Tuesday, November 13, 2012

Autonomous Drones; The Next Wave In Robotic War


Image: Terminator hunter-killer drone

By Dan De Luce,


The Pentagon is investing heavily in ‘autonomy’ for robotic weapons, with researchers anticipating squadrons of drones in the air, land or sea that would work in tandem with manned machines – often with a minimum of supervision.
"Before they were blind, deaf and dumb. Now we're beginning to make them to see, hear and sense," Mark Maybury, chief scientist for the U.S. Air Force, said.
Instead of being "in the loop”, humans will be "on the loop”, said Maybury, explaining that operators will be able to "dial in" when needed to give a drone direction for a specific task.
"We're moving into more and more autonomous systems. That's an evolutionary arc," said Peter Singer, an expert on robotic weapons and author ofWired for War.
"So the role moves from being sort of the operator from afar, to more like the supervisor or manager, and a manager giving more and more of a leash, more and more independence," he said.
"These [technological] responses that are driven by science, politics and battlefield necessity get you into areas where the lawyers just aren't ready for it yet," Singer said.
"We're not far away from having a single piloted Apache or other helicopter system and a larger number of unmanned systems that fly with that," said Dahm, a professor of mechanical and aerospace engineering at Arizona State University.
"These require very high levels of machine reasoning. We're much closer to that than most people realise," Dahm said.
The Air Force is planning for scenarios in which different tasks would be divided up among manned and unmanned ‘systems’, with drones jamming enemy air defences, tracking targets and assessing bomb damage, while piloted warplanes oversee the launching of bombs and missiles.
"You have to be able to show that the system is not going to go awry – you have to disprove a negative," Dahm said. "It's very difficult to prove that something won't go wrong."
One veteran robotics scientist, Ronald Arkin, a professor at the Georgia Institute of Technology, believes that countries will inevitably deploy independent robots capable of killing an enemy without a human pushing a button.
Arkin, who has worked on U.S. defence programs for years, argues that robotic weapons can and should be designed as “ethical" warriors, with the ability to distinguish combatants from innocent civilians.
Without emotions to cloud their judgment and anger driving their actions, the robots could wage war in a more restrained, "humane" way, in accordance with the laws of war, Arkin said.
"It is not my belief that an unmanned system will be able to be perfectly ethical in the battlefield, but I am convinced that they can perform more ethically than human soldiers are capable of," he wrote.

No comments: