Zeus AI awarded DOE SBIR Phase I to develop multimodal foundation models for urban meteorology

Zeus AI is pleased to announce a new funding award from the U.S. Department of Energy’s Small Business Innovation Research (SBIR) Program. Our Phase I project, A Meteorological Foundation Model for Gap-filled High-Resolution Data in Urban Environments, will leverage generative models trained on multimodal earth observations to seek breakthroughs in neighborhood-level heat monitoring.

The urban heat island effect presents a growing threat to human health and wellbeing, particularly in vulnerable communities. Heat islands also increase both overall and peak energy consumption, but are difficult to monitor with current tools due to heat’s significant variability in space and time.

Our project aims to fill this gap by employing multi-resolution satellite data and next-generation AI. Foundation models, trained on multimodal Earth observations, and image super-resolution via diffusion models will allow us to predict urban heat patterns at 70-meter resolution.

Through our partnership with Wet Dog Weather, we’ll develop a real-time web visualization tool to make this data readily accessible to stakeholders. A case study collaboration with the City of Boston will provide a real-world testing ground for our technology, paving the way for broader deployment in cities across the nation.

The DOE SBIR program provides non-dilutive funding for early-stage research and development with significant commercialization potential. By harnessing the power of multimodal AI and satellite data, we aim to equip city planners, public health officials, and residents with actionable insights to mitigate heat stress and create more livable, resilient communities.