Federal Project Grant Award Summary

Award Overview and Funding Details

The National Science Foundation's Division of Computing and Communication Foundations awarded a $333,333 Project Grant to the Trustees of the Colorado School of Mines (doing business as Colorado School of Mines) under the Computer and Information Science and Engineering program (CFDA 47.070). The award, effective September 1, 2026, with completion anticipated August 31, 2029, supports collaborative research focused on developing advanced machine learning theory and algorithms for multi-sensor autonomous sampling missions.

Research Deliverables and Technical Focus

The project delivers theoretical advances and practical algorithms in Gaussian Processes (GPs)—a widely-adopted machine learning methodology—that accommodate heterogeneity across sensing systems, autonomous agents, and data sources. The research addresses three primary areas: (1) heterogeneity of sensing agents with differing movement capabilities, costs, and motion constraints; (2) heterogeneity of sensor operations and characteristics; and (3) heterogeneity of sensed phenomena. By extending GP-based active learning algorithms to handle these real-world operational variations, the project enables autonomous systems to gather environmental and observational data more efficiently and cost-effectively than current homogeneous-assumption methods, producing high-quality maps and datasets with fewer total samples required. The work directly supports the NSF's broader mission of advancing computing and information science research with applications across multiple scientific and engineering disciplines.