“Overland Assessment of Physical Environments for Mobility in Complex Ecological Landscapes” | W81EWF 23 SOI 0022

Opportunity Information: Apply for W81EWF 23 SOI 0022

The Department of Defense (Army Corps of Engineers), through the U.S. Army Engineer Research and Development Center (ERDC), is offering a cooperative agreement opportunity focused on improving how well synthetic, AI-generated geospatial landcover products match real conditions on the ground. The core idea is straightforward: AI, machine learning, and deep learning are increasingly used to rapidly produce maps and other geospatial datasets that support government planning and operations, but those products can only be trusted to the extent they are validated against high-quality field observations. In many real-world or contested situations, traditional large-scale ground-truthing is difficult, dangerous, or simply not feasible, so ERDC is looking to build a rigorous, science-based approach for collecting targeted field data and using it to measure and improve the accuracy of synthetic landcover and related derivative products.

The work centers on detailed overland landscape data collection and accuracy assessment in forested areas and mixed terrain, specifically in the upper coastal plain and lower piedmont setting at Fort A.P. Hill, Virginia. The selected awardee will conduct field-based research that documents physical and biotic environmental characteristics relevant to mobility, particularly in support of assured position, navigation, and timing (APNT) activities and tests. In practice, this means collecting precise information about terrain and vegetation that can affect how people and vehicles move through a landscape, and then using those observations to evaluate how well different provided datasets (including AI/ML/DL-derived synthetic datasets and more traditional foundation landcover datasets) represent reality.

A major deliverable is a comprehensive, traditional scientific accuracy assessment that compares three things side by side: (1) synthetic datasets produced using AI/ML/DL approaches, (2) existing foundational landcover datasets, and (3) field-verified ground truth data collected during the project. The accuracy assessment is expected to be detailed enough to produce meaningful confusion matrices and other standard accuracy metrics, including analysis of where and why misclassifications occur. ERDC is explicitly interested in capturing differences between leaf-on and leaf-off conditions to understand seasonal impacts on classification performance, which is often a major source of error in forest and mixed-cover environments.

Field collection is expected to be highly detailed and spatially precise, with a minimum mapping unit on the order of about 0.5 meters accuracy or better. The opportunity calls for collecting information across multiple land cover types and about 15 representative areas spanning different landforms and locations within the Fort A.P. Hill area of interest. Examples of the types of field data requested include surface soil characteristics, vegetation type and structure, landforms, percent cover estimates, stem counts and stem density estimates, canopy size or canopy characteristics, barriers or obstacles that may affect mobility, edge habitats, and human-made (anthropogenic) features. Collectively, these measurements are meant to support both mobility-focused interpretation and robust validation of landcover products and their derivatives.

On the data management side, ERDC wants the resulting datasets to be usable immediately in mapping and modeling workflows. Data must be delivered in standard geospatial formats suitable for ingestion into common GIS and modeling programs. The awardee is also expected to provide access to ongoing spatial data during the collection phase, including current and detailed aerial imagery and web-accessible, survey-grade GPS field collection outputs so ERDC can monitor progress and data quality as the work is underway. ERDC (specifically USACE-ERDC-GRL) will assist in developing a proposed data dictionary to standardize how observations are recorded, but the awardee is expected to generate clear documentation of processes and workflows. That documentation may include written manuals, videos, and other media that explain how data were collected, processed, quality-checked, and prepared for accuracy assessment.

The broader intent goes beyond a one-off validation exercise. ERDC wants results that help the community understand the strengths, limitations, and appropriate uses of AI-generated landcover products, especially for operational planning contexts where mistakes can have real consequences. The outputs are also meant to inform widely used land cover products and updates, explicitly including the USGS National Land Cover Database (NLCD) and the Virginia Land Cover Database, while also contributing findings to the wider scientific literature on landcover mapping and validation.

Administratively, the opportunity is listed as a discretionary funding opportunity under a cooperative agreement mechanism (Funding Opportunity Number W81EWF 23 SOI 0022; CFDA 12.630). The expected number of awards is one, with an award ceiling of $200,000. The posting was created on June 13, 2023, with an original closing date of August 4, 2023. The eligible applicant category is listed as "Others," with additional eligibility details referenced in the full announcement.

• The Department of Defense, Dept. of the Army -- Corps of Engineers in the science and technology and other research and development sector is offering a public funding opportunity titled "“Overland Assessment of Physical Environments for Mobility in Complex Ecological Landscapes”" and is now available to receive applicants.
• Interested and eligible applicants and submit their applications by referencing the CFDA number(s): 12.630.
• This funding opportunity was created on Jun 13, 2023.
• Applicants must submit their applications by Aug 04, 2023. (Agency may still review applications by suitable applicants for the remaining/unused allocated funding in 2026.)
• Each selected applicant is eligible to receive up to $200,000.00 in funding.
• The number of recipients for this funding is limited to 1 candidate(s).
• Eligible applicants include: Others (see text field entitled Additional Information on Eligibility for clarification).

Apply for W81EWF 23 SOI 0022

[Watch] Creating a grant proposal using the step-by-step wizard inside the applicant portal: