Fraunhofer Italia Research

Background/Objectives: The global syndemic of obesity, undernutrition, and climate change highlight the complex health and environmental challenges faced by young adults. These challenges may intensify during the transition to university. As a matter of fact, limited budgets, time constraints, and insufficient culinary skills often lead to unbalanced diets and increased risk of obesity. University cafeterias, serving large numbers of students, represent an ideal setting to promote healthier and more sustainable eating behaviors. The FOOD-HACK Project aimed to design and implement a cafeteria-based intervention using nudging strategies to promote healthier and more sustainable lunch choices among university students. Methods: This pilot study employed a pre-post design with two independent phases in the Polo Cravino cafeteria at the University of Pavia. Food consumption was assessed over 12 non-consecutive days across four weeks. During the intervention, three nudging strategies were implemented: (1) choice architecture, (2) salient labeling, highlighting healthy and sustainable options, and (3) educational prompts. Results: Across both phases, 2400 tray photographs were collected. Post-intervention, the proportion of trays aligned with the Harvard Healthy Eating Plate and EAT-Lancet Planetary Diet models increased, reflecting higher consumption of vegetables and fruit. Legume-based first courses increased; however, legumes did not substantially replace animal proteins as the main protein source, and meat remained predominant in second courses. Reductions in trays containing multiple carbohydrate sources were also observed. Conclusions: The nudging intervention improved overall meal quality, demonstrating that subtle environmental modifications can guide students toward healthier dietary choices, particularly by increasing fruit and vegetable intake. However, the persistent preference for animal proteins highlights the challenge of shifting protein consumption toward more sustainable sources. These findings suggest that nudging can be an effective tool to promote healthier and more balanced eating behaviors in university settings, though complementary strategies may be needed to foster substantial changes in protein choices.

The rapid expansion of artificial intelligence (AI) and cloud computing is creating a significant but often overlooked impact on global water resources. This paper presents a global assessment of water consumption in AI-driven data centres, distinguishing between water consumption for operational use at the facility, off-site water consumption related to electricity generation, and embodied water consumption associated with hardware manufacturing and supply chains. To anticipate future demand, a scenario-based probabilistic forecasting framework inspired by Bayesian methods is developed, combining sparse empirical data with expert-informed assumptions and policy-relevant growth trajectories for the years 2030 and 2050. Results suggest that, without mitigation, global water consumption associated with data centres could increase more than seven times by mid-century, with cooling-related operational consumption accounting for the majority of demand. Several mitigation pathways are identified, including improvements in cooling efficiency, adoption of alternative technologies, and infrastructure planning that takes into account regional water availability. A sensitivity analysis highlights the strong influence of compute growth and efficiency trends on future outcomes. The findings offer a transparent and adaptable basis for aligning AI infrastructure development with long-term water sustainability goals. • Provides probabilistic forecasts of global water demand from AI data centres to 2050. • Introduces a scenario-based framework grounded in Bayesian-informed assumptions. • Highlights cooling efficiency, electricity sourcing, and siting as critical mitigation levers.

This work addresses the problem of reducing the Sim2Real gap of vacuum grasping for physics-based simulated environments. We introduce a grasp contact computational model, integrating it into the recently proposed Nvidia Isaac Sim robotics platform. The proposed approach enables realistic assessment of the seal formation between the gripper’s suction cup and the object of interest, enhancing the simulation of gripper-object attaching and detaching behavior. We define five conditions that must be satisfied to accurately replicate realistic seal formation between the gripper and the object, using a quasi-static spring model for the suction cup and custom rules to detect holes and bulges on the object’s surface. This enhancement is essential for complex object geometries, where default grasp contact models fail to account for seal loss due to surface discontinuities. The grasp contact model is compared to the default Surface Gripper in Isaac Sim. Simulations and real-world experiments show that our approach outperforms the default model in accurately reflecting the real seal formation in the presence of holes, bulges and discontinuities of the object’s surface. A link to the code repository is also provided.

In situ robotic automation in construction is challenging due to constantly changing environments, a shortage of robotic experts, and a lack of standardized frameworks bridging robotics and construction practices. This work proposes a holistic framework for construction task specification, optimization of robot morphology, and mission execution using a mobile modular reconfigurable robot. Users can specify and monitor the desired robot behavior through a graphical interface. In contrast to existing, monolithic solutions, we automatically identify a new task-tailored robot for every task by integrating Building Information Modeling (BIM). Our framework leverages modular robot components that enable the fast adaption of robot hardware to the specific demands of the construction task. Other than previous works on modular robot optimization, we consider multiple competing objectives, which allow us to explicitly model the challenges of real-world transfer, such as calibration errors. We demonstrate our framework in simulation by optimizing robots for drilling and spray painting. Finally, experimental validation demonstrates that our approach robustly enables the autonomous execution of robotic drilling.

Classical robot-programming approaches often require the definition of rigid sequences of elementary motions and actions by expert robot programmers through coding. Such solutions are suitable for environments that are more structured and less dynamic with respect to construction ones. We propose a different approach with a novel software architecture that exploits Building Information Modelling (BIM) data for facilitating robot-programming and deployment in construction. It includes mission parametrization and customization tools that abstract elementary robot functionality in terms of behaviors, that can be easily understood, assembled, and adapted on-site by non-expert programmers. We verify the effectiveness of our proposed approach by considering the execution of a spray-painting use-case for interior walls and a usability study.

ABSTRACT In this paper, an approach for computing online safety zones for collaborative robotics in a robust way, despite uncertain robot dynamics, is proposed. The strategy implements the speed and separation monitoring paradigm, and considers human and robot enclosed in bounding volumes. The human-robot collaboration is monitored by a supervisory controller that guides the robot to stop along a path-consistent trajectory in case of collision danger between human and robot. The size of the robot safety zone is minimized online according to the stop time of the manipulator, and the uncertain robot dynamics is considered using interval arithmetic to ensure compliance with the joint torques limits even in case of imperfect knowledge of the dynamic model parameters. The results verify the effectiveness of the proposed approach, and evaluate the influence of dynamics variations on human-robot collaboration.