An F-35 releases a joint direct attack munition, which contains inertial measurement units for backup guidance. Some researchers are looking at improved IMU technology to replace or supplement existing GPS. (Air Force)
WASHINGTON — GPS navigation: Can’t live without it, can’t trust it.
That’s the problem facing military planners as they try to sort through the increasing reliance of equipment and troops on positioning data from satellite networks at a time when the vulnerabilities of GPS are becoming more apparent.
This month, a group of hackers managed to break into ship-tracking software to spell out the word “pwned,” hacker speak for a successful conquest. And over the summer, a group of students at the University of Texas figured out how to send ships off course using fake GPS signals.
That doesn’t mean the US military isn’t working to strengthen the encryption on GPS equipment, which was never designed with security as a top priority. But Pentagon research and development chief Al Shaffer listed GPS security as one of his top concerns.
“It’s becoming easier and easier with modern electronics to do things like jam GPS signals,” he said. “Our military is reliant on GPS both for precision navigation and for time, and most of our weapon systems need very precise time.”
Shaffer mentioned work on inertial measurement units (IMUs) as a possible solution. It’s not a new technology, having been the subject of major research before the invention of GPS, but it’s getting a close look.
The idea is that a unit uses three gyroscopes and three accelerometers to gauge direction and changes in velocity. If given an initial starting point, the mechanisms can be used to provide relative movement data and tell where something has gone.
In 2009, Andrei Shkel started looking into the improved use of IMUs for the Defense Advanced Research Projects Agency (DARPA), where he worked as a program manager on the microtechnology for the positioning, navigation and timing program.
He started focusing on one of the most demanding problems: IMUs for missiles. The challenge is that the units have to be both small and cost effective, given that they’re set to be destroyed.
IMUs were already in use, but were capable of maintaining accurate location data for only about 20 seconds, meant as a backup measure for the temporary loss of a GPS signal.
Shkel wanted to see if a unit could maintain accuracy for the duration of a missile flight. His research found that 70 percent of missiles flew for three minutes or less, with 98 percent flying for 20 minutes or less.
“The conclusion was that we have very precise inertial instruments that would be able to solve this problem probably 100 percent, but at a very high cost,” he said. “You can’t put a million-dollar inertial measurement unit on a missile that’s under a million dollars.”
Not only were the units expensive, but also quite large.
“The state-of-the-art, low-cost devices were the size of an apple,” Shkel said. “We said, ‘Can we build devices that will solve the problem, but will have the size of an apple seed?’ ”
They largely succeeded. Using semiconductor technology, Shkel and his team came up with a system that could provide three minutes of accurate data that was a fraction of the size of a penny with an anticipated cost of less than $1,000 per unit to manufacture.
That would cover the duration of missile flights for 70 percent of missions, but Shkel and his team wanted to see if the technology might be applied elsewhere.
So they came up with a system where a variant of the IMU could be inserted into a boot, taking advantage of the natural start and stop of footsteps to keep the system calibrated.
“When you walk, when you touch the ground, the sensors go through this zero velocity point,” Shkel said. “If you use this phenomenon to reset the sensor, we showed that you could remain accurate using this approach for up to four hours.”
But because of the limited durations, these systems still worked largely as backups, meant to collaborate with GPS. That can be useful, as GPS and an IMU can be used to constantly compare data, which would help show when the GPS is tampered with.
“You can basically analyze your GPS receiver and confirm if there is a match with what your inertial navigation gives you and what your GPS navigation gives you, and draw comparisons about spoofing and jamming,” Shkel said.
The Atomic Approach
But to extend the capability of the inertial systems, Shkel is looking at pairing the data with that from another technique that uses Larmor frequencies of atoms to detect motion change.
The idea is that one can detect changes to an atom’s magnetic field using lasers when the atom moves. The technique has the advantage of remaining accurate over much larger durations and longer distances, so long as the measurements aren’t heavily affected by outside magnetic fields. To that effect, the systems, called chip-scale combinatorial atomic navigators, are shielded from outside fields.
The downside to the atomic measurements is that they don’t react very well to sudden movements. That’s not a problem if the IMUs are used in tandem.
“Combining this fast response and this ability to measure a large dynamic range, and the long-term stability coming from the atomic devices, will basically put together the best of both worlds and extend the duration of missions,” Shkel said.
Jointly, the technology should be able to allow nearly any missile to hit its target, and combined with algorithmic data, allow a soldier an extended period of use without any connection to a GPS signal.
“It’s currently four hours, but it could go for days,” Shkel said.
His only major concern? Funding and interest. Shkel left DARPA earlier this year and continues his work at the University of California, Irvine, but he sees the history of inertial systems as a warning.
“It’s happened historically where people realized that inertial navigation was a big deal; back in the ’70s and ’60s, they put a lot of emphasis on it, but then GPS came along and people said, ‘Well, we don’t need inertial sensors so much,’” he said. “Fast forward 30 years, and we start questioning whether we can rely on GPS, and the next 20 years, we’ll be dealing with jamming and spoofing in military applications.”
And the technology, while advancing, isn’t at the point of production yet.
“DARPA jumps in and funds the program, but now efforts are entering this valley of death,” Shkel said.
What may help the science is that the commercial side of navigation also is getting attention.
“The industry is well aware of ‘spoofing,’ ” Michael Toscano, CEO of the Association for Unmanned Vehicle Systems International, wrote in testimony before Congress in 2012.
“Spoofing has implications for any technology that depends on GPS for guidance and timing, whether it is manned or unmanned aircraft, your cellphone or your car. In fact, commercial airliners are relying more and more heavily on GPS signals to locate the runways at airports and, with the advent of the next generation air traffic control systems, all aircraft — manned and unmanned — will rely on GPS for navigation.”