-built models) which are equipped with high-quality components, but whose capabilities cannot be fully utilised. Existing solutions from manufacturers are either highly integrated and remain very expensive in the OEM sector. As part of the project, the control system will be further optimised, validated in existing TIRF and wide-field microscopes, and tested in collaboration with partners from research and industry. The expected outcome is the provision of a validated technology platform for single-molecule fluorescence detection that is scalable, flexible and transferable – with high potential for technology transfer.

Title:
Solar-powered latent heat storage (SOLALA)

Project Leader:
Prof. Dr.-Ing. Lutz Rauchfuß,
Chair of Drive and Control Engineering

Project:
KVP

Abstract:
We aim to validate and optimise the results of our completed ZIM project ‘ELAN’ (ending 31 December 2024), in which we demonstrated the triggerability of sodium acetate trihydrate solution (NaAc) as a loss-free, long-term solar storage medium. We intend to build on the project concept from ‘ELAN’ because it contributes to the energy transition by ensuring that surplus solar radiation is not left unused, but is stored latently without loss and only retrieved when heat is required. The description of the system, from the energy source to the storage unit, begins with the solar collector (solar thermal). This absorbs the sun’s thermal radiation to heat water, which is then stored in the residential property’s buffer tank. The surplus solar energy, which is available even during the transitional period, is to be stored as latent heat so that it is retained without loss until the next time heat is required. The innovative core of the system lies in the utilisation of the phase transition, which is reached in sodium acetate trihydrate (NaAc) as early as 58°C. However, the major advantage of NaAc lies in the ability of the charged melt to be supercooled. Unlike water, once the temperature falls below the melting point of NaAc, the absorbed energy remains stored in the liquid phase down to –15°C. Consequently, no special thermal insulation is required; an uninsulated storage tank is sufficient. When the heat is required again, an actuator developed by us can be used to introduce a nucleation site, causing the supercooled melt to crystallise whilst a temperature of 53°C is reached in the NaAc. Modern houses are almost always fitted with underfloor heating, which requires a flow temperature of 45°C. Achieving this temperature is the aim of our project proposal. We aim to demonstrate the successful improvement of the heat exchanger design, which transfers the temperature of the NaAc to the surrounding water via the largest possible surface area. Model calculations and simulations carried out support this approach. Project partners such as energie.depot from Radeberg have been manufacturing efficient thermal hot-water storage tanks for many years, but are unable to tap into new market segments without innovative long-term storage solutions. Our project partner’s commitment to bringing this innovation into series production remains unwavering.

Title:
Stereomicroscopic positioning aid for inclined laser structuring

Project Leader:
Prof. Dr. rer. nat. Steffen Weißmantel,
Chair of Physics/Physical Technologies

Project:
KVP

Abstract:
Project name & objective The
aim of the project is to validate a stereomicroscopic method for the precise alignment of workpieces within the focal region of a pulsed laser system. This is necessary for the industrial implementation of a patent-pending process (EM-061) for the laser-based creation of recesses with steep wall angles. Innovative Core The novel laser structuring process enables, for the first time, the targeted creation of steep or vertical wall surfaces through controlled tilting of the workpiece relative to the laser beam. The challenge lies in the precise, reproducible positioning of curved or free-form surfaces at an angle to the focal plane. 

Relevance
Conventional digital or reflected-light microscopy fails to capture steep wall structures optically, as these techniques rely on perpendicular observation and limited depth of field. In contrast, a stereomicroscope operates with two beam paths that are slightly inclined towards each other, thereby creating a three-dimensional impression of the surface. This three-dimensional stereo imaging enables the visual assessment of even obliquely oriented surfaces and structures – in particular steep wall surfaces, such as those encountered in the planned procedure. Furthermore, high-quality stereomicroscopes have a large depth of field compared to reflected-light microscopy, which enables a focused view of different planes within the inclined structure. This allows both pre-positioning relative to the laser’s focal plane and in-line inspection of edge angles to be carried out with a high degree of reproducibility – a crucial step towards industrial process integration.

