ACME ATRONOMATIC, LLC — Department of Defense STTR Phase I: N23A-T025
ACME ATRONOMATIC, LLC — STTR Phase I award from Department of Defense.
Phase I STTR feasibility signal
- Phase I awards fund proof-of-concept work. For capture teams, they mark early interest from Department of Defense in a technical approach.
- Watch for Phase II follow-ons from the same firm/topic family — that conversion path is where budgets and transition pressure rise.
- Obligated amount $139,946. Cross-check similar awards in the same agency and technology tags for going-rate context.
- Topic code N23A-T025 links this award to a solicitation family — search the same topic stem for incumbents and recompete timing.
- Amount
- $139,946
- Agency
- Department of Defense · Navy
- Program / Phase
- STTR · Phase I
- Topic
- N23A-T025
- Solicitation
- 23.A
- NAICS
- —
- Place of performance
- FL
- Period
- 2023-07-17 → 2024-01-16
Description
In a partnership with the Space Sciences and Engineering Center at the University of Wisconsin-Madison, ACME AtronOmatic d/b/a MyRadar, proposes to develop an end-to-end, nowcasting-to-NWP (N2N) suite that incorporates an artificial intelligence nowcasting model, output from high-quality numerical weather prediction (NWP) models, and an established blending algorithm to create a seamless experience for users. This proposal will highlight the strengths of both modern Artificial Intelligence (AI) techniques and traditional NWP methods, creating a superior product compared to using either alone. Our method will employ data fusion, as we combine passive remote sensing observations with characterizations of the thermodynamic instability of the environment from model initial conditions. We first employ machine learning to create a cloud classification algorithm which will be used to correct HRRR cloud layer initial conditions. A second generative AI system will create the nowcasting fields. Training and validation datasets will be sourced from co-located space-borne LIDAR measurements with geostationary satellite retrievals. Finally, we convert these deterministic models to probabilistic using stochastic sampling strategies as well as developing direct probabilistic methods.