Within the profile Technical Information Technology (ICT Department) the most important specializations are Embedded Software and Industrial Automation. About half of the Technical Information curriculum consists of learning modules, the other half is organized in projects. The whole study lasts four years. After two-and-a-half year students choose a specialization. Before the choice is made students have several occasions in which they learn something about the possible fields of specialization. In the first and second year there are two modules about Industrial Automation. First there is a module on actuators, sensors and interfacing, later a module on production systems. Finally there is an Industrial Automation project. In this project groups of students get the assignment to develop the control for a scale model flexible automation cell or to develop a monitoring system for this cell. In the last year of their studies students participate in a larger Industrial Automation project, often with an assignment from Industry. Here also the possibility exists to join multidisciplinary projects (IPD; integrated product development).
DOCUMENT
Automation surprises in aviation continue to be a significant safety concern and the community’s search for effective strategies to mitigate them are ongoing. The literature has offered two fundamentally divergent directions, based on different ideas about the nature of cognition and collaboration with automation. In this paper, we report the results of a field study that empirically compared and contrasted two models of automation surprises: a normative individual-cognition model and a sensemaking model based on distributed cognition. Our data prove a good fit for the sense-making model. This finding is relevant for aviation safety, since our understanding of the cognitive processes that govern human interaction with automation drive what we need to do to reduce the frequency of automation-induced events.
DOCUMENT
A description of our experiences with a model for education in innovative, interdisciplinary and international engineering. (Students from different (technical) disciplines in Higher Education are placed in industry for a period of eighteen months after completing two-and-a-half year of theoretical studies). They work in multi-disciplinary projects on different themes, in order to grow to fully equal employees in industry. Besides students, teachers and company employees participate in the projects. The involvement of other level students, both from University and from Vocational Education, is recommended. The experiments in practice give confidence in the succesful implementation of this model.
DOCUMENT
The increasing amount of electronic waste (e-waste) urgently requires the use of innovative solutions within the circular economy models in this industry. Sorting of e-waste in a proper manner are essential for the recovery of valuable materials and minimizing environmental problems. The conventional e-waste sorting models are time-consuming processes, which involve laborious manual classification of complex and diverse electronic components. Moreover, the sector is lacking in skilled labor, thus making automation in sorting procedures is an urgent necessity. The project “AdapSort: Adaptive AI for Sorting E-Waste” aims to develop an adaptable AI-based system for optimal and efficient e-waste sorting. The project combines deep learning object detection algorithms with open-world vision-language models to enable adaptive AI models that incorporate operator feedback as part of a continuous learning process. The project initiates with problem analysis, including use case definition, requirement specification, and collection of labeled image data. AI models will be trained and deployed on edge devices for real-time sorting and scalability. Then, the feasibility of developing adaptive AI models that capture the state-of-the-art open-world vision-language models will be investigated. The human-in-the-loop learning is an important feature of this phase, wherein the user is enabled to provide ongoing feedback about how to refine the model further. An interface will be constructed to enable human intervention to facilitate real-time improvement of classification accuracy and sorting of different items. Finally, the project will deliver a proof of concept for the AI-based sorter, validated through selected use cases in collaboration with industrial partners. By integrating AI with human feedback, this project aims to facilitate e-waste management and serve as a foundation for larger projects.
In INCEPTION (INdustrial roboChEmic PlaTform ImplementatiON) Zuyd Hogeschool, the Noël Research Group (University of Amsterdam) and partners will develop and implement automated robotic platforms relying on advanced AI algorithms to accelerate reaction optimization, based on the RoboChem platform. Thus, synthesis of active pharmaceutical ingredients within the drug development process will be optimized. This will diminish the time-to-market for new medicines and improve the sustainability of this development process. To develop and implement these RoboChemic platforms, a consortium of chemical and high-tech partners will cover all aspects related to required hard and software, e.g. automation (Beartree Automation), reactors (Chemtrix) and analysis (Mettler-Toledo). The development and implementation will be guided by pharmaceutical Contract Research Organization end-users Ardena and Symeres. The mix of partners from academia (Noël Research Group), Center of Expertise CHILL, Zuyd and multiple companies ensures an efficient and integrated development. The overarching question: “How can AI-assisted optimization and RoboChemic platforms efficiently be implemented in the chemical industry?” and research question: “What improvements on set-ups, programs, and capabilities are necessary for optimal industrial use?” will be answered by: i) Extending the applicability of RoboChemic platforms in industry by exploiting the modularity of its hardware control platform by incorporating additional equipment, and exploiting its software flexibility by adding optimization objectives and human-in-the-loop functionalities. (WP2) ii) Exploring identification of pharmaceutical relevant side-products, and subsequent rapid optimization. (WP3) iii) Enabling efficient upscaling using data at small scale by coupled learning at higher scales. (WP4) iv) Allowing enzymatic catalyst screening by an automated platform. (WP4) v) Accelerating uptake of RoboChemic platforms in industry by dissemination of demonstrator applications and by evaluating a prospective start-up implementing and servicing RoboChemic platforms. (WP5) Thus, by implementing RoboChemic platforms in industry we will make the pharmaceutical CROs and equipment companies more competitive.
Automating logistics/agrifood vehicles requires dependable, accurate positioning. Automated vehicles, or mobile robots, constantly need to know their exact position to follow the trajectories required to perform their tasks. Precise outdoor localization is helped by the increased price/performance ratio of RTK-GNSS solutions. However, this technology is sensitive to signal deterioration by e.g. biomass and large structures like poles/buildings. Robust localization requires additional localization technologies. Several absolute and relative positioning technologies exist and available sensor fusion solutions allow for combining these technologies. However, robot developers require modularity, and no integral solutions exist. Commercial solutions are either customized or high-priced testing solutions. Academics mainly propose specific sensing combinations and lack industrial applicability. Market demand articulation expresses the need for redundancy besides modularity, both for vehicle safety and system resilience, referring to the current geopolitical GPS jamming reality. MAPS aims for an open-source, ROS2-based, multi-modal, robust and modular localization solution for outdoor logistics and agrifood applications, enabling dependable and safe vehicle automation, allowing both sectors to handle labor shortages, introduce durable solutions and enhance resilience. MAPS focuses on a sensor fusion approach allowing modularity, with integrated redundancy. It includes online confidence level estimation, supporting both continuous fusion and modality switching, aiming for location/situation aware behavior and allowing for market-requested hybrid in-vehicle/infra solutions. MAPS intents to maximally utilize the consortium’s vehicle dynamics knowledge - including vehicle-(soft)soil interaction - in the solution for plausibility and dead reckoning. An accompanying PhD/EngD research is foreseen. With project partners enabling scalable, industry-grade solutions MAPS aims to bridge the gap between academic-level research and market-desired applicability. MAPS is independent, though aims to cooperate with AIFusIOn from Saxion on re-usable architectures and integration of AIFusIOn specifics, like AI-based situational awareness and indoor-outdoor switching, if both are granted.
