Top 5 DARPA Robot and AI Projects of the Past Year

Robotics development and commercialization reached new highs in 2016, with notable events including Google DeepMind’s victory at board game Go, Midea Group’s acquisition of industrial automation leader KUKA, and national strategies such as “Make in India.” As with other technologies, the Defense Advanced Research Projects Agency is at the forefront of robotics innovation. DARPA robot funding could have wide-ranging effects.

Among other things, the U.S. Department of Defense’s research and development agency launched the Internet in 1969, the Global Positioning System in 1973, and some of the first digital maps in 1979. These three inventions make modern-day robotics possible.

“That’s what we’re looking for every day: breakthrough technologies that will give an advantage to our war fighters,” said Stephen H. Walker, acting DARPA director, at an event at Harvard University last week. “Part of what we do is to invest in technology, but we never forget who we’re working for. The war fighter is our customer, and what we do is about protecting the nation.”

DARPA hasn’t slowed down but in fact has been actively pursuing numerous robotics and AI projects in the past year. Here are five DARPA robot projects that start with military systems and could affect the entire industry.

The global military robotics market will grow to $21.11 billion by 2020, predicts Markets and Markets, and the U.S. defense budget is also expected to grow.

[note style=”success” show_icon=”false”]

Business Takeaways

DARPA appears to be positioning its innovations towards solving staff shortages and enhancing communication and operation between advanced technologies.
Multiple DARPA robot projects support the advancement of drones in air, on land, and in the sea.
As with AUVs and fighter pilots, innovations coming out of DARPA labs could challenge the status quo and cause friction between different military departments.

[/note]

5. ACTUV

The Anti-Submarine Warfare Continuous Trail Unmanned Vessel (ACTUV) project began in 2010 and is developing autonomous military vessels with the initial goal of tracking diesel-electric submarines from foreign nations. The latest product of ACTUV is a vessel called “Sea Hunter.”

An estimated 132 feet long, Sea Hunter can stay at sea for months at a time and uses international conventions to decide autonomously what to do in certain situations.

DARPA partnered with the Office of Naval Research to develop Sea Hunter, so this vessel may be a sign of the capabilities the U.S. Navy wants: fully-autonomous vessels operating in waters around the world.

4. ALIAS

A common theme in many DARPA robot research is reducing the need for humans to constantly be present in certain environments. One way to do this is through software or drones. But what about giving pilots extra robotic arms?

DARPA’s Aircrew Labor In-Cockpit Automation System (ALIAS) solves the problem of not having enough military pilots by replacing co-pilots with a robot arm from Universal Robots. Multiple cameras share data, and then the robot controls aircraft functions such as the throttle. Could this be an extension of autopilot technologies and eventually reach commercial aircraft?

Speaking of robotic arms, last July, DARPA gave Space Systems Loral a $20.7 million contract as part of its efforts toward robotic servicing of geosynchronous satellites, replacing some of the capabilities of the space shuttle. However, Orbital ATK Inc. claims to be working on a similar system.

3. TERN/UFP

Northrop Grumman’s Tactically Exploited Reconnaissance Node (TERN) drone is unlike any other military drone in existence because it can take off and land vertically like a helicopter and then shift into horizontal winged flight and perform a surveillance once it reaches a certain height (see image above.

TERN will be tested this year and is expected to provide every U.S. warship with Predator-like capabilities. General Electric supplies the engine, and the company has 3D-printed parts for other aircraft.

An equally impressive drone project is called “Upward Falling Payload.” The idea is to place drones in oceans around the world that will lie dormant until activated.

2. CODE

Collaborative Operations in Denied Environment, or CODE, is designed to solve one of the biggest problems when it comes to controlling robots in complex environments — decision making.

CODE would enable multiple drones to carry out entire missions on their own by assessing environments and situations in real time and then request permission for certain actions from a single human operator.

CODE is so aggressive that it seeks to allow unmanned aerial vehicles (UAVs) to even identify and engage targets on their own.

1. Eyeriss

The most audacious DARPA robot project of the past year is code-named “Eyeriss.” It is a chip that can be installed in smartphones.

However, unlike smartphone chips that have just four cores, Eyeriss has 168 cores, and it would be programmed with artificial intelligence to supposedly match the human brain. Through this chip, the Pentagon plans to enable robots and drones to learn in real time, without any human input.

The Defense Department has tried to reassure the public and nervous observers that its robots and AI will still be limited and accountable to humans.

“We’re very clear that the lethality decision should always be made by a human,” Walker said. “But we can develop technology for the human war fighter to make his decisions more quickly and effectively.”

[note style=”success” show_icon=”true”]

More on Military, Security Robotics:

[/note]

DARPA robot projects to watch in 2017

What might DARPA unveil this year? New initiatives would likely fall under two broad themes. They could be technology for replacing humans in harm’s way, or they could address the challenges of controlling robotics and advanced systems in complex environments.

For instance, the N-ZERO program intends to reduce power consumption of Internet of Things (IoT) devices so that they “wake up” in an environment when needed (see also UFP above). The agency is also funding research into self-healing materials.

Cost savings are another issue, although the Trump administration’s federal budget calls for a $20 billion increase in military spending.

The real question, however, how far the DoD is willing to go with certain projects. Take CODE and ACTUV, for example. CODE allows robots to execute entire missions on their own, including engaging targets, while ACTUV seeks to develop autonomous sea vessels. Combine these two, and you have a not only new naval vessel, but you could replace entire fleets.

This leads to an even bigger issue: Will the branches of the U.S. armed forces align with DARPA’s push toward automation or challenge the agency as it alters the makeup of the U.S. military?