DARPA's Cybernetic Binoculars Tap Soldiers' Brains To Spot Threats

The U.S. Army and DARPA have concluded field tests on next-generation binocular replacements that read human brain signals and have a 91% threat detection success rate. They might just help you control your car with your thoughts too (seriously).

Binoculars on the battlefield are fine, as long as soldiers know what they're looking at. But when a target's not so clear or, say, a shopkeeper with a broom could easily be mistaken for an insurgent with an RPG, the eyes--even the conscious, rational mind--might not be the best tool for threat-spotting and quick reaction.
So a new system from military think tank DARPA is instead going straight to soldiers' brainwaves to spot real threats--from far away, or amid a crowded landscape.

The concept might sound familiar to science fiction readers: Augmenting human soldiers with brainwave-reading computers. The Cognitive Technology Threat Warning System (CT2WS) is a threat detection system for troops in the field that simultaneously scans warfighters' brainwaves while a camera surveys the area. The binocular replacement system detects a specific kind of brainwave (the P300, which is involved in stimulus evaluation and categorization), combines that info with a camera feed, and processes it all through an algorithm in near-real time to feed back an almost-instant threat assessment. (Think: every cyborg POV shot in every Terminator movie ever made.) Sounds pretty out there, but testing indicates 91% of enemy targets were identified in the field, compared with the 47% spotted by U.S. warfighters in action today who aren't using the new system.

The CT2WS project started in 2008, with the goal of developing next-generation portable visual threat detection devices for use in warzones. The University of California San Diego's bioengineering department and several California biotech and hardware firms partnered with DARPA to develop the brain-scanning enemy detection device.

As currently developed, CT2WS consists of three parts. There is an electroencephalogram (EEG) headset (below) worn by the user which records electrical activity in the brain and sends a ping to an outside computer system when the subconscious evaluates a visual threat.

Additionally, there is a separate 120 megapixel electro-optical video camera with a 120-degree field of view (below).

Lastly, both the camera and EEG unit are connected to a computer system that uses proprietary algorithms to identify potential targets and cue images for review. The software behind CT2WS can be run on a laptop as well, according to DARPA.

HRL Laboratories is a Malibu-based R&D house jointly owned by Boeing and General Motors which worked on CT2WS. One of HRL's specialties is developing cognitive-neural algorithms that allow computers to interpret human thoughts. According to HRL, the end result is far superior to conventional enemy-spotting technologies like binoculars. “CT2WS automatically scans a field of view more than ten times as wide as that is available using standard army binoculars. This is coupled with digital techniques that provide far higher resolution and greater effective visual distance than today's binoculars,” HRL's Deepak Khosla tells Fast Company.

In testing for desert, tropical, and open terrain, CT2WS was able to identify 91% of targets successfully. DARPA is also considering combining the system with a commercial radar--during field tests, the combination of CT2WS and a commercial system, the Cerberus Scout surveillance system, was able to identify 100% of the targets encountered.

The EEG sensor component of CT2WS was developed by San Diego's Quasar. Quasar used special wireless EEG sensors for the project that don't require the use of conductive gels and which don't cause skin abrasion. The lightweight EEG sensors and accompanying headset are small enough to wear under a bike helmet, according to Quasar's Walid Soussou. The CT2WS headset is also designed for easy cleaning, and meets the blunt and ballistic impact safety requirements for a military helmet.

DARPA, for their part, is playing up the fact that human and machine can complement each other on the battlefield. Project literature claims that “humans are inherently adept at detecting the unusual,” while algorithms are successful at detecting commonplace phenomena that are potential indicators of threats or targets--such as birds in flight or tree branches swaying. When the camera and sensor were tested, sensor and cognitive algorithms returned 810 false alarms per hour. However, once a testee began wearing an EEG cap and feeding in results, false alarms dropped to only five per hour.

Development of CT2WS is currently being transitioned from DARPA to the U.S. Army Night Vision and Electronic Sensors Directorate. According to HRL, the military is interested in CT2WS for situational awareness in reconnaissance, force protection surveillance, and standard infantry tactical fighting. The transfer of CT2WS technology to the U.S. Army indicates that the brain-wave-reading binoculars have progressed past testing and into the sweet spot of Pentagon bureaucracy.

Of course, CT2WS also has civilian applications: According to Khosla, HRL (which, again, is partly owned by General Motors) believes that the EEG decoding and cognitive algorithms used by CT2WS can also be used for controlling buttons inside cars or breaking in sudden emergencies--all using, well, human thought.