28 Nov 2023

Discover ERDC 101 and 201 Training Modules User’s Guide

Abstract: Discover ERDC is a web-based tool that functions as a knowledge management hub by enabling employees of the US Army Engineer Research and Development Center (ERDC) to access valuable resources such as detailed employee profiles, organizational details, and links to other knowledge stores. This document covers the update of the ERDC 101 and 201 video player systems, the addition of a training component to those modules, and the integration of the systems into Discover ERDC. The updated video systems contain a collection of onboarding video presentations that give new employees critical information about their careers at ERDC. In addition, Discover ERDC 101 and 201 provide progress-tracking mechanics for asynchronous learning, as well as the ability to certify that employees have completed the training modules. This document serves as a user guide for these tools, providing an overview of the content and functionality.

28 Nov 2023

Investigation of Steam Adsorption Chillers to Modernize Existing Central Steam Plant Systems

Abstract: This report investigates the integration of steam adsorption chillers as a modernization strategy for conventional central steam plant systems. Our objective is to assess the feasibility, advantages, and challenges of incorporating steam adsorption chillers into existing steam plant setups to enhance energy efficiency and cooling capabilities. Central steam plant systems have historically been used for steam-based heating but often lack cooling capabilities, necessitating additional cooling infrastructure. Steam adsorption chillers offer a potential solution by using waste steam for cooling, optimizing energy utilization and reducing reliance on traditional cooling methods. Through a comprehensive analysis, this report evaluates the technical compatibility and potential cost implications of implementing steam adsorption chillers. It explores factors such as system integration, operational dynamics, and maintenance requirements to provide a holistic view of the feasibility and benefits of this modernization approach. The findings aim to offer valuable insights to decision-makers and Army facility managers seeking innovative ways to upgrade central steam plant systems. By considering the technical and economic aspects of adopting steam adsorption chillers, this report contributes to the knowledge base for sustainable and efficient energy utilization in central plant operations.

8 Nov 2023

AI on Digital Twin of Facility Captured by Reality Scans

Abstract: The power of artificial intelligence (AI) coupled with optimization algorithms can be linked to data-rich digital twin models to perform predictive analysis to make better informed decisions about installation operations and quality of life for the warfighters. In the current research, we developed AI connected lifecycle building information models through the creation of a data informed smart digital twin of one of US Army Corps of Engineers (USACE) buildings as our test case. Digital twin (DT) technology involves creating a virtual representation of a physical entity. Digital twin is created by digitalizing data collected through sensors, powered by machine learning (ML) algorithms, and are continuously learning systems. The exponential advance in digital technologies enables facility spaces to be fully and richly modeled in three dimensions and can be brought together in virtual space. Coupled with advancement in reinforcement learning and computer graphics enables AI agents to learn visual navigation and interaction with objects. We have used Habitat AI 2.0 to train an embodied agent in immersive 3D photorealistic environment. The embodied agent interacts with a 3D environment by receiving RGB, depth and semantically segmented views of the environment and taking navigational actions and interacts with the objects in the 3D space. Instead of training the robots in physical world we are training embodied agents in simulated 3D space. While humans are superior at critical thinking, creativity, and managing people, whereas robots are superior at coping with harsh environments and performing highly repetitive work. Training robots in controlled simulated world is faster and can increase their surveillance, reliability, efficiency, and survivability in physical space.

7 Nov 2023

Integrating MOVEit Motion Constraints on a Novel Robotic Manipulator

Abstract: MOVEit, a widely used Robot Operating System framework, plans composite tasks, where the high-level sequence of actions is fixed and known in advance. However, these tasks need to be tailored and adapted to the environmental context. This framework uses custom trajectory planners, known as controllers, to solve goals that are fully defined within the configuration space. Libraries, such as the Open Motion Planning Library, provide a collection of motion planners that can solve task-space goals. An exact spatial and joint replication of the robotic manipulator’s mechanics, typically Universal Robot Description Format and Semantic Robot Description Format files, is required. Common arms such as the Panda-Manipulator and OpenMANIPULATOR-X provide these files in their respective public repositories, but custom arms require significant modification or even a complete rewrite of these files.

