At the core of the project is the idea of using machines to identify launcher-shaped objects buried within the staggering amount of digital imagery collected by US spy satellites, manned and unmanned aircraft. A senior official in the Department of Defense explained that it is this vast amount of data that makes manual research inefficient. "What was largely a manual process for intelligence analysts has to become an automated one," he said, cited by Defense One. The ultimate goal is to train computers spot what are called transporter-erector-launchers (TELs). North Korea used these kinds of launchers during missile tests conducted over the last few months.