Presentations

RobDT: AI-enhanced Digital Twins for Space Exploartion Robotic Assets

September 01, 2022

Whorkshop, 2nd International Conference on Applied Intelligence and Informatics (AII2022), Reggio Calabria, Italy

The ROBDT project aims at developing a Robotic Digital Twin framework that combines data-driven models, physics-based and symbolic models and uses online data and data analytics to adapt models at runtime. The digital twin will support the robotic asset operations by providing timing and reliable prediction and by supporting what-if analysis to assess multiple scenarios. More information here

Epistemic Planning applied to Robotics

March 28, 2022

Whorkshop, AI4HRC Worskhop, Udine, Italy

Davide Soldà, neo laureato magistrale in Informatica, illustra brevemente la sua tesi di laurea nella quale ha realizzato una architettura software chiamata EPICO che permette di modellare ad alto livello problemi di planning epistemico, di trovare le soluzioni (i piani) attraverso la combinazione di due solver e di eseguire questi piani con i Panda Robot. More information here

Deep Learning for forecasting Thunderstorms on NE Italy

September 05, 2019

Poster, European Sever Storms 2019, Cracow, Poland

Forecasting thunderstorm is challenging, in particular in a complex area, like NE Italy, where complex orography (Alps chain), plain, lagoon and sea deeply interact with each other. In this work a completely new approach is proposed, based on machine learning and in particular using Convolutional Neural Network, for forecasting cloud-to-ground lightning occurrence and number, every 6 hours. Many fields forecasted by the ECMWF IFS model, or computed from them (as for example the equivalent potential temperatures), are used as candidate predictors. The first results seem encouraging and will be compared with those obtained using a more “classical” approach, that is, implementing an analogue method with the AtmoSwing software. More information here