Award Data

DeepVerify is a suite of tools to verify and certify robustness of different types of AI applications with a specific focus on currently popular neural networks such as deep CNN, reinforcement learning, recurring neural networks. However, our tools can be deployed on other types of AI/ML frameworks as well. Upon automated in-depth assessment of AI applications and exploration of input data space a ...

To address DARPA’s need for a lightweight, wearable laser alert sensor system, Intellisense Systems, Inc. (Intellisense) proposes to develop a new Laser Irradiance Sensor and Alert (LISA) system based on a multi-aperture detector array that covers a 2-pi steradian (360 by 90 degree) field of view (FOV) and distinguishes laser sources from non-laser sources by the correlations in intensity among ...

Attollo Engineering proposes to design and build a TRL6/TRL 7 prototype laser sensing device spanning the spectrum from 450 nanometers (nm) to 1600nm that is man or vehicle mount capable.  Attollo has a TRL 8/9 device called the OMNir (https://www.attolloengineering.com/laser-sensing.html) that senses laser designators (1064nm) and laser rangefinders (1550nm) designed originally for fratricide av ...

Lasers present a threat to military forces. They can damage the eyesight of dismounted soldiers, operators of ground vehicles and pilots, as well as damage EO/IR sensors. Existing solutions operate over a narrow spectral band, have small field-of-view (FOV), have large size, weight and power (SWaP) and do not meet the requirements of the solicitation. Trex proposes a wearable Laser Alert System (L ...

Underwater Acoustic Modems (UW-A) have seen significant R&D and commercial product introductions in recent years. Standards including JANUS (NATO) and JSATS (USACE), have emerged.  Yet, interoperability remains limited due to the generally proprietary nature of commercial products, the complexity and vulnerability of high bandwidth options, and high costs/size/power consumption limiting modem use ...

Prediction of complex dynamical systems is an important task that has implications in a wide variety of technological and policy domains.  Accurate prediction of processes such as terrestrial and space weather, climate change, financial and economic markets, and pandemics can have profound policy and engineering impacts.  As a result, diverse methods of time series prediction have been developed ...

Software and hardware flaws can be exploited to make programs perform unintended computations or leak sensitive data. We propose to counter these threats by isolating libraries and other program units inside a single process. The developer will insert source-level annotations that i) map code and data units to compartments and ii) capture how each compartment is intended to interact with others, i ...

Hypersonic vehicles will be flying in conditions that subject their structure and propulsion systems to harsh environments not previously seen by other aircraft. Conventional instruments cannot survive in these extreme environments and do not provide reliable data. optoXense proposes to design and develop a new pressure measurement system that will survive harsh (2000psi, 1800°C) hypersonic envir ...

Clostra proposes to create a tool which will allow instructors or content developers to select 2D videos with which neural networks can create 3D immersive content. This tool will also permit instructors to add figures and objects to the environment. This will dramatically decrease the time required for the U. S. military to create challenging training scenarios across a range of environments. Thi ...

Localization of over water airborne contacts by a ship at sea or a land-based collector, typically requires either active sensing of the target or a combination of passive measurements from two or more cooperative platforms sharing information about either the time and frequency of arrival or angle of arrival of the RF energy from the aircraft. In both of those cases the friendly platforms are una ...