Smart Warrior Analytics Enable Faster Decisions

While operating at the tactical edge, warfighters must quickly make life-or-death decisions. They rely on the intelligence available to them via communications, observation, and other sources. However, as warfighting systems become more interconnected, massive amounts of data are available at any given point in time—resulting in information overload. 

Adding to the data challenge, rival nations have publicly demonstrated an increasing use of technologies like warfighter-worn sensors, artificial intelligence-aided decision chains, and algorithmic warfare. This makes quick, accurate decisions by allied warfighters even more imperative.

To stay ahead of the adversary, the Department of Defense (DOD) must empower warfighters with technology at the tactical edge that puts intelligence into context, informs decisions, and connects military teams with relevant information. From fully connected sensor suites and edge processing to artificially intelligent wearables and robotic teammates, a vast breadth of technologies is available. 

But loading warfighters up with massive amounts of data, and more and more technology and hardware leads to increased physical and cognitive load and opens them up to fatigue, distraction, and vulnerability. With near-peer adversaries operating at machine speed, how can the human warfighter outthink the enemy? 

Modern warfighting increasingly operates at a pace faster than what human analysis and decisions can support unaided. Within intelligence analytics centers, technologists rely on machine learning (ML) to automate and augment human analytical capabilities. 

The same is true at the tactical edge. The nation's warfighters need a fully connected technology suite that empowers the human as a mission-adaptable platform and provides only the most important insights and analytics when and where needed—what we call Smart Warrior Analytics—to enable rapid decision making and ensure optimization of warfighter performance. In the age of algorithmic warfare, this suite empowers the warfighter to act with precision while decreasing distraction and stress caused by cognitive and physical overload.

Body-worn sensors feed into a common operating picture that in turn delivers better intelligence to inform operational decisions. This model is similar to the familiar example of commercially available smartphone-based navigation apps, where crowd-sourced geolocation data seamlessly integrates with satellite-based route information to help drivers avoid traffic en route to their destination. 

These warfighter-worn sensors integrate into the existing system to gather information like biometrics, environmental data, acoustics, thermal images, and video in an unobtrusive way. A small body-worn computer combs through this captured data to process it, analyze it using AI algorithms, and determine what the warfighter needs to see to rapidly make decisions and identify threats. 

This tech suite is a system of sensors, computation, and power built on open architectures that enable warfighters to gather and analyze data at machine speed while being fully connected to the machine learning operations (MLOps) pipeline to distribute insights instantly.

There are many ways that ML could be used at the tactical edge—from managing power usage across soldier-worn systems to extend battery life to changing the heads-up display inside a soldier’s helmet to either increase situational awareness or streamline information delivery for combat situations.  

ML algorithms have the capacity to learn and improve over time—but doing so requires a feedback loop. This is easy to achieve with MLOps at an analytical center, but deploying an algorithm to thousands of operators at the tactical edge and returning feedback to strengthen the model is a challenge. This data in motion is also more vulnerable to cyber attack and requires significant bandwidth to relay. 

The dismounted warfighter has to have unobtrusive systems to gather the data (electro-optical and infrared cameras, acoustic sensors, biometric sensors), analyze it with little computational power, and deliver feedback to the MLOps pipeline. Power is also a key consideration to enable these on-body platform capabilities. The use of emerging technologies to include novel textile-based and power harvesting solutions is critical to mitigating the physical load. While these are not easy tasks to achieve, the result is a warfighter equipped with unprecedented compute power, using technology that helps to make decisions at the speed of war. 

Thanks to the open architecture approach, warfighters are able to integrate additional hardware seamlessly and share resources across the human platform, reducing the physical load while remaining connected at the tactical edge. Unlike integrating new technologies into proprietary, closed, or not truly open architectures, this approach enables DOD to get away from piecemeal, disconnected equipment and software while allowing the warfighter to tailor the technologies to fit the mission need. 

The warfighter has a common operating system and can seamlessly integrate products—whether electronic textiles, equipment, hardware, or software—from multiple vendors to create a body-worn technology suite. This common foundation reduces acquisition and connectivity challenges for the program office while speeding insight delivery and decision making for the warfighter. 

An urgent truth underscores the need for Smart Warrior Analytics: Information superiority is essential to warfighting dominance. That means that protecting our data, especially when it is at the tactical edge, is of paramount importance. 

Keeping warfighters safe requires ongoing protections against algorithmic warfare. As more AI systems are used to inform battlespace decisions, it becomes more important that we understand and trust the algorithms—and know that they haven’t been compromised. 

Integrating smart technologies into the tactical edge requires a security-first approach, one that recognizes that cyber protections are an immutable component of battlespace readiness. Booz Allen’s approach is to “bake in” cyber protections during system design—creating inherent resilience with a cyber approach that can be updated as the threat landscape changes. This also allows for the integration of new protections against adversarial attacks on algorithms.

The nation's adversaries are getting smarter, often out-investing the U.S. in critical technologies like AI and ML, and sometimes taking the human out of the decision loop altogether. DOD must find a path forward that takes into account the ethics of the country and the need to quickly develop and train algorithms to operate instantaneously.

In terms of initiatives like Joint All-Domain Command and Control (JADC2), smart warrior technologies shorten the chain from sensor to effector and make warfighting forces more responsive, agile, effective, and survivable. 

Achieving overmatch requires a clear focus on how to develop, train, and maintain up-to-date algorithms to be brought into operations faster, to integrate new technologies seamlessly, and to reduce the physical and cognitive load on warfighters.