ROSIN will create a step change in the availability of high-quality intelligent robot software components for the European industry. This is achieved by building on the existing open-source “Robot Operating System” (ROS) framework and leveraging its worldwide community. ROS and its subsidiary ROS-Industrial (European side led by TU Delft and Fraunhofer) is well-known, but its European industrial potential is underestimated. The two main critiques are (1) is the quality on par with industry, and (2) is there enough European industrial interest to justify investing in it? Partially, the answer is “yes and yes”; ample industrial installations are already operational. Partially however, the two questions hold each other in deadlock, because further quality improvement requires industrial investment and vice versa. ROSIN will resolve the deadlock and put Europe in a leading position. For software quality, ROSIN introduces a breakthrough innovation in automated code quality testing led by IT University Copenhagen, complemented with a full palette of quality assurance measures including novel model-in-the-loop continuous integration testing with ABB robots. Simultaneously, more ROS-Industrial tools and components will be created by making 50% of the ROSIN budget available to collaborating European industrial users and developers for so-called Focused Technical Projects. ROSIN maximizes budget efficacy by alleviating yet another deadlock; experience shows that industry will fund ROS-Industrial developments, but only after successful delivery. ROSIN provides pre-financing for developers which will be recovered into a future revolving fund to perpetuate the mechanism. Together with broad education activities (open for any EU party) led by Fachhochshule Aachen and community-building activities led by Fraunhofer, ROSIN will let ROS-Industrial reach critical mass with further self-propelled growth resulting in a widely adopted, high-quality, open-source industrial standard.

Organizations are making massive investments in artificial intelligence technologies such as Machine Learning (ML). Yet, ML project fail often due to high uncertainty. In this research, we conduct a survey to examine how organizations can cope with the high uncertainty of ML projects. The project is a collaboration between the IT University of Copenhagen and the University of Auckland.What? Why? How?

The question investigated in this project is: How do we best develop adaptive digital tools for technical knowledge acquisition? Adaptive, in this context, promises that such tools are appropriate for diverse cognitive learning styles, and technical knowledge entails a combination of 1) Procedural knowledge (knowing how to complete tasks and solve programming problems) and 2) Complex problem solving skills (knowing how to understand and approach an open-ended programming problem). The project is founded on the premise that machines and tools should make humans smarter and that technology is not the goal, but rather the means to an end. The objectives of the project are to produce new general methods for defining and diagnosing learning styles, new evaluation measures for whether tools support technical knowledge acquisition, as well as prototype novel tools that measurably enhance human cognitive capacity and strengthen technical knowledge.

Sharing data is the foundation of open science, helping efficiency and reproducibility of research, and a fertile ground for a vibrant new data-driven economy. Yet, whatever the value created by sharing data, we cannot ignore the risks that it presents to privacy protection. The best known data protection measure, differential privacy, is so complex that neither data scientists, lawmakers, nor data subjects can appreciate the guarantees it gives! In contrast, probability is a commonly understood measure of risk. Can we use the probability of re-identification of a subject by an attacker to measure anonymity? We will answer this question by building Bayesian models for a range of standard statistical procedures and anonymization algorithms. We will then test the effectiveness of existing Bayesian inference algorithms for assessing the reidentification risk. If successful, we will be able to initiate building new foundations for data sharing in science

Engagement with the art and music can have a range of positive effects on physical and mental health. However, access to the many benefits of musical experience can be limited due to various reasons: (1) the geographic location of the venue, (2) personal limitations in access, (3) limitation on the schedule of the concerts and performances, (4) cost of the experience. Due to these limitations the effort and cost of getting access to the concerts and performances override additional value the participant would get from the experiences. In XTREME project, we remove these limitations by developing a human-centered mixed reality (MR) solution that, instead of requiring the participant to be physically present in the venue, brings the concerts and performances to a remote location via MR technology while retaining the immersive experience, and even giving additional dimensions to it by integrating real and virtual contents. Our solution also allows for virtually shared, realistic experience together with selected social circles of the user. The solution removes geographic limitations as by the MR solution the concerts and performances can be accessed virtually from anywhere and they hence are more inclusive for people with limited access. Likewise, providing a global access to the art and music provides a wider audience for musicians and artists, not limiting the number of seats by the size of the physical venue. As this solution removes the necessity to travel to concerts and performances, it also provides a greener alternative. In contrast to the current state-of-the-art, our solution targets at making an MR solution to the concerts and performances where those are in in 3D, both for visuals and sound, blended into the room of the users. We will also provide maximised experience of presence via reconstruction of the users and their behaviour in the shared access, and enhance the immersion by providing an additional visual layer to the sound.