h+ Magazine

It’s all in the legs. Boston Dynamics’ new PETMAN is billed as an anthropomorphic robot for testing chemical protection clothing used by the U.S. Army. Now available as a prototype, PETMAN is built from the same balancing technology that makes the four-legged mule-like BigDog so formidable in hostile terrain – except that PETMAN has two legs:

While this new robotic leg technology is both scary and impressive, PETMAN is a far cry from eight-foot-tall Terminator T-600 model from the Terminator movie franchise. In the movies, the rubber-wrapped T-600s use their somewhat human-like appearance to get high-caliber weapons into striking range.

The finished PETMAN –- expected in 2011 –- is being designed to mimic human physiology, for example sweating in response to temperature and humidity changes, to make it a realistic testing device for the chemical protection suits. At a top speed of 3.2 mph, the PETMAN prototype is not going anywhere too fast –- yet.

But what exactly is the Army up to? A recent Time article quotes U.S. Lieut. General Michael Vane, who directs the Army Capabilities Integration Center in Fort Monroe, Va.: "I am starting out with the idea of having a technology-enabled human. [But] we might someday come up with [separate] IT doctrine and robot doctrine... We want to make the people or humans in charge under command and control in a 'whole of government' approach."

The conflicts in Iraq and Afghanistan have led the Army to increasingly look for rapid delivery of anything that can lighten a soldier's load. P.W. Singer, a Senior Fellow at Brookings Institute, characterizes the current state of U.S. military robotics in a TED talk earlier this year (see the h+ interview with Singer, “Wired For War or How We Learned to Stop Worrying and Let Dystopian SF Movies Inspire our Military Bots” in Resources):

An Army Robotics Strategy White Paper published earlier this year makes it clear that the Army has big plans for robotics systems:

1. Reduce risks to soldiers in war through IED detection and neutralization,
2. Reduce the workload on soldiers by having robots perform routine tasks and tasks that required sustained “high tempo” operations, such as routine surveillance of bases, and
3. Enable extended range for unattended ground sensors for mobile reconnaissance.

The White Paper also makes the case that, “many of the robotics systems in use by soldiers and small units in Iraq and Afghanistan have proven their worth – they have saved dozens, perhaps hundreds of lives.” It sets the following priorities:

• Reconnaissance and surveillance
• Target Identification and Designation
• Counter-Mine Warfare
• Chemical, Biological, Radiological, Nuclear, Explosive Reconnaissance

Daniel Goure, vice president of the Lexington Institute, a public policy think tank, questions the ability of robotic systems to distinguish between a friendly ally, a local civilian, or a hostile fighter. "It is tough enough for us to train human soldiers to distinguish between combatants and noncombatants on the battlefield. It is much more difficult to write software that does that," suggests Goure.

Dr. Ronald Arkin, director of the Mobile Robot Laboratory at Georgia Tech and under contract with the U.S. Army, is working on ethical guidance software for battlefield robots that is intended to help with such delicate battlefield discrimination. (See the h+ interview with Dr. Arkin, “Teaching Robots the Rules of War” in Resources) “I don't believe there is any fundamental scientific limitation to achieving the goal of these machines being able to discriminate better than humans can in the fog of war, again in tightly specified situations,” says Arkin.

The PETMAN prototype is literally all legs -– there is no cognitive processing. The autonomous T-600 Terminators, like the fictional Cylons from the TV series Battlestar Galactica, have sophisticated AI software to guide their actions. Robotics expert Dr. Noel Sharkey, a computer science professor at the University of Sheffield is skeptical that such intelligence can be achieved. In a New Scientist interview, he suggests that AI is a dangerous myth that could lead to a dystopian future of unintelligent, unfeeling robot caregivers and soldiers. “I'm an empirical kind of guy, and there is just no evidence of an artificial toehold in sentience,” says Sharkey. “It is often forgotten that the idea of mind or brain as computational is merely an assumption, not a truth. When I point this out to ‘believers’ in the computational theory of mind, some of their arguments are almost religious. They say, 'What else could there be? Do you think mind is supernatural?' But accepting mind as a physical entity does not tell us what kind of physical entity it is. It could be a physical system that cannot be recreated by a computer.”

This is an ongoing debate in both cognitive and computer science. Proponents of strong AI in robotics, such as Hans Moravec and Ray Kurzweil, see a different future than Sharkey, one in which computer processing speed will eventually overtake that of the human brain, and in which human consciousness can be simulated electronically and uploaded into a robotic substrate.

The Army’s goal of using “technology-enabled humans” to reduce risk to soldiers on the battlefield certainly does not preclude a possible future of robotic warriors similar to the fictional T-600 Terminators or Cylon Centurions. But, regardless of which side of the strong AI debate you’re on, it’s clear that PETMAN’s robotic leg technology makes military bots start to look more like bipedal humans than four-legged mules.