Gulfstream 450/550 Sensor Logic Changed with ASC-912C

The EPIC system looks for the sensor with the best accuracy (i.e., the lowest Estimated Position of Uncertainty (EPU)). It’s very common to see an EPU below 0.1 nm with either Hybrid or GPS as the position data source. This could result in nuisance shuffling between them since the selection tolerance is very small. To ensure the FMS position did not bounce back and forth, Honeywell incorporated logic to prevent the sensor from shuffling due to the similar EPUs. The Hybrid IRS function utilizes filtered GPS inputs to independently calculate position data producing a tightly integrated GPS/IRS position. A Hybrid filter gain was also added that is equal to that of GPS-D but will have a higher priority for sensor selection than GPS.

Why it Works Well

Honeywell published an article explaining Inertial Reference Systems in detail several years ago that can be found on the Honeywell Pilot Gateway. Hybrid IRS is often described technically as a “tightly coupled” position sensor which provides extremely accurate position calculations. Better explained, Hybrid IRS uses components in the Inertial Reference Unit to receive GPS data from the GPS receiver.  “Why is this good” you ask?  The idea (and benefit) of the hybrid system is that it uses both position sources, IRS and GPS, to remove error from one another.   Consider IRS – it is extremely accurate in short periods of time yet accumulates drift error that compounds over several hours.  GPS position is accurate in most parts of the world but does have some errors introduced due to atmospheric conditions, satellite coverage and geometry, and interference. Hybrid blends the two, using sensor-specific algorithms to filter out error in each.

Cutting Through the Noise

Prior to Hybrid technology, the FMS received a position input on power-up that was extremely accurate, but that accuracy decreased as the IRS would drift at a rate of 1-2 miles per hour. Hybrid IRSs have algorithms that use multiple Kalman Filters to target specific inaccuracies, resulting in significantly refined position, velocity and attitude performance. Measurement data are sent through a series of Kalman filters that analyze the statistical probability of errors and continuously refine the hybrid position solution. The Hybrid solution provides for better overall position accuracy from the sensors and, in the case with later systems (referred to as “High Step II”), provide more filtering that produces a RAIM like function and also extends integrity protection levels. This allows the system to “coast” using the Hybrid Inertial Position in an environment, that for many reasons, may lose GPS. This makes RNP navigation, especially approaches using low RNP, more robust to protect against GPS outages that would otherwise lead to unexpected missed approaches.