Gustavo Tontini

Addressing the energy-environment nexus, our oceanographic lens on flow characterisation will pave the way for more efficient and sustainable development of tidal energy, contributing to a cleaner #energy future. #OceanEnergy Link to paper: nature.com/articles/s4146…

A study in @NatureComms uses images collected during @NASA’s DART mission of the asteroid Didymos and its moon, Dimorphos, to explore the origin and evolution of the binary system. go.nature.com/3Wj6QTi

@NatureComms Presentation by Dr Esther Lagemann about the work. youtu.be/ddomrz1cz_U?si…

Towards extending the aircraft flight envelope by mitigating transonic airfoil buffet #engineering @NatureComms nature.com/articles/s4146…

#EditorRant: Can the #nanogenerators communities please stop with the "lighting XXXX LEDs" demonstrations? These are NOT technically sound demonstrations, nor valid benchmarks, and just look silly. *Same with running a car over a #material or #device to show "how strong" it is.

Easy-to-actuate multi-compatible truss structures with prescribed reconfiguration #mechanical #metamaterials #engineering @NatureComms nature.com/articles/s4146…

⚡️ Real-Time Outage Management in Active Distribution Networks Using Reinforcement Learning over Graphs #energy #infrastructure #gridresiliency @NatureComms nature.com/articles/s4146…

Early warning and proactive control strategies for power blackouts caused by gas network malfunctions #energy #infrastructure #resiliency #engineering @NatureComms nature.com/articles/s4146…

Transparent radiative cooling cover window for flexible and foldable electronic displays @NatureComms nature.com/articles/s4146…

Limit State Equation and Failure Pressure Prediction Model of Pipeline with Complex Loading #MechanicalEngineering #StructuralEngineering @NatureComms nature.com/articles/s4146…

Lightweight and drift-free magnetically actuated millirobots via asymmetric laser-induced graphene #materials #engineering @NatureComms nature.com/articles/s4146…

An effective and rapidly degradable disinfectant from disinfection byproducts #environmental #chemistry #microbiology #engineering @NatureComms nature.com/articles/s4146…

Development of a micro-combined heat and power powered by an opposed-piston engine in building applications #engineering @NatureComms nature.com/articles/s4146…

Towards silent and efficient flight by combining bioinspired owl feather serrations with cicada wing geometry #bioinspired #materialsengineering #engineering @NatureComms nature.com/articles/s4146…

Radiative cooling and indoor light management enabled by a transparent and self-cleaning polymer-based metamaterial #thermalmanagement #engineering @NatureComms nature.com/articles/s4146…

Great coverage by @IFLScience of recent paper in our "Autonomous Driving and Traffic Systems" collection @NatureComms iflscience.com/autonomous-veh…

Autonomous vehicles were involved in fewer accidents than human-driven vehicles in most scenarios, says study in @NatureComms. The research supports that autonomous technology could improve road safety, while also identifying specific areas for improvement nature.com/articles/s4146…

Our work on 3D wind field estimation aboard small drones 🛩 has been published in @NatureComms! 📰: nature.com/articles/s4146… Most drones fly by actively fighting the forces of nature. Our project is a step towards enabling UAVs to embrace these phenomena to their advantage with the help of #AI. We trained a model called WindSeer 👀 that can run aboard a UAV in flight and, based on the UAV's wind measurements at its current location, predict the wind tens or hundreds of meters away, at a much finer spatio-temporal resolution than weather models. In the future, #AI will enable fixed-wing drones to plan their routes based on these predictions in order to fly like frigatebirds, which cover great distances without a wing flap by relying almost fully on wind energy. Our study started before the gen AI revolution and focused on an entirely different problem. Yet, at a high level our method ended up being surprisingly similar to the key insight behind training SLM models like Phi 💡: take a high-quality but expensive generative data source (computational fluid dynamics simulation, in our case), synthesize data from it in a controlled manner, and use it to train a small but powerful model. I am in awe of my collaborators from @ASL_ETHZ who made this result possible. Florian Achermann (not on X/Twitter) flew all aspects of this project through research challenges, COVID, Swiss aviation regulations, and GPU driver installations ad astra. I am also deeply thankful to @erichorvitz for his support of this type of -- literally, blue-sky! -- research at @MSFTResearch. Paper summary🎥: youtube.com/watch?v=SRKmbe… Drones dance over the Swiss Alps 🎥👇

🛰️☄️ The fast-spinning near-Earth Didymos #asteroid was found to have a degraded top shape by the #DART mission. This work reveals the surface features observed on it are more likely to have been caused by collisional effects than by YORP #planetaryscience nature.com/articles/s4146…