A new system makes "lost at sea" an obsolete concept
It's easy to think of GPS as a ubiquitous resource reliably blanketing the Earth. But actually, there's a region making up two-thirds of the globe where many travelers routinely have trouble picking up any GPS signal at all, sometimes for up to weeks at a time, while traveling thousands of kilometers.
That region is made up of the world's oceans, and the travelers who need alternative ways of navigation are those who cruise in submarines below the water's surface, where GPS signals can't penetrate. But it's not just submariners who risk getting lost at sea without a backup to GPS. Increasingly, the U.S. Armed Forces are concerned that adversaries might purposefully disable, jam, or corrupt GPS signals during hostilities. For the military's ships and other sea-faring vessels and platforms with nothing but open water in sight, the sudden inability to know exactly where they are could place their missions — and their safety — at risk.
But that problem is on the way to being solved, thanks to a Leidos precision navigation system that can instantly kick in when GPS goes down. Making use of everything from ocean-depth information to tiny changes in deep-sea magnetic and even gravitational fields, our Assured Data Engine for Positioning and Timing (ADEPT) can provide any craft that operates above or below surface with near-GPS-precision position data. "Even if GPS is denied for days or weeks at sea, you can still stay on mission," says David Stark, a senior navigation Subject Matter Expert on the Leidos ADEPT development team.
Of course, sailors have, for millennia, been able to navigate out at sea without GPS by using the sun and stars, as well as by keeping track of their speed and direction. Those old-fashioned techniques are still useful and have even become fairly accurate with modern equipment such as high-powered optics, and inertial measurement units (IMUs) that gauge acceleration.
ADEPT can use those techniques, too. But weather can obscure the sky, and IMUs tend to drift into inaccuracy over hours, especially the low-power, inexpensive ones used with small, light craft. That's why ADEPT brings in other, more sophisticated sensor data to close the gap left by a missing GPS signal. One of those data streams, for example, can come from acoustic sensors that bounce sound waves off the ocean floor to measure depth, currents, and the ship's velocity relative to the bottom. ADEPT can take that information and compare it to maps of the ocean floor — available for much of the worlds' oceans — in order to determine where the craft is.
Getting the best location fix possible
To get an even more precise reading and resolve any uncertainty in the depth data, ADEPT can add in data from magnetic- and gravitational-field sensors that underwater vehicles and surface vessels often carry. Such sensors look for how those fields change over different locations in the ocean. “There are all kinds of anomalies in the seafloor that have been mapped,” notes Doug Stranghoener, a Program Manager on the ADEPT team specializing in maritime navigation. "The system can use any of them to get a more accurate fix."
But getting good sensor data is only half the job. At the heart of ADEPT is a software "sensor fusion engine" that pulls all the available data together, using machine learning, to determine how accurate and trustworthy the different types of data are, and how to blend it all together to get the most precise and reliable fix possible. Results are updated continuously throughout the mission.
As new and improved types of sensors are placed on ships and submarines, as well as on platforms such as the underwater unmanned vehicles that the Navy is deploying, ADEPT can work them into the data mix with little or no modification. "The more sensors you have in your toolkit, the better the fix you can get in different circumstances," says Stark. "We can rapidly integrate any combination of sensor data that's available." The software can even detect errors in sensors or on published maps, he adds, and adjust them on the fly to yield accurate locations.
The picture gets even better if ADEPT is being used across a group of vessels, such as a surface fleet or Unmanned Underwater Vehicle (UUV) swarm, thanks to ADEPT's ability to enlist collaborative positioning techniques. The software combines the various sensor data streams from the different members of the group, integrates it with ranging data that gives the position of the vessels relative to one another and comes up with fixes for the entire group that can be more accurate than any one craft could develop on its own. That technique is typically good for ships separated by up to about 20 miles, notes Stranghoener — and if an aircraft is overhead to relay signals between ships that are beyond the line of sight, it can work up to sixty miles or more.
Now the ADEPT team is looking at ways to sharpen the software's ability to guide ships through harbors that can be crowded with marine traffic and other hazards, says Stark. But he concedes that for now, it's hard to beat the judgment of a sharp-eyed sailor when it comes to that particular challenge.
There are still some situations where the old seagoing ways are best.