29 Sep 2023

Low Size, Weight, Power, and Cost (SWaP-C) Payload for Autonomous Navigation and Mapping on an Unmanned Ground Vehicle

Abstract: Autonomous navigation and unknown environment exploration with an unmanned ground vehicle (UGV) is extremely challenging. This report investigates a mapping and exploration solution utilizing low size, weight, power, and cost payloads. The platform presented here leverages simultaneous localization and mapping to efficiently explore unknown areas by finding navigable routes. The solution utilizes a diverse sensor payload that includes wheel encoders, 3D lidar, and red-green-blue and depth cameras. The main goal of this effort is to leverage path planning and navigation for mapping and exploration with a UGV to produce an accurate 3D map. The solution provided also leverages the Robot Operating System

20 Sep 2023

Mapping and Localization Within a Mock Sewer System

Abstract: Herein, we explored a robot’s ability to localize and map, both in simulation and on a physical robot, within a mock sewer system. Mapping and localization techniques were first developed and tested in simulation and were then transitioned to the actual robot for additional physical testing. Several odometry and simultaneous localization and mapping (SLAM) techniques, including gmapping, SLAM toolbox, elevation mapping, and RTABMap, were evaluated for this particular environment. The results of the odometry and the various SLAM approaches are discussed in detail.

18 Sep 2023

Accelerated Corrosion of Infrastructural Seven-Strand Cables via Additively Manufactured Corrosion Flow Cells

Purpose: The purpose of this project was to generate an accelerated corrosion methodology capable of producing seven-strand cables with simulated corrosive defects for calibration of nondestructive analysis (NDA) techniques. An additively manufactured accelerated corrosion cell was motivated and designed. Previous attempts at accelerated electrochemical corrosion used a large cable area with a current density that was too low (i.e., 1 A/m²)* to effectuate efficient corrosion. The accelerated corrosion cell presented here takes advantage of the restricted area within the corrosion flow cell to maximize the corrosion rate in a consistent and calibrated manner (i.e., 2,000 A/m²).

28 Aug 2023

Scaling and Sensitivity Analysis of Machine Learning Regression on Periodic Functions

Abstract: In this report we document the scalability and sensitivity of machine learning (ML) regression on a periodic, highly oscillating, and 𝐶∞ function. This work is motivated by the need to use ML regression on periodic problems such as tidal propagation. In this work, TensorFlow is used to investigate the machine scalability of a periodic function from one to three dimensions. Wall clock times for each dimension were calculated for a range of layers, neurons, and learning rates to further investigate the sensitivity of the ML regression to these parameters. Lastly, the stochastic gradient descent and Adam optimizers wall clock timings and sensitivities were compared.

18 Aug 2023

Unmanned Ground Vehicle (UGV) Path Planning in 2.5D and 3D

Abstract: Herein, we explored path planning in 2.5D and 3D for unmanned ground vehicle (UGV) applications. For real-time 2.5D navigation, we investigated generating 2.5D occupancy grids using either elevation or traversability to determine path costs. Compared to elevation, traversability, which used a layered approach generated from surface normals, was more robust for the tested environments. A layered approached was also used for 3D path planning. While it was possible to use the 3D approach in real time, the time required to generate 3D meshes meant that the only way to effectively path plan was to use a preexisting point cloud environment. As a result, we explored generating 3D meshes from a variety of sources, including handheld sensors, UGVs, UAVs, and aerial lidar.

17 Aug 2023

A/E/C Graphics Standard: Release 2.2

Abstract: The A/E/C Graphics Standard Release 2.2 has been developed by the Computer-Aided Design/Building Information Modeling Technology Center to document how proper hand-drafting practices can be achieved in advanced modeling. It is through the collection and documentation of these practices that consistent models and drawings shall be achieved throughout the US Army Corps of Engineers (USACE), as well as other federal agencies. In the collection of these practices, various historical USACE District drafting manuals were consulted and compared against practices contained in industry and national standards, with consideration toward whether software can achieve those practices. The documentation of these practices will help to achieve both clear and aesthetically pleasing construction documents.