For most of human history, the ability to pinpoint your location would have been nothing short of magic. These days, thanks to Global Navigation Satellite System (GNSS) technology, it barely raises an eyebrow. Yet, as ubiquitous as GNSS has become in commercial, civil, and military applications, it’s still not a perfect solution. The biggest issue: unreliability.
As long as receivers have clear line of sight to GNSS satellites, positioning works great. And most of the time, they do — especially in the air, on sea, or on highways. Unfortunately, in many locations, buildings and terrain can interfere with satellite signals, leading to significant errors and even loss of fix. And positioning systems typically have no way to predict when and where these significant errors will occur.
Technologists have long viewed this as an inherent, if unfortunate feature of GNSS. After all, we can’t change the laws of physics to make satellite signals immune from reflection or interference. With new 3D-mapping techniques, however, we can do something just as useful: understand exactly where and when GNSS can be trusted, and where it can’t. Now, industry leaders are using these techniques to radically improve positioning for current and emerging use cases. And they’re rewriting the rules of what’s possible with GNSS.
Navigating GNSS Challenges
The strength of GNSS technology lays in its use of constellations of satellites. Receivers only need clear line of sight to a few of them to accurately establish position, and most of the time they can get it—even when most satellite signals are blocked. The problem: GNSS satellites never stop moving, orbiting the Earth at about one degree per minute.
If you’re calculating position in an urban region with tall buildings then, you could usually trust GNSS signals, but there’s been no way to know when you shouldn’t. Reflections off buildings in particular can throw off positioning by tens of meters (Figure 1). It’s not like positioning systems can just avoid bad spots either, since they change as satellites move, literally second by second.
So far, this issue hasn’t been a deal breaker for most commercial GNSS applications. As we trend towards more ambitious use cases though—such as autonomous vehicles or drone-based delivery—reliable positioning data becomes absolutely critical.
Changing the Game with 3D Mapping
Today, industry leaders are applying new techniques to solve this problem. By combining GNSS data with high-resolution 3D maps, they can identify exactly where GNSS signals will be available and can be trusted, and where they can’t, second by second. And they can offer these insights (typically as a cloud service) to any commercial customer or government agency that could benefit from more reliable positioning.
Here’s how the technology works: Companies aiming to provide GNSS reliability services start with a fine-grained 3D map of a given area. For aviation and automotive use cases, organizations are using mapping solutions that deliver native resolutions down to 30 centimeters, with the ability to derive even higher-definition resolutions. They also use techniques like multispectral imaging to differentiate between natural and man-made materials, penetrate through smoke, and more.
Once these companies have a high-fidelity map, they create an overlay of observable points across that area, such as breaking the region up into one-meter squares. Then, from each of those points, they determine which satellites will be clearly visible at a given time. There are multiple ways to make this determination. The most popular include:
- Shadow matching: This technique, often applied to smartphones, creates a 3D map-aided (3DMA) grid showing predicted GNSS signals for a given area. It then compares those predictions against measured signals in real time to identify and correct for errors.
- Ray tracing: In other venues—especially aviation, where more granular details on elevation and azimuth are needed—leading solutions use ray tracing from each observable point up to each satellite in a constellation. They do this repeatedly, building a data set that shows which satellites are line-of-sight (and therefore, reliable) second by second, across every meter of a 3D map (Figure 2).
Next-Generation Positioning in Action
GNSS reliability solutions apply different techniques for different use cases. In addition to how a system calculates line of sight, the biggest factor dictating a solution’s approach is how quick the calculation needs to be and how many points. There are two main options:
- Device Computation for a single point: For applications with ample local computing resources, where some amount of latency in calculating positioning is acceptable, and the device only wants to know a single (or few) points a device centric solution is appropriate. This approach is used in most smartphone-based scenarios, where applications have a powerful endpoint device, typically a strong network connection, and do not require precision measured in milliseconds.
- Cloud based scaled solution: There are many scenarios, however—such as aerial drones and driverless vehicles—where the entire environment around the vehicle or route are needed quickly and the vehicle does not have computation dedicated to GNSS. In these cases, GNSS cloud services can offer real-time predictive insights. They perform all calculations for an area ahead of time in the cloud, forecasting some period into the future. Positioning engines can then understand in advance and real-time exactly where they can trust GNSS signals and where they can’t, and proactively mitigate issues. Or better yet, use he predictions to extract the best signals possible and improve performance.
Looking Ahead
Implemented effectively, early integrations have shown that these techniques can literally double the accuracy of GNSS systems. But the potential goes much farther across multiple aviation, automotive, and military scenarios:
- Aviation: 3D mapping techniques enable smarter flight planning, allowing organizations to know ahead of time where and when it’s safe to fly, and adjust flight times to optimize positioning reliability. Or, in large-scale drone applications like drone-based delivery, companies can choose to routes known to be reliable at specific times.
- Automotive: Automakers can use the predictions in real-time to improve the performance of the GNSS system and improve safety by knowing where and when GNSS is reliable as a part of all the sensors on the car.
- Military: Mission planners can improve risk analysis by understanding best- and worst-case scenarios for GNSS performance in a given area. They can also evaluate the integrity of GNSS signals in the field, including detecting when positioning data is being blocked or spoofed.
These are just a few of the possibilities. With new 3D mapping techniques, we can transform GNSS from “usually reliable” to highly accurate. And we can start tapping the potential of positioning for applications we’ve only begun to imagine.
Disclaimer: Views Expressed are Author's Own. Geospatial World May or May Not Endorse it
