10 Jul 2026

Microsoft Azure Artificial Intelligence / Machine Learning Hackathon for Development of Retrieval-Augmented Generation Large Language Model

Abstract: The US Army Corps of Engineers (USACE) Civil Works (CW) research and development (R&D) mission is to address challenging environmental sustainability problems through innovative science and engineering, which helps to ensure a safer, more prosperous, and more resilient nation. To achieve this, the US Army Engineer Research and Development Center (ERDC) plans, executes, leads, and directs many R&D programs in coordination with USACE Headquarters, Districts, and Divisions through its multiple strategic focus areas, which include infrastructure, water modeling, crisis preparedness, ecosystem, sediment management, data, artificial intelligence, and robotics. In this process, much information is generated, including internal progress reviews, financial reports, scopes of work, work package planning, and success stories.

7 Jul 2026

Engineer Research and Development Center Process Automation System (E-PAS) Database Checker

Abstract: The purpose of this document is to specify the software requirements, architecture, and design for the US Army Engineer Research and Development Center (ERDC) Process Automation System (E-PAS) Database Checker, a tool that monitors the E-PAS database and provides notifications based on its size. This document is designed for the software engineers and developers maintaining Database Checker and is intended to aid them in understanding its architecture and underlying functionality.

7 Jul 2026

Technical Regional Execution Center No Effect Table (TREC NET)

Abstract: The purpose of this document is to specify the software requirements, architecture, and design for the Technical Regional Execution Center (TREC) No Effect Table (NET) macro suite, a collection of automated software routines and functions developed to manage and operate the NET. This document is designed for the software engineers and developers maintaining the macro suite and is intended to aid them in understanding its architecture and underlying functionality.

8 Jun 2026

MoistViT: A Vision Transformer Model for Moisture Content Prediction of Wood Chips

Abstract: Moisture content in wood chips is a critical parameter for industries such as pelleting mills, bio-refineries, paper mills, and renewable energy production. The moisture level significantly influences both the quality of the final product and the efficiency of the production process. Consequently, accurate knowledge of moisture content is of substantial importance to wood chip-reliant industries. However, current methods for determining moisture content are either time-consuming or require costly equipment and specialized setups. Therefore, developing a quick and reliable method for assessing wood chip moisture content is imperative. To address this need, we evaluate fourteen Vision Transformer (ViT) architectures and introduce an optimized model, MoistViT, developed using Bayesian Optimization Hyperband (BOHB) for efficient hyperparameter tuning. Experiments on two wood chip image datasets (1600 total images) show that MoistViT achieves 91% accuracy and 92% F1-score on Source 1 and 93% accuracy and 93% F1-score on Source 2, outperforming all baseline models. Subsequently, a thorough analysis of failure cases has been carried out, including the identification of the most challenging groups of moisture levels. These analyses provide valuable insights into the complex task of determining moisture content from inherently heterogeneous wood chips. The proposed MoistViT demonstrates significant potential for real-time applications in relevant industries, which could ultimately lead to a streamlined production process.

7 May 2026

An All-Hazards Return on Investment (ROI) Model to Evaluate U.S. Army Installation Resilient Strategies

Abstract: The paper describes our project to develop, verify, and deploy an All-Hazards Return of Investment (ROI) model for the U.S. Army Engineer Research and Development Center (ERDC) to provide army installations with a decision support tool for evaluating strategies to make existing installation facilities more resilient. The need for increased resilience to extreme weather caused by climate change was required by U.S. code and DoD guidance, as well as an army strategic plan that stipulated an ROI model to evaluate relevant resilient strategies. During the project, the ERDC integrated the University of Arkansas designed model into a new army installation planning tool and expanded the scope to evaluate resilient options from climate to all hazards. Our methodology included research on policy, data sources, resilient options, and analytical techniques, along with stakeholder interviews and weekly meetings with installation planning tool developers. The ROI model uses standard risk analysis and engineering economics terms and analyzes potential installation hazards and resilient strategies using data in the installation planning tool. The ROI model calculates the expected net present cost without the resilient strategy, the expected net present cost with the resilient strategy, and ROI for each resilient strategy. The minimum viable product ROI model was formulated mathematically, coded in Python, verified using hazard scenarios, and provided to the ERDC for implementation.

4 May 2026

An Investigation of Causes of Inaccuracy of Infrared Radiation Cameras for Large-Scale Additive Manufacturing Applications

Abstract: In additive manufacturing, accurate temperature data are needed for both real-time feedback for print operators and understanding the thermomechanical behavior for prediction and part quality characterization. Through the collection of accurate temperature data, thermal models can be validated to predict process-induced properties of parts. Infrared radiation (IR) is used to determine the temperature of a surface. Because IR cameras measure thermal radiation from a distance without contact, they are safe to use in high-temperature environments like 3D printing. An investigation of reported temperature values for multiple cameras during one print showed a decreasing trend for cameras close to the printer’s heat sources, which was not reflective of the printing process, and a discrepancy of ±20°C when printing at 200°C across overlapping camera views. Two more prints were studied to determine whether this camera behavior was unique to that print and geometry. The analysis showed the same results across all three prints, with camera-reported values having inconsistencies for a single layer, a subset of layers, and the scale of the print. Multiple possibilities for the cameras’ variances were explored. The IR cameras were determined to require further calibration and experimentation before reported temperature values can be treated as physical temperature values.

