GPS and Its Limitations

Trilateration: the core idea behind satellite positioning

GPS, formally the Global Positioning System, is one specific example of a broader category of technology called GNSS, or Global Navigation Satellite System. Every GNSS satellite continuously broadcasts a signal containing its own precise position and an extremely accurate timestamp. A receiver on the ground, or on a drone, picks up signals from multiple satellites and measures how long each signal took to arrive. Since radio signals travel at the speed of light, a known travel time translates directly into a known distance to that satellite. With a distance measurement from a single satellite, the receiver only knows it is somewhere on a sphere centered on that satellite; with distance measurements from four or more satellites at once, the overlapping spheres narrow down to a single point, giving the receiver its own three-dimensional position. This process is called trilateration. A fourth satellite is needed in addition to the three that would mathematically be sufficient for a pure 3D position, specifically to correct for the receiver's own less-precise internal clock, which is not accurate enough on its own to measure signal travel time correctly.

Multi-constellation GNSS: more than just "GPS"

In everyday conversation, "GPS" is often used loosely to mean any satellite positioning system, but GPS specifically refers to the United States-operated constellation. Three other major constellations operate the same way and are commonly used together with GPS by modern receivers: GLONASS (operated by Russia), Galileo (operated by the European Union), and BeiDou (operated by China). A drone's positioning receiver that can use multiple constellations simultaneously has access to far more satellites in the sky at any given moment than a GPS-only receiver, which generally means faster position locks, more consistent satellite geometry, and improved accuracy, especially in marginal conditions such as partial sky obstruction.

Source of error one: multipath

Multipath error occurs when a satellite signal does not travel directly to the receiver, but instead bounces off a nearby surface, such as a building facade, a metal structure, or even the ground, before arriving. Because that reflected signal traveled a longer path than a direct signal would have, it produces an inaccurate distance measurement, which throws off the resulting position calculation. Multipath error is most severe in "urban canyon" environments, dense areas with tall buildings close together, but it also affects operations near large industrial structures, bridges, and other reflective surfaces relevant to many commercial inspection jobs.

Source of error two: ionospheric and atmospheric delay

GNSS signals travel from satellites in orbit down through the Earth's atmosphere, including the ionosphere, a layer of the upper atmosphere containing electrically charged particles. That layer slows the signal slightly and unpredictably, depending on solar activity and time of day, introducing a small timing error that translates into position error. Consumer-grade receivers generally cannot fully correct for this effect on their own, which is one reason baseline GPS accuracy for a typical sUAS is measured in meters, not centimeters, without additional correction technology.

Source of error three: satellite geometry and DOP

Even with a large number of visible satellites, the geometric arrangement of those satellites in the sky matters. If the satellites being used are clustered close together in one part of the sky, the resulting position calculation is far less precise than if the same number of satellites were spread widely across the sky. This concept is measured as Dilution of Precision, or DOP: a low DOP value indicates good, well-spread satellite geometry and a more reliable position fix, while a high DOP value indicates poor geometry and reduced positional confidence, even with an otherwise adequate satellite count.

RTK and PPK: how mapping-grade accuracy is achieved

Standard consumer GNSS positioning is typically accurate to somewhere in the range of a few meters, which is entirely adequate for general position hold and return-to-home, but not precise enough for applications like survey-grade mapping, where centimeter-level accuracy matters. Real-Time Kinematic (RTK) positioning solves this by using a second, stationary receiver, called a base station, placed at a precisely known location. The base station continuously compares its own known position to what GNSS alone reports, calculates a correction, and transmits that correction in real time to the drone's receiver (the "rover"), which applies it immediately during flight to achieve centimeter-level accuracy live. Post-Processed Kinematic (PPK) uses the same underlying concept, but instead of applying corrections in real time, it logs raw GNSS data on both the base station and the aircraft during the flight, then combines and corrects that data afterward in software. PPK does not depend on a continuous real-time radio link between the base and the aircraft during flight, which makes it more resilient in situations where maintaining that link reliably would be difficult, at the cost of not having corrected positioning available live during the flight itself.

• RTK: Real-time correction, requires a continuous live link between base station and aircraft during flight.
• PPK: Corrected after the flight in software, does not require a continuous live link during flight.

Why sUAS depend on GPS for position hold and return-to-home

Most multirotor sUAS use GPS positioning to hold a stable hover automatically, continuously making small corrections to counteract wind drift by comparing the aircraft's current GPS position against where it is supposed to be. This is what makes a modern drone feel like it "locks in place" in the air rather than drifting the way a purely manual aircraft would in any wind. GPS is also the foundation of return-to-home (RTH): at takeoff, the aircraft logs its GPS coordinates as the home point, and if RTH is triggered, either manually by the pilot or automatically after a lost control link, the aircraft uses its current GPS position and the stored home point to navigate itself back autonomously.

The deeper technical picture behind losing GPS signal in flight

Module 11 covers the operational, in-the-moment procedure for a loss-of-GPS-signal emergency; this lesson explains the technical reality underneath that scenario. GPS signal loss or severe degradation in flight generally comes from the same error sources already covered here reaching a critical level at once: too few visible satellites due to obstruction (flying close beside a tall building, under a bridge, or near dense tree cover), a temporary spike in ionospheric disturbance, severe multipath in a dense urban environment, or, less commonly, deliberate interference or jamming. When a modern sUAS loses reliable GPS lock, it typically does not simply fall out of the sky; instead, many aircraft automatically switch from GPS-assisted position-hold mode into a mode often called "attitude" or "ATTI" mode, in which the flight controller still stabilizes the aircraft's orientation but no longer has a positional reference to hold it steady against wind or execute an accurate return-to-home. In this mode, the aircraft will drift with the wind unless the remote pilot takes over manual control, which is exactly why understanding this technology matters operationally: a pilot who has never practiced flying without GPS assistance is far more likely to struggle at the exact moment a real loss-of-GPS event occurs.

Scenario: inspecting a building facade between two tall high-rises

A remote pilot is hired to inspect the facade of a mid-rise building situated between two much taller high-rises downtown. During the flight, the aircraft's GPS accuracy indicator degrades noticeably, and the pilot notices the position hold feels less locked-in than usual, drifting slightly before correcting. This is a textbook multipath and satellite-geometry problem: the surrounding high-rises block a meaningful portion of the sky, reducing the number of directly visible satellites and increasing DOP, while reflective glass and metal surfaces on the buildings introduce multipath error on the signals that remain. The pilot, understanding this is a known limitation of the environment rather than an aircraft malfunction, flies more conservatively, gives the aircraft more separation from the reflective surfaces where practical, and stays ready to switch to manual attitude-mode flying if the position hold degrades further, rather than assuming GPS assistance can be fully trusted in this specific environment.