Award Data

This project aims to create a general-purpose readout architecture that will allow the rapid deployment of next generation detection systems. The system will be based on the development of a programmable general-purpose integrated circuit (GPIC) that has the front-end electronics required to read out signals from a variety of radiation detectors, especially next generation scintillators and semico ...

The Nu-Trek team is proposing to develop µDet, a low Size, Weight, and Power (SWaP) read out integrated circuit (IC) for gamma and neuron detectors. µDet offers pulse shape digitization, which in turn enables gamma-neutron discrimination. This is a game changing capability that brings laboratory-level functionality to the field. In Phase I the Nu-Trek Team will develop a baseline design for the ...

The Synthetic Aperture Radar (SAR) Image Generation Data Augmentation (SIGDA) system is achieved using SAR simulators and the Arete’s Reification approach. Large, realistic datasets will be generated using the Arete Reification capability. These large Reified datasets are then used to train machine learning or Artificial Intelligence (AI), Automatic Target Recognition (ATR) classification algori ...

Current radiation detection algorithms are based on the concept that each detector operates independently. The Probitas Project, Inc. (Probitas) and the Lawrence Berkeley National Laboratory (LBNL) propose to show the benefits of data fusion to improve the identification, localization, and characterization of a radioactive source in a complex scene as compared to a singular detector algorithm. We ...

This STTR Phase I project proposes novel deep learning models for generating realistic multi-spectral remote sensing imagery, specifically in the infrared (IR) and near-infrared (NIR) bands. The proposed system enables synthesis of semantically realistic imagery and provides parametric control of synthesizing objects-of-interest, type of terrain and land cover, time or season, weather, cloud cover ...

The overall goal is to develop numerics-informed neural networks (NINNs) and DeepOnets for chemical reactions and for PDEs with spatial derivatives improve the computational efficiency of the chemical kinetics models for chemical weapon agents and simulants. Based on the first NINN developed by the Karniadakis’s group in 2018, which blends the multi-step time-stepping with deep neural networks, ...

To address the challenge DTRA faces in identifying rare objects of interest to defeat improvised threat networks using multispectral imagery, small business ARiA and research institution Michigan Technological University (MTU) will develop and demonstrate the feasibility of the Generative Augmentation Process (GAP). The Phase I effort will (1) conduct a proof-of-concept study for GAP by developing ...

The project will develop numerically inspired deep neural nets (NINNs) in order to replace the stiff ordinary differential equation (SODE) solvers currently being used to integrate chemical species in high-fidelity computational fluid dynamics simulations. Unlike traditional deep neural nets, the architectures and optimization strategies used to learn the physics of a problem will be based on the ...

Improving the effectiveness of counter-WMD operations requires improved understanding of weapon-target interaction. Specifically, time-resolved measurements of temperature and composition are required to allow temporal evolution of a detonation fireball. To address this need, SA Photonics will develop MONITOR, a laser-based temperature diagnostic that will enable wide dynamic range temperature mea ...

This STTR Phase 1 project proposes development of a statistical relational learning framework for identification of the function of laboratory equipment from imagery. The proposed framework uses semantic reasoning to incorporate evidence from multiple classifiers and feature extractors, domain knowledge, and scene context for scene understanding and labeling. The Phase I effort will focus on imple ...