Summary of implementation & expected impact
As part of the project, a suitable stereomicroscope will be procured, set up and integrated into an existing laboratory environment. Selected test geometries will be used to verify whether reproducible focusing and review of the inclined wall surfaces is feasible. The results will be incorporated into the planning of a system-integrated positioning and process control system. Validation using the stereomicroscope not only represents an intermediate technical step but also lays the foundation for a subsequent long-term validation project involving industry partners.

Title:
Technology Assessment through AI-Supported Patent Searches

Project Leader:
Prof. Dr. rer. nat. Thoralf Gebel
, Chair of Innovation Management and Change
Management 

Project:
KVP

Abstract:
The planned project ‘Technology Assessment through AI-Supported Patent Searches’ aims to systematically investigate the potential of AI-supported patent searches and analyses for supporting innovation and research processes at universities. The starting point is the growing importance of patent analysis as a key source of knowledge for technological developments, competitive analysis and innovation strategies.
Traditional patent searches are often time-consuming, complex and require a high level of specialist knowledge. Modern AI approaches, on the other hand, enable a significantly more efficient and intuitive analysis of large patent datasets and related scientific publications. This allows comprehensive analyses and searches to be carried out in a short space of time, which not only reveal the potential of new technologies but also highlight their limitations and constraints more quickly. 
As part of the project, the Scopri platform will be used as an example, as it allows for rapid, topic-specific analysis of patent landscapes. The aim is to evaluate this platform in comparison with other AI methods and traditional database-based research approaches. The applicant aims to establish AI-supported patent analyses as a strategic tool to support internal university research and innovation projects in the future. This should enable technology trends to be recognised at an early stage, innovation potential to be identified, and research activities to be targeted more effectively. Furthermore, the expertise gained through this project will not only contribute to internal optimisation at HSMW but also enhance the university’s external profile – including through collaborations with industry – thereby providing an advantage for future projects in terms of both their planning and implementation. In addition, by incorporating the project results into teaching, better-qualified graduates can be provided for the Saxon labour market, thereby strengthening the innovative capacity of Saxon companies.

Title:
AutoWell – Validation of a 96-well plate implementation for high-throughput confocal fluorescence measurements

Project Leader:
Prof. Dr. rer. nat. Richard Börner,
Chair of Biophotonics/Physical Engineering

Project:
KVP

Abstract: The
aim of the project is the rapid commissioning and validation of a proprietary 96-well plate holder on the MT200 confocal fluorescence microscope manufactured by PicoQuant GmbH. The microplate holder is to be inserted into the existing scanner adapter in a mechanically reproducible manner and calibrated via the software within the MT200’s wide-range scanner in such a way that individual wells or defined well areas can be specifically targeted. Following manual referencing of the confocal focus to the cover glass, a defined Z-position within the solution is to be set in order to acquire fluorescence time series in the respective wells.

The innovative core lies in the adaptation of a 96-well plate concept—which has already been scientifically demonstrated—into a cost-effective, robust and, in the long term, platform-independent hardware extension for existing inverted confocal fluorescence microscopes. Of particular relevance is the combination of the well plate holder with an anti-evaporation device for water-immersion objectives, which has also been developed in-house. In contrast to more complex solutions involving an active supply of immersion water – for example, via syringe pumps – this approach is intended to enable simple, efficient and low-maintenance long-term operation.

The project is important because confocal fluorescence microscopes and single-molecule FRET measurements have, to date, often been operated with only low sample throughput. However, for the characterisation of fluorescently labelled biomolecules, particularly RNA-based assays, automated measurement series in standardised multi-well formats are of great importance. An established potassium-dependent FRET series of an RNA construct, for which ensemble and single-molecule data are already available in the Biophotonics Research Group, will be used as a validation application.

During the three-month project period, the microplate holder on the MT200 is to be mechanically integrated, calibrated within the scanner coordinate system and experimentally validated using fluorescence time series. In addition, tests will be carried out to determine whether the evaporation protection functions reliably for at least 10 hours without the immersion medium needing to be topped up. The expected outcome is a validated proof of concept on the MT200, as well as a robust assessment of the technical limitations of the existing wide-range scanner. In the short term, the microplate holder and evaporation protection are to be incorporated into the technology and sales portfolio of the Biophotonics Group. In the medium term, the project will form the basis for a platform-independent complete solution with its own control system and wide-range scanner.