Wednesday, October 28, 2009

Failure To Field The Right Robots Costs Lives

(via National Defense Magazine)

Lt. Gen. Rick Lynch, commanding general
of Fort Hood, Texas, and the Army’s 3rd
Armored Corps, has lost 155 soldiers in
combat in Iraq. The lives of 80 percent,
or 122 of them, would have been saved if
the right robots had been in place, he said
at a recent industry conference.

“There’s got to be a sense of urgency
here. I am so tired of going to
[technology] demonstrations,” Lynch

Not only has the Army failed to field the
right kinds of robots that would save war
fighters’ lives, it has not fielded any
robots to Iraq or Afghanistan at all, he
said at the Association for Unmanned
Vehicle Systems International conference
in Washington, D.C.

That claim may have come as a shock to those at the conference. Hundreds of unmanned aerial
vehicles fly over the two theaters of operations today. The exploits of the explosive ordnance
disposal teams and the robots they employ to clear roadside bombs are well known. But Lynch
doesn’t consider those devices robots because they require humans to control them remotely.

“I’m talking about a system that has a certain degree of autonomy,” he said.

Lynch has unique credentials he can use to back up his claims. He has commanded troops in
combat and holds a master’s degree in robotics from the Massachusetts Institute of Technology.

He came to the conference in the nation’s capital to deliver a few blunt messages: that the
military has failed to field life-saving robotics technologies; research and development dollars are
being spent unwisely; and many of the robotics systems that could be saving lives are mature
enough to be fielded today.

“Senior leaders are told they have robots in the battlefield. So the senior leaders think we don’t
need any more,” he said.

Robot developers expressed support for Lynch.

“I love Gen. Lynch because he brings his absolute passion as a commander… for how important
it is for us on the acquisition side to get the tools in the hands of the war fighter,” said Marine
Corps Col. James Braden, project manager at the robotics systems joint project office. His
organization is responsible for fielding non-EOD ground robots in the military.

Braden said he has had several philosophical discussions with Lynch on the definition of the word
“robot” and whether any had been fielded in Iraq or Afghanistan.

Lynch likened the military’s research and development robotics strategy to fertilizing a flower
bed. If one doesn’t have enough fertilizer, it’s not productive if one spreads it so thin that nothing
will grow. The gardener has to concentrate his limited resources on a few flowers, he said.

There is so much money being thrown at high-flying ideas that essential life-saving robotics
technologies are not receiving the dollars they need, he added.

“Everybody has got an idea and I’ve seen these ideas for the past 25 years,” Lynch said. “You
can’t make everybody happy. You got to be able to say, ‘That’s interesting, but not that
interesting. But, by God, this is important.’”

There are four applications where the military needs to place all its robotics research and
development funds, he said.

The first is route clearance.

The armored Buffalo and Husky vehicles are excellent at clearing routes, he said. The problem is
that they have soldiers inside them.

“So when they come across that [improvised explosive device] and the IED detonates, there is a
real good chance that a soldier was killed or severely injured,” Lynch said. “We’ve got to get
those kids out of those route clearance vehicles, and I know the technology today allows us to do

Col. Greg Gonzalez, project manager of the Army’s unmanned aerial systems office took
exception to Lynch’s assertion that unmanned systems hadn’t been applied effectively to the
IED problem.

Lynch’s assertion ignores the work that unmanned aerial vehicles have done in spotting
insurgents planting roadside bombs and then calling in ground forces to react, Gonzalez said at
the conference. “To say that we haven’t provided funding or capability to do those types of
things is a little shortsighted.”

The second application Lynch wants to see more focus on is persistent stare.

To defeat the IED threat, soldiers are tasked to keep watch on “hotspots” — places where
insurgents repeatedly go back to plant the roadside bombs.

Lynch said seven soldiers under his command were killed and three were captured because they
were out watching IED hotspots. “That didn’t have to happen.”

Robots can take the soldiers’ places, he said. They can continuously keep watch on an area, and if
nefarious activity is spotted, “We can take appropriate action. … We can kill those bastards
before they plant the IEDs,” he added.

That includes mounting a weapon on the robot, he said.

Armed ground robots have sparked controversy during the last two years. Army combatant
commanders have asked for armed tele-operated ground robots, and one version, the special
weapons observation remote reconnaissance direct action system, or SWORDS, was fielded. The
concept was to have the robots venture out into exposed areas and fire weapons at targets
without exposing troops on the ground to danger. However, these robots were never used as
intended. Senior Army officers ordered that they be stationary and placed behind sandbags.

Lynch insisted after the speech that this technology is ready to be used.

“There’s a resistance saying that armed ground robots are not ready for the battlefield. I’m not
of that camp,” he told National Defense.

That includes the robot autonomously firing the weapon or, in other words, shooting without a
human in the decision loop, he said. SWORDS never had that feature, and the idea of armed
autonomous robots firing guns on the battlefield remains controversial. But Lynch was steadfast.

“I believe we can do automatic target recognition ... to allow that capability. Autonomously,” he

Lynch’s third priority is “convoy following.” The military needs to move massive quantities of
supplies overland, including fuel and water. Insurgents have taken advantage of this by planting
roadside bombs. These trucks can drive themselves, Lynch said.

“Why in the world does every cab have to be occupied by a human being?” he asked. “Why can’t
we just have the lead vehicle manned? I’ve seen that technology demonstrated many times over
the last 20 years, but it’s still not fielded.”

The top three requirements on his list are all IED related, Lynch noted. Studies show that this
will be the insurgent’s weapon of choice for the foreseeable future.
Indeed, recent casualty reports from Afghanistan have borne this out. The Taliban has taken up
this tactic, and roadside bomb attacks have increased as troops have surged into the provinces
this summer.

Furthermore, Lynch said, studies have also shown that the Army will be deploying 10 brigades
in the next 10 years. Investing in these counter-IED technologies and fielding them as soon as
possible will save lives, he said.

The final requirement is for a “robotic wingman” — an unmanned vehicle that will mirror the
movements of others with a certain degree of autonomy.

It’s wrong to assume that major combat operations will never return and future battles will all
mimic the Iraq insurgency, he said. Combat vehicles such as tanks can operate with autonomy
alongside manned vehicles.

“If you have four tanks in a platoon, and each has four kids in a tank, is that really necessary?”

Braden said when Lynch’s troops deploy next year he will receive some of the technologies on his
wish list.

“Will they be as good as he wants? Will they meet his metrics? Never.” Lynch will continue to
ask program managers to push the technology forward, Braden said.

Convoy-following and persistent stare, “I think we can do. Route clearance is going to be a really
tough technical challenge,” he added.

The wing-man concept? “That’s probably closer to doable. We need to push testing to show
people what we got,” Braden said.

“The resourcing will be a challenge because as good as we get, as quick as we get, we still have to
marry the requirement to the resource. But I am optimistic we can give him some great tools to
take forward into an operational assessment.”

Lynch, despite his stated weariness of technology demonstrations, set up a “robotics rodeo” at
Fort Hood the first week of September. He wanted vendors to come to the base and show their
wares as long as they fit into the four “life-saving” categories.

Key to the rodeo’s success would be the presence of decision makers who could see for
themselves that these technologies were mature and ready to go into battle.

“If we’re not fielding, we’re failing,” said Lynch, repeating an acquisition community mantra.

Braden agreed. “We need people like him pushing. We need the passionate three-stars to stand
behind how important this is. We need to push the things he’s asking for.”


No comments:

hit counter script