How to make decisions repeatable at the military tactical edge

Missions require innovative approaches to unlocking the data that maintains battlespace advantage. By leveraging key technologies such as API-based data interch...

“The best way to get from point A to point B is to already be at point B,” General John M. Murray told attendees of the Global Force Next symposium, hosted by the Association of the United States Army (AUSA) in March 2021. He was speaking about delivering equipment to soldiers in the field. But his words apply equally to data.

Military missions are increasingly dependent on processing and contextualizing data. To create an advantage in all battlespace domains, mission teams must manage and exploit proliferating data sources, data types and data volumes, and share information in multiple directions across U.S. forces and coalition partners.

Much of the data is already at point B, generated at the tactical edge. The challenge is processing and analyzing it to make informed decisions within a short response window, while the data is still accurate and relevant.

Traditionally, data is sent to a centralized datacenter that can handle voluminous data, with information and insights then transmitted back to the field. But that process is too slow and cumbersome for today’s battlespace, especially in denied, disrupted, interrupted and limited (DDIL) edge environments. Instead, teams need to gather, analyze and generate decision-support information closest to the warfighter.

Fortunately, new technologies and approaches can overcome the limitations in processing, bandwidth and other degradation constraints of edge environments. API-based data interchange, microservices architectures, artificial intelligence and machine learning can all be key enablers in data-driven decision support that is automated and repeatable at the edge.


Achieving interconnected data availability

The volume and types of mission are increasing from various sensors and systems – some new, some potentially decades old. That requires dealing with larger and larger datasets. What’s more, mission and warfighter needs change unexpectedly, requiring immediate integration of new data sources and stakeholders. Those demands call for ways to use and share data while constantly recomposing mission systems and capabilities.

Achieving rapid data integration within disaggregated forces requires three key technologies: automation, API-based data interchange and microservices.

First, automated systems can quickly process and analyze massive amounts of data and then prioritize that data, increasing the speed, relevance and effectiveness of information exchange.

With API-based data interchange, data-generating sensors produce data in their proprietary format while simultaneously translating it into a standard API format. Data-receiving systems then accept the API-format data and translate it into their own proprietary format. The result is only two translations, no matter how many datatypes are involved.

But data integration isn’t the only hurdle to overcome. Mission teams also need a way to dynamically adapt their edge infrastructures as tactical situations change in real time. The solution is a microservices architecture, which implements applications as a collection of loosely coupled software services that can be deployed rapidly and independently. The result is modular, interoperable capabilities that can be rapidly assembled and recomposed.

By deploying microservices, organizations can take advantage of containers at the edge, combining each application with its supporting code and data into a single package. When coupled with an automation platform, warfighters can transition from reactive to proactive ops at the edge. This approach can create real-time actionable information and recommendations, better predictive analysis, and real-time decision support.


Leveraging AI and ML for decision support

Of course, it’s not enough to simply make data available at the edge. Mission teams need a way to separate the signal from the noise and look only at information that’s timely, relevant and most likely to drive real-time decision support.

Data-centric collaboration powered by AI and ML at the edge can provide mission teams with a single collaborative environment to access geospatial data combined with the latest analytic models and tools. These tools can, for example, automatically tag data by type, source and purpose as it’s generated. The tags can include start and end dates for when the data is relevant; beyond that timeframe, the data is allowed to expire. The result is a data armory at the edge – a datastore that’s pre-positioned, rectified and prioritized.

Teams can further enhance data utilizing AI to intelligently extrapolate where data is coming from and what it’s intended for. That enables systems to deliver the right data to the right people in the right location at the right time. Stakeholders see only the data they need at the time they need it, enabling decision-quality information.

AI is already proving itself at the edge. For instance, unmanned aerial systems (UAS) are being tested with advanced software that enables AI software tools to operate within the low size, weight and power (SWaP) constraints of the extreme edge. The UAS uses onboard sensors and AI-enhanced software to detect and better classify military targets. In practice, the result is enriched intelligence, surveillance and reconnaissance (ISR) data and improved situational awareness for decision-makers.

In an even more extreme edge environment, NASA is leveraging analytical tooling for scientific analysis on the International Space Station (ISS). Using a containerized environment to conduct AI analysis at the edge will reduce the dependence, delays and failure rates of transmitting large datasets to Earth-deployed assets. By transmitting only high-quality information back to Earth, the organization will accelerate time to value on actionable results.

Missions require innovative approaches to unlocking the data that maintains battlespace advantage. By leveraging key technologies such as API-based data interchange, microservices architectures, AI and ML, warfighters can make decision support more automated and repeatable at the edge.

Travis Steele is principal chief architect, U.S. Air Force & U.S. Space Force, for Red Hat.


Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.

Related Stories