Army looks towards gaming & big data to create prepared future force

In a TRADOC Mad Scientist presentation, Dr. Robert Smith, a senior research engineer at the U.S. Army Tank Automotive Research Development and Engineering Center (TARDEC), laid out his vision for an army which uses big data to improve its odds on the battlefield.

With the amount of data generated in simulated environments and through learning systems aboard autonomous technologies, Smith believes the Army can better understand even the most minute behaviors and patterns of soldiers and generate lessons from them, allowing for improved training and technologies. These concepts are ones that have been accepted and refined within the commercial video game, sports, and autonomous car development communities.

Smith cited William Roper, the director of the Strategic Capabilities Office (SCO) in the Department of Defense, who said some of the top technology companies in the world believe the organization with the most data is going to be able to train the most intelligent machine. Through programs in their early stages of development, he believes the Army can start taking advantage of this fact.

Specifically, Smith discussed the ways in which this data could be used by those who acquire technologies. According to his research in 2008, 30 percent of DOD acquisitions programs did not include end users of technologies in their development. He quoted Gen. David Perkins, commanding general of TRADOC, who cited “the soldier-technology interface” as a necessary advantage on the battlefield, but one which could be improved through the earlier interactions of servicemembers with platforms.

Smith believes that Operation Overmatch, an early synthetic prototyping effort, is one way to generate this data and bring about the cross-learning of soldiers with technology developers. The platform takes accurate models of early stage technologies and places them within a virtual gaming environment accessible to soldiers through their computers. TRADOC takes the data gleaned from players in game and analyzes behaviors, technologies, and tactics with the aim of refining capability requirements and platforms. Overmatch is still in alpha testing, but Smith believes the game will eventually result in roughly 12 million hours of gameplay per year.

From these data points, Smith envisioned a future Army which would include these same features in training and the employment of technologies. “The future complex world is going to be so complex that there is no training memo you can write...like you can in the past...if you’re going to chase a bad guy into a large building in a megacity on the 40th floor, there might be 40 different ways of getting that bad guy,” Smith said, but with big data, the best ways might rise to the top.