Soldiers gain combat edge with smart helmets
A DARPA program is funding efforts to refine wearable sensor networks designed to track enemy fire
A number of advances to shooter-location systems that U.S. ground forces use to track the source of enemy fire could produce even more sophisticated tools, including one that has the ability to network sound-gathering nodes for even greater accuracy.
Vanderbilt University researchers are constructing a prototype that takes the concept further. Building on an concept known as PinPtr, it uses smart nodes perched on soldiers’ helmets — or elsewhere on their body — to form ad hoc, self-organizing networks with other nodes to pinpoint where fire is coming from and then build a 3-D image that it relays to soldiers' handheld devices.
The system can do more than identify shots fired in direct line-of-sight. It can detect multiple shooters at a time and identify the caliber and type of weapons fired.
Each node uses four separate microphones instead of just one like most other location systems use, said Ákos Lédeczi, senior research scientist at Vanderbilt’s Institute for Software Integrated Systems (ISIS), which developed the shooter-location system.
If just three of those microphones can detect the acoustic wave created by a muzzle blast, each node's microprocessor can calculate the shooter’s location, though not with great precision. More precise feedback emerges when several nodes communicate with one another, constantly swapping information about shooter and location.
The more nodes that are part of the network, the more accurate the whole ensemble is, Lédeczi said. When two or more nodes pick up the sound of a rifle firing, it can locate the shooter’s position to better than one degree of accuracy as far as 300 yards away in less than a second.
The Vanderbilt network also uses a patented filtering technique to distinguish primary shooter sounds from those that bounce between buildings, which is necessary for urban environments.
“Our system can handle echo in an urban environment, which is something others systems can’t do yet,” Lédeczi said.
The network was developed under the Defense Advanced Research Projects Agency’s Advanced Soldier Sensor Information Systems and Technology program. The program's goal is to improve battlefield awareness through things such as soldier-worn sensors.
The system uses sensor technology invented at University of California at Berkeley and manufactured by Crossbow Technology. Vanderbilt’s contribution came through development of the algorithms used to detect and filter acoustic data and a circuit board that carries a node’s microprocessor and other electronics.
An entire ISIS node can be produced for about $1,000, substantially less than what other systems cost. It weighs only slightly more than the four AA batteries used to power it.
However, the system still needs more work, Lédeczi said.
Because the helmet is an exposed area during battle, which puts the nodes at greater risk of damage, the Army has been experimenting with sensors that are worn on various other parts of the body. However, that can hamper the accuracy of the system because it doesn’t provide for a 360-degree field of sound capture for the sensors.
Also, tests carried out by the National Institute of Standards and Technology to determine the accuracy of the system used stationary nodes and no live people. Researchers need to perform more tests to prove it can work as well in a mobile environment, Lédeczi said.