Wired for War: The Future of Military Robots

Peter W. Singer

The coming war-bots by land

The robotic systems just rolling out or already in prototype stage are far more capable, intelligent, and autonomous than ones now in Iraq and Afghanistan. They range in size from tiny eight-pound robots to the world’s current biggest robot, a 700-ton robotic dump truck capable of hauling 240 tons of earth at a time, which also served as the model for the character Long Haul in the Transformers series. But even they are just the start. As one robotics executive put it at a demonstration of new military prototypes, “The robots you are seeing here today I like to think of as the Model T. These are not what you are going to see when they are actually deployed in the field. We are seeing the very first stages of this technology.”

The next wave of new robots to be deployed on land will mostly be “new and improved” versions of existing platforms. For example, iRobot’s original Packbot just had a digital camera that sent back views of what the robot was seeing, making it essentially a mobile pair of binoculars. Now, most Packbots perform EOD roles with the ensuing addition of fairly simple effector arms and grippers.

But as new add-ons are developed, the same robot will be able to take on a wider set of battlefield roles. For example, the company has already tested out a Packbot armed with a good-old fashioned shotgun, because it is “so versatile.” The robot can now fire a variety of ammunition, including non-lethal rubber bullets, rounds that can blow down a door, and even powerful “elephant killer” bullets.

Another version is called the REDOWL (Robotic Enhanced Detection Outpost with Lasers), which uses lasers and sound detection equipment to find any sniper who dares to shoot at the robot or accompanying troops, and then instantly targets them with an infrared laser beam. ‘”You’ll actually see the sniper before the smoke disappears from the shot,” said retired Admiral Joe Dyer, who leads the military programming at iRobot. He adds that in tests, it’s been 94% accurate and is smart enough that ‘”It can tell the difference between a 9 millimeter pistol and an AK-47 or an M-16.”

Foster Miller has similar plans to upgrade its current generation of ground robots. For example, the first version of the armed SWORDS still needs the remote human operator to be situated within a mile or two, which can keep the human in the danger zone. Vice President Robert Quinn describes how the company plans to vastly extend the range of communications to get ground robot operators completely off the battlefield. “It is not an insurmountable problem. It is nothing that money and time can’t solve.”

The SWORDS itself is being replaced by a new version named after the Roman god of war. The MAARS (Modular Advanced Armed Robotic System) carries a more powerful machine gun, and 40mm grenade launchers. For non-lethal settings, it packs a green laser “dazzler,” tear gas, and a loudspeaker.

As these systems evolve, we will also soon see entirely new unmanned combat vehicles hit the battlefield. One such prototype was the Gladiator. Described as the “world’s first multipurpose combat robot,” it came out of partnership between the Marine Corps and Carnegie Mellon University. About the size of a golf-cart, the vehicle is controlled by a soldier wielding a Playstation video game controller, but software plug-ins will allow it to be upgraded to semi-autonomous and then fully autonomous modes. Fully loaded, it costs $400,000 and carries a machine gun with 600 rounds of ammunition, anti-armor rockets, and non-lethal weapons. “It is just fucking nasty,” raves Noah Shachtman.

Not all ground robots will take on combat roles. For instance, medics have long been one of the most dangerous jobs on the battlefield. A former Army Special Forces officer explains how this is generating a pull for robotic solutions, “If you can avoid unnecessary situations where you expose them [medics] to fire and you end up with two dead guys, then we have a responsibility to the American people to avoid that.”

An early entry into the “med-bot” field is yet another improved version of the Packbot, known as the Bloodhound. Whenever a soldier is hurt, an alert will go out and the robot will find the wounded soldier on its own. Once there, the robot’s human controller, who might be located anywhere in the world, will check out the soldier via the video link and treat them using the robot’s on-board medical payload, which will include a stethoscope (likely very cold, with no one to breathe on it), liquid bandages, and even automatic syringes to dispense morphine or medicines.

Of course, robots will have a hard time replicating the compassion of a real life medic. As one Special Forces soldier tells, “The last thing I want to see if I’m about to die is a robot coming for me. I want to see a human.” On the other hand, that robot may be able to go where humans could not, so their lack of a bedside manner may be viewed as an acceptable tradeoff.

Wet and wild, robot style

A broad new set of robots are also being introduced for war at sea, where the main differentiation is whether they are designed to operate on the surface, like a boat, or under water, like a submarine.

