From Space to Safety : CS GROUP Innovates at BiDS 2025

Riga, where science meets space and data

The air buzzes with excitement at the Big Data from Space 2025 conference in Riga. Researchers, engineers, and innovators from across Europe gather around bright stands and glowing screens.
Among them, one booth catches the crowd’s attention CS GROUP, proudly showcasing its latest breakthroughs from the ExtremeXP project, funded under the European Union’s Horizon Europe programme.

Flash Floods

Predicting Flash Floods Through Artificial Intelligence

On display is an intriguing poster: “Surrogate modelling and user-in-the-loop experimentation for urban flood prediction: the EXTREMEXP approach.” Behind this technical title lies a powerful idea using AI to forecast urban flash floods before they happen. The team focused on the city of Nîmes, in southern France, an area frequently struck by sudden and devastating Cévenol floods. By combining hydrodynamic simulations, meteorological data, and high-resolution topography, CS GROUP developed a deep learning model capable of predicting water depth up to 30 minutes in advance. This model, built on a U-Net neural network enhanced with temporal attention layers, learns how rainfall evolves over time and transforms these insights into accurate flood predictions. A step forward not only for environmental science, but also for the protection of cities and their inhabitants.

Keeping Humans in the Loop

What truly sets ExtremeXP apart is its human-centered philosophy. Here, machines don’t operate in isolation they collaborate with experts. Scientists can interact with the model, run simulations, visualize results, and even provide feedback to refine the system. This iterative process, known as human-in-the-loop experimentation, makes AI more transparent, explainable, and trustworthy essential qualities when dealing with crisis management and public safety.

A Shared Vision: Innovation Serving the Planet

At BiDS 2025, CS GROUP showcased a new era of space and data innovation, where AI, human expertise, and sustainable technology work together to better understand and protect our planet reminding us that space and Earth are deeply connected through knowledge and collaboration.

Discover the full poster here
The ExtremeXP project is co-funded by the European Union Horizon Program HORIZON CL4-2022-DATA-01-01, under Grant Agreement No. 101093164

© ExtremeXP 2023. All Rights Reserved – Privacy Policy