Marines will soon share the battlefield with gun-wielding robots, swarms of micro drones, driverless cars and self-unloading supply aircraft — and the Corps wants the machines to do more of the thinking.
Learning to trust autonomous machines will be one of the key tasks for U.S. troops over the next 20 years, according to a panel of industry experts at the Sea-Air-Space expo outside Washington.
The future of human-machine interaction will fundamentally transform the way the military does business, said retired Brig. Gen. Frank Kelley, the deputy assistant Navy secretary for unmanned systems.
“One of the things we want to emphasize: We are not trying to replace Marines and sailors with unmanned systems — that is not the goal,” he said during the Tuesday panel. “We need to optimize unmanned systems and think about to what degree we are going to allow these systems to be fully autonomous.”
Marines already incorporate an array of automated systems into their missions, but many — like pocket-sized reconnaissance drones or explosive ordnance disposal robots — need troops to operate them. Autonomous machines wouldn't require the operator to constantly manage their systems, said Col. James Jenkins, director of science and technology at the Marine Corps Warfighting Laboratory.
“When bullets start flying, the Marine either becomes so absorbed in operating the robot that he loses sight of what’s happening around him, or he drops the controller and becomes a rifleman,” he said.
The goal for the Marine Corps, in line with its future operating concept Expeditionary Force 21, is to be able to integrate machines that act independently after a squad or company commander issues them a task. Ideally, Jenkins said the Marines would be able to give the machines orders that support their missions, and the robots would just go off and do them.
“These systems will interface with a human just like a subordinate fire team leader who goes back to their squad leader when they have something to report or need new orders,” he said.
The ability to deploy such autonomous entities ultimately comes down to a question of trust in the human-machine team: the individual system, the people who develop the systems and in the people who actually operate them, Kelly said.
That raises questions about how much trust Marines should place in inanimate objects, though, Jenkins said.
“If we're talking about killer robots, do I give that robot a life or death decision without a person in the loop to make that decision?” he asked.
Matthew L. Schehl covers training and education, recruiting, West Coast Marines, MARSOC, and operations in Europe, Africa and the Middle East for Marine Corps Times. He can be reached at email@example.com