Robots of the first type, unmanned boats, are called USVs (Unmanned Surface Vessels). They actually have a great deal in common with land robots as they both primarily operate in a 2-dimensional world. Many of the simplest USVs merely entail taking sensors and a remote control unit and plugging them into a boat.

However, many think the sea is actually a far more difficult environment for robots than land. “Everything’s working against you,” says Robert Wernli of the Space and Naval Warfare Systems Center (SPAWAR) in San Diego. Waves and currents can pull a boat off course. Visibility is lower, and sometimes communications more difficult. Plus, robots can get seasick; the constant motion and corrosive effects of salt water cause breakdowns much more rapidly than on land.

So far, the prototype USVs tend to be smaller boats rather than large warships. One example is a 36 foot robotic motorboat called the “Spartan Scout,” which the Navy has spent some $30 million developing. Guided by a GPS navigation system, the boat can be on its own for up to 48 hours, and speed up to 50 miles per hour. Filled with sensors (including day and night video cameras), Spartan Scout is designed to carry out surveillance, patrol harbors, and inspect any suspicious ships that might be tying to pull another USS Cole type attack by sneaking up on a Navy vessel. If it finds something fishy, the robot boat is also packing a .50 caliber machine gun. Spartan Scout got its first real world use in the Iraq war in 2003, inspecting small civilian boats in the Persian Gulf, without risking sailors’ lives. The boat also mounts a loudspeaker and microphone, so an Arab linguist back on the “mothership” would interrogate suspicious boats that the Spartan Scout had stopped. As one report put it, “The civilian sailors were somewhat taken aback when they were interrogated by this Arab speaking boat that had no one aboard.”

The other type of sea-bots are UUVs (Unmanned Underwater Vehicles). These are designed for underwater roles such as searching for mines, the cause of most naval combat losses over the last two decades. Many UUVs are biologically-inspired, like the “Robo-lobster,” which operates in the choppy waters close to shore. But others are converted torpedoes, like the Remus, which was used to clear sea-mines in Iraq, or even mini-submarines, which are launched from manned submarines to hunt down the enemy.

The sea will also prove to be a new platform for robots to fly from. The Navy plans to equip many of its ships with the MQ-8 Fire Scout, a sister version of the robotic helicopter used in the US Army’s Future Combat Systems plan. Able to take-off from and land autonomously from any warship with a small deck, the Fire Scout can fly more than six hours. It packs thermal imagers, radar, high-powered video cameras, and a laser designator that can target for the mothership’s weapons, or fire its own rockets. With a range of over 200 miles, the robotic chopper can take the ship captain’s eyes further than ever before, including even inland.

The most novel of the drones at sea may be the Cormorant, DARPA’s design for a submarine-launched flying drone. Operating a plane off a submarine may sound new, but it actually dates back to World War II; indeed, the very first air attack on the mainland U.S. was in 1942, when a submarine-launched Japanese plane bombed Brookings, Oregon. What is new about the Cormorant is not only would it be unmanned, but also able to be both launched and recovered while the submarine stays hidden underwater. Having wings like a sea-gull, the drone would be squeezed into a missile launch tube. Whenever the sub commander wants to scout about or launch a surprise air attack, the drone would be fired from the tubes, float to the surface, and then launch into the air using converted rocket boosters. The drone would then fly back to a rendezvous location on its return. It then lands in the water, sinks back down, and the submarine scoops the robot plane back inside.

Top (unmanned) guns

As with ground robots, the next wave of robot planes, also known as “unmanned aerial vehicles” or “systems” (UAV or UAS), will be a mix of upgraded current systems, converted manned vehicles, and brand new designs. For example, the Predator drone today does surveillance and also some ground attack missions. New versions are being reconfigured for electronic warfare, submarine hunting, and even air to air combat. Thomas Cassidy, a former Navy fighter pilot (so respected that he even had a cameo in Top Gun) and now CEO of Predator’s manufacturer General Atomics, declares, “I want to see a Predator coming back here with MiG kills painted on its side; and that will happen soon.”

The next generation of the Predator is the even more menacing-sounding Reaper, an Air Force drone about four times bigger and nine times more powerful. Among its improvements is a Microsoft Windows software package that has “automatic man-made object detection” and “coherent change detection.” Not only can the plane come close to flying itself, but its sensors can recognize and categorize humans and human-made objects. It can even make sense of changes it is watching, such as being able to interpret and retrace footprints or even lawnmower tracks. As of 2009, 28 Reapers were in service, with many deployed to Afghanistan, “standing alert somewhere in case a certain high-priority target pops his head out of his cave.” There are also plans for them to patrol above London during the 2012 Olympics.

