We're so accustomed to using the global positioning System (or its European and Russian counterparts, the Galileo GNSS and GLONASS systems) for location and navigation that it's easy to forget or ignore the obvious: GPS does not have a 100% assurance of availability.
What can impede GPS? Three classes of things can:
- Physical barriers such as underground tunnels, bodies of water, or even ordinary buildings, for starters
- Unpredictable solar events such as those huge “bursts” of high-energy particles that cause electronic storms and upsets in our atmosphere
- Unintentional local interference or intentional jamming
Given these weaknesses — and they are very real, not just “the sky is falling” crisis mongering to get more funding — the military and its Defense Advanced Research Project Agency (DARPA) are supporting projects that may lead to very precise navigation without GPS. A recent article in the always readable and enjoyable Physics Today provided a fascinating overview of these efforts, the progress thus far, and the problems ahead. Early in the article, there was a basic statement that's always worth keeping in mind: “Three pieces of information are needed to precisely navigate between two known points: orientation, acceleration, and time. Once an initial position is provided, inertial systems determine where to go.”
These next-generation inertial navigations systems would provide GPS-like accuracy on the order of a meter or better, but they would be completely independent of any outside signal reception. That's why they are called “inertial,” after all. The goal is provide position, navigation, and timing (PNT) information via inertial measurement units (IMUs) that are simultaneously orders of magnitude smaller, lighter, and lower-power than today's best IMUs.
IMUs, of course, are not new. Using extremely sophisticated mechanical gyroscopes, they have been used for missile guidance, and ring-laser gyros are now standard on commercial airliners. They are helping look for distortions in the space-time continuum caused by gravity. And MEMS-based IC gyros are now available at low cost, using little power and offering very good performance –- but not good enough for the DARPA project goals.
To get a sense of what inertial guidance and IMUs are all about, check out these two books: Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance by Donald MacKenzie (a somewhat unusual subtitle but a great read) and Modern Inertial Technology: Navigation, Guidance, and Control by Anthony Lawrence (more technical and quantitative).
Regardless of which IMU developments help DARPA reach the goals, there is no doubt that PNT instrumentation has advanced a long way in just a few decades, when precision analog circuits centered on low-drift op-amps were key building blocks. Regardless of how the next stage of IMU performance is implemented, it still starts with sensors (analog transducers) and analog signal-conditioning circuitry, which must somehow extract meaningful signals from an electrically challenging environment plagued by low signal level and low SNR (to cite just two issues) and do so with precision and stability.
The outputs of these analog stages are then enhanced and further enhanced in the digital world of processors and algorithms, including additional compensation and calibration, but these objectives can be reached only if the analog front end (sensor plus conditioner) has the required performance. The old maxim of “garbage in, garbage out” still carries a lot of truth in this situation, as it does in nearly all test and measurement scenarios.
How do you think non-GPS, next-generation IMUs will perform, compared to today's IMUs and GPS-based systems? Will there be a fundamental new-technology “breakthrough” (admittedly, a hard-to-assess benchmark, except in retrospect) that radically advances the state of the art? Or will it be the result of a series of modest but still significant incremental improvements that combine and leverage one another to yield substantial improvements along one or more dimensions of weight, performance, or power consumption?