GPS-free systems to spur highly advanced sensors, fusion

We’ve come to expect the U.S.-based global positioning system (GPS) to be available and ubiquitous for the countless military, commercial, and consumer applications dependent on it. Its diverse uses represent a huge leap from its original military-centric objective for determining an object’s precise location (positioning), chart its path to a destination (navigation), and manage its movement along that path (guidance)—usually summarized as PNG.

Applications that were not even conceived, let alone doable, are now enabled by tiny GPS ICs and systems that provide amazingly accurate and precise results—you can make your own list here.

If you want some insight into the people who made GPS happen despite severe technical and bureaucratic obstacles, check out Pinpoint: How GPS Is Changing Technology, Culture, and Our Minds by Greg Milner. Though somewhat dated now in its discussion of social implications, this fascinating book from 2016 tells the story of GPS from its conceptual origins as a bomb guidance system to its presence in almost everything we do.

Despite the sense that GPS is everywhere, the reality is that it was never the situation. Underwater, tunnels, indoor sites, and similar RF-blocked locations simply can’t receive enough of the relatively weak satellite signals to provide a viable result.

Now, we’re seeing many more situations where GPS signals are also being “denied” due to deliberate interference or spoofed via false signals by players with various motives. Some of the consequences are modest (lost dogs can’t be found), but others have more serious implications.

One possible solution is to increase the power of the transmitted signals, but that’s technically difficult and won’t happen for years even if and when it does—and doing so will still not help in many of these cases.

Alternatives to GPS

There’s a significant amount of research and product development toward devising ways to provide PNG using non-GPS, non-RF techniques driven by sensors for which jamming or signal access is not an issue. All of them require a considerable amount of computation to make sense of the sensed signals and transform data into results; none of them provide the performance of a GPS-based system—at least not yet. Much of the R&D work is being done by startups and innovators, in addition to traditional sensor vendors.

Among the non-GPS possibilities are:

• Inertial sensing

This is not new, of course, and has been used for decades, beginning with gyroscopes and accelerometers. Both sensors are now reduced to small, low-power MEMS devices that are orders of magnitude smaller, lighter, and lower-power than their electromechanical predecessors of just a few decades ago and even compared to the laser and fiber-optic versions that leverage the Sagnac effect and interferometry. Still, their accuracy is not as good as a high-end GPS system, but it’s improving.

For example, ANELLO Photonics has developed a silicon photonics optical gyroscope—dubbed SiPhOG—that uses an on-chip waveguide manufacturing process, integrated with a patented silicon photonic integrated circuit (Figure 1). Together, they claim these offer fiber-optic gyro performance with a standard silicon manufacturing process.

Figure 1 This silicon photonics optical gyroscope uses an on-chip waveguide manufacturing process that is integrated with a patented silicon photonic IC. Source: ANELLO Photonics

• Magnetic sensors

The Earth’s magnetic field is pervasive, ubiquitous, and unjammable. It’s also uneven, with highly localized variations due to differences in the Earth’s outer-crust and under-crust layers as well as deeper causes (literally) from flows of conducting material within the Earth (Figure 2).

Figure 2 This geomagnetic map of part of the Northern hemisphere is a starting point for more detailed, higher-resolution images and variations, and changes that must be captured for effective magnetic navigation. Source: Geomag

Using supersensitive quantum-based magnetic sensors based on optically pumped, cesium-based, split-beam scalar magnetometers, which have an absolute accuracy between one and three nanoteslas, it’s possible to read that field with high precision. The Earth’s core field has values ranging from 25 to 65 microtesla (that’s 0.25 to 0.65 gauss) at the surface while magnetic anomaly field of interest typically varies by just hundreds of nanotesla.

The readings are then matched to pre-existing maps of Earth’s field. This scheme has the disadvantage of not being very accurate compared to GPS, partially because the Earth’s magnetic field is not static and matching maps need constant updating.

Despite these challenges, companies such as SandboxAQ have developed a navigation technology (AQNav) that leverages proprietary large quantitative models (LQMs) and powerful quantum sensors to make use of the Earth’s crustal magnetic field. By combining high-sensitivity magnetometers with AI algorithms to identify unique magnetic patterns and locate position in real time, it’s possible to determine position in that field. The sensing is entirely passive, so users remain undetected.

• Visual matching

This uses a simple concept of matching what a camera sees to the verified landmarks on a map. Visual terrain-following has been used for decades in cruise missiles which follow a precise terrain-image pattern. Orders-of-magnitude improvements in imaging quality and the associated algorithms needed to process and match the observed image to the map now make this technology even more precise.

One vendor pursuing this approach is Vermeer Corp. Their system uses between one and four electro-optical/infrared camera feeds simultaneously to map real-time video to a locally stored 2.5D or 3D map database to generate an accurate location signal.

• Celestial navigation

This classic approach to navigation now uses modern, automated versions of the transit, celestial charts and precise clocks, aided by computerized calculations. This is a case of “back to the future” but in a new form and implementation.

• E-LORAN

LOng-RAnge Navigation was a hyperbolic radio navigation system developed by the United States during World War II. The third iteration, LORAN-C, was initiated in the late 1960s, but the stations and system were decommissioned in the 1990s due to the availability and performance of GPS.

It uses the differences in timing of received signals from multiple high-power transmitters in the 100-kHz band (yes, that’s kilohertz) to developed positioning information.

Enhanced LORAN is a standard which builds on the now obsolete LORAN system by putting more information into the modulation of the carrier as well as adding a data channel. Like LORAN, E-LORAN offers some benefits such as near-impossibility of jamming and spoofing, but it also requires many high-power transmitters and many of these need to be in inhospitable or remote locations which are difficult to support (Figure 3).

Figure 3 Like its predecessor LORAN, the enhanced LORAN system will require an extensive physical infrastructure located around the world. Source: UrsaNav

While E-LORAN proponents are eternally hopeful, the project has had difficulty getting traction and support due to technical challenges (primarily at the transmitter side), very high up-front infrastructure costs, and best-case accuracy of about 50 to 100 meters (although there are proposed ways to improve that number).

The realities of dealing with a GPS-unavailable world

Many of these alternatives are being enabled by advances in quantum-based sensors. Some may even require supercooled arrangements with all the obvious downsides of that requirement. Each of them offers the virtue of not being jammable or denied.

At the same time, none offers the amazing accuracy and simplicity of GPS for the user. No single technology offers anything close to GPS. A viable alternative, even with reduced accuracy, will require advances in sensors and gigabytes of support data such as maps. Any GPS alternative will also require tight fusion and merging of unrelated sensor technologies and outputs, huge datasets, and extensive use of AI and machine learning to create useful results.

It will be fascinating to see which one of these, if any, takes a dominant role in non-GPS settings, or will it be a balanced fusion? Perhaps some unexpected physical phenomenon will come from behind, as has happened so often in the past. As they say, “predictions are very hard to make, especially about the future.”

Related Content

• When your sensors mislead you
• Sensors Without Wires, But Not “Wireless”
• Navigating without GPS requires advanced sensors, intensive analog
• Sophisticated Sensors, Extreme Conditioning, Advanced Algorithms Yield Amazing Geolocation Results

The post GPS-free systems to spur highly advanced sensors, fusion appeared first on EDN.