In turn, just as the Reaper hit the battlefield in 2009, Cassidy’s company turned out a prototype of its successor, called the Avenger. Powered by a jet engine, it can fly twice as fast than the turboprop-powered Reaper, and more than three times as quick as the Predator, while carrying over 3,000 pounds of weapons. It is also specially designed to be stealthy, with radar absorbing materials, a swept wing, and internal bomb bay. Of note, the prototype also came equipped with a tailhook, potentially allowing it to land on aircraft carriers at sea, another indicator that it is being planned for operations not merely against insurgents in the Middle East, but in potential contingencies against a certain large Asian landpower that shall go nameless.

The Avenger is one of the many contenders for the role of a UCAS (Unmanned Combat Aerial Systems). This type of drone is specially designed to replace the ultimate of pilot roles, the fighter jock. Other key UCAS prototypes include the Boeing X-45, which was described by one author as “flat as a pancake, with jagged 34-foot batwings, no tail and a triangular, bulbous nose” that make it look like “a set piece from the television program Battlestar Galactica.” X-45 also has a cousin, the Northrop Grumman X-47, which is roughly the same size, but designed to land on aircraft carriers.

These drones were designed to be especially stealthy for the most dangerous roles, such as targeting enemy air defenses. Already, the prototypes have been able to launch precision guided missiles, have two planes flown by one human operator, have the controls “passed off” between different remote human operators 900 miles away from each other, and in one war-game autonomously detected unexpected threats (missiles that “popped-up” seemingly out of nowhere), engaged and destroyed them, and then did battle damage assessment on their own. Follow-ons to the X-45 are now believed to be funded under the Pentagon’s “Black budget,” while the X-47 will start its trials of flying off US Navy aircraft carriers in 2011. The program is also credited with inspiring similar unmanned fighter jet prototypes in China, France, Israel, Italy, Russia, South Africa, and the United Kingdom.

The pattern with unmanned planes in the early 21st century seems to be mirroring what happened with manned planes in the early 20th century. There was initial skepticism and opposition to them in general, followed by using them just for observation and spotter roles. Soon, however, a new line was crossed, and they began to be used for ad-hoc attack roles, often out of the pilots’ frustration at just watching the enemy.

The stories from World War I were of early observation plane pilots taking grenades and homemade bombs up with them in their biplanes to drop on the enemy and the same happened with the early jury-rigged arming of the Predator in the air and the Marcbots on the ground. Perhaps the most telling parallel story from the Iraq war was when an enlisted soldier flying an unarmed Raven drone spotted an enemy insurgent planting an IED. He tried to show his commanding officer the danger, but the officer couldn’t pick out the image of the insurgent on the view screen. So, the operator kept circling the drone closer and closer to the insurgent. Still, the officer couldn’t see the Iraqi. Finally, the soldier just got frustrated and flew the drone directly into the insurgent’s chest.

Of course, just as World War I pilots couldn’t just watch each other going about their business of bombing their side on the ground, and so started taking pot-shots at each other which very quickly led to fighter planes specially designed for the task, so too is the next step of advancement unmanned drones that are specially designed to take on other robotic planes. In 2006, DARPA budgeted $11 million for the “Peregrine UAV Killer.” Like the peregrine falcon that hunts other birds, it is designed to loiter over an area, stealthily gliding about, until it sees an enemy UAV and then quickly dive down and blast it. Drone vs. drone is the step that follows in air warfare.

As new prototypes of unmanned planes hit the battlefield, though, they will begin to look less current planes. For instance, one trend will be for the size extremes to be pushed in two directions. Not having pilots that need to be replaced every 10 hours or so will also allow unmanned planes to become far bigger than any created so far. For example, Boeing is at work on a glider powered by solar energy and liquid hydrogen that could stay aloft for 7 to 10 days. It has a wingspan almost the length of a football field.

We are also seeing the return of blimps to warfare. Raytheon has just received a contract to build 12 unmanned aerostats for the US Army, which carry radars that can detect targets on the ground as far as 300 miles away. In turn, Lockheed Martin has been given $150 million to design and build a robotic “High Altitude Airship,” 25 times larger than the Goodyear blimp. Such huge, long-endurance, unmanned blimps open up a whole new range of roles not normally possible for planes. For example, airships could literally be “parked” in the air, as high as 100,000 feet up, for weeks, months, or years, serving as a communications relay, spy satellite, hub for a ballistic missile defense system, floating gas station, or even airstrip for other planes and drones.