29 Apr 2026

Deep Learning Approaches for Buried Object Detection in Infrared Imagery

Abstract: Artificial intelligence and machine learning techniques are increasingly utilized to detect buried objects in thermal infrared imagery. This task relies heavily on the quality and diversity of the training dataset, requiring datasets that capture variability present in real-world environments. Synthetic imagery offers a means to expose algorithms to a greater range of conditions than is often available in real-world data alone. This study evaluates the performance of three open-source object detection models—Faster Region-Based Convolutional Neural Network (R-CNN), You Only Look Once (YOLOv8), and Single Shot Multibox Detector—trained using real-world, synthetic, and hybrid datasets. Real-world imagery was collected from a single field site, while synthetic data were generated using the Virtual Environmental Simulation for Physics-Based Analysis software suite. Model performance was evaluated using Intersection over Union and confidence scores. Models trained exclusively on synthetic datasets with limited scene diversity, when tested on real-world imagery from the same location, produce high false-positive and false-negative rates. Detection performance im-proved significantly for Faster R-CNN and YOLOv8 when trained using a hybrid dataset combining real-world and synthetic data. Analysis of red-green-blue histograms revealed differences in pixel intensity distributions between real and synthetic imagery, indicating areas for improving synthetic data generation.

29 Apr 2026

Forward Operating Remote Camera for Engineering—Construction Assurance and Monitoring (FORCE-CAM), Generation 1

Abstract: This research delivered a first-generation, real-time construction monitoring capability, enabling visual situational awareness for off-site subject matter experts. The live-streamed and recorded data can be visualized from a remote computer desktop to aid in identifying non-conformance issues during active paving operations or during concrete damage assessment and repair operations. Experimentation on asphalt paving and skid-steer construction equipment using direct electro-optical and thermal sensors provided validation of the efficacy of this solution.

22 Apr 2026

Statistical Analysis of Large Format Additively Manufactured Polyethylene Terephthalate Glycol with 30% Carbon Fiber Tensile Data

Abstract: In large format additive manufacturing (LFAM), a keener understanding of the relationship between the manufacture method and material temperature dependency is needed for the production of large polymer parts. Statistical analyses supported by material properties and a meso-structural understanding of LFAM are applied to elucidate tensile data trends. The data from LFAM polyethylene terephthalate glycol with 30% carbon fiber (CF) (PETG CF30%) panels (diagonal, horizontal, and vertical in the x-y print plane) and injection-molded specimens tensile tested at six different testing temperatures (room temperature, 40 ◦C, 50 ◦C, 60 ◦C, 70 ◦C, and 80 ◦C) were used for statistical analyses. A standard deviation, a coefficient of variation, and a two-way and one-way analyses of variance (ANOVA) were conducted. The manufacturing method (44.2%) and temperature (47.4%) have a strong effect on the ultimate tensile strength, in which temperature (82.6%) dominates Young’s modulus. To explain the difference between the ultimate tensile strength of vertical, diagonal, and horizontal specimens at room temperature, a visual inspection of the specimen failure was conducted and the maximum stress at the crack tip was calculated analytically. The decreased strength in the diagonal specimens resulted from the reliance on interlaminar adhesion strength. Future work will consider the effect of the void space variation on tensile strength variance.

8 Apr 2026

Powering the Monitorization of Uninterruptible Power Supplies

Purpose: The danger of invasive species and the ecological impact on natural environments can be seen throughout the world. In the United States, the invasive species problem is being addressed in the Mississippi River and the tributaries that feed it where invasive carp were introduced and invaded, threatening native species and ecosystems. To battle invasive species’ movement into naïve watersheds, the underwater Acoustic Deterrent System (uADS) was developed by the US Army Corps of Engineers (USACE) to stem the migration of the invasive carp through navigation locks. This project serves as a vital effort to preserve the natural balance of aquatic life in the Mississippi River and those waterways that are connected to it. The project highlights the crucial need for systems that monitor the health of the hardware that keeps the project alive in the event of power failure or other disasters. The ability for researchers to quickly check the health of various systems, receive notification of failure, and see visualizations of hardware data is indispensable when a failure with a poor response time could allow these species to move through points where the system is in place. This paper will discuss the process of using containerization to address the monitorization needs of such systems and how containerization may allow for systems to be created quickly while still allowing for easy access to the needed data.