At the other end of the scale, a host of robotics are being miniaturized. For instance, US Air Force lab researchers have built a rocket engine that fits on the tip of a pencil. Some believe that micro systems could eventually go down to the nano-scale, that is to the molecular level. While the idea has been bandied about in such fiction as Michael Crichton’s novel Prey, many think it could come to fruition in the coming decades. Boston College researchers have already built a chemically powered nanomotor that is just 78 atoms in size, while those at a university in the Netherlands have made a solar-powered engine just 58 atoms in size.

Tiny engines allow tiny machines. And, tiny machines may mean teeny, tiny robots, or “nanobots.” A major advancement in these happened in 2007, when David Leigh, a professor at the University of Edinburgh revealed that he had built a “nanomachine,” whose parts consisted of single molecules. When asked to describe to a normal person the significance of his discovery, Leigh said it would be difficult to predict. “It is a bit like when stone-age man made his wheel, asking him to predict the motorway,” he said. Leigh would make one venture, however. “…Things that seem like a Harry Potter film now are going to be a reality.”

Astro-bots go to war

If space is to become a new potential zone of conflict, its unique nature demands that unmanned systems play a key, and perhaps near-exclusive, role. Not only do weapons in space need to stay up there a long time, but the major challenge of fighting in space is first getting things into space. It costs roughly $9,100 a pound to launch anything into space with the Space Shuttle. So, if a system is to be manned, the human and each and every pound of water, food, and oxygen tanks to keep them alive is expensive to send. Likewise, manned systems in space are incredibly vulnerable (one bullet or laser hole and there goes all the air).

Instead, the U.S has already started work on a number of unmanned systems for potential use in space. One example is the X-37, an orbital test vehicle about a quarter of the size of the Space Shuttle, which flew its first test flight in 2006. The military’s strong interest in it is perhaps best illustrated by the fact that, while the program was originally run by NASA, its development was then transferred over to DARPA.

Another program is the X-41 Common Aero Vehicle, also known as the Falcon program. Planned for testing in 2010, it is a cross between an intercontinental ballistic missile and the Space Shuttle. It is designed to travel at the border between space and the atmosphere, around 100,000 feet, but, unlike a missile, the Falcon will be able to come back after a mission if it finds no targets. As John Pike of the Global Security organization comments, the aim is to give the U.S. the ability to “to crush someone anywhere in the world on 30 minutes notice, with no need for a nearby airbase.”

This weaponization of space, unmanned or not, is certainly controversial. Former U.S. Senate majority leader, Tom Daschle, described the plans to investigate the possibility of war in space, then being pushed by Sec. of Defense Donald Rumsfeld, as “the single dumbest thing I have heard so far from this administration…It would be a disaster for us to put weapons in space of any kind under any circumstances. It only invites other countries to do the same thing.” Lt. Col. Bruce Deblois of the U.S. Air Force published a detailed study that concurred with Daschle. The report argued that while being the first to deploy weapons into space might seem advantageous, it would only open up the floodgates for others to the same.

These fears do appear to be playing out. In 2007, after a test of their own anti-satellite missile, senior colonel Dr. Yao Yunzhu of the Chinese Army’s Academy of Military Science, issued a not thinly veiled warning. If the U.S. thought it was going to be “a space superpower, it is not going to be alone…It will have company.” This debate will likely rage on for years, if not decades, or at least until the Vulcans arrive to resolve it for us.

But what is interesting is that governments are not the only ones looking at space as a new unmanned battleground. In 2007, the Tamil Tiger group of Sri Lanka became the first, but likely not last, terrorist group to takes its operations into space, hijacking the signal from an Intelsat satellite and using the commercial satellite to beams its own messages back to Earth. And, just as private companies like Blackwater have reentered the conflict game on this planet, we should not be surprised if privatized conflict also rises one day in space, especially with the growth of private space businesses. Examples include Richard Branson’s “Virgin Galactic” space-line plan or Google’s $30 million prize to the first private team able to land a robot on the moon (one of the competitors is actually also the team behind some of the Pentagon’s energy beam weapons programs). Robo-One, a robot competition event held in Japan every year, may provide a taste of what’s to come. The organizers have announced plans for a new division in 2010: robot combat in space. A small satellite, carrying humanoid robots will be blasted into the heavens. “Once safely in orbit, the satellite will release its robotic passengers, who will proceed to fight each other in the vacuum of space.” Space, the final frontier, for humankind, and robot war.


Get daily updates from Brookings