FMS Navigation and Position Updating
FMS navigation is the way a flight management system builds a route from its database and the crew's entries, computes the aircraft's position by blending inertial, satellite and radio sources, estimates the accuracy of that position, and generates lateral, vertical and time guidance along the route.
A flight management system navigates in three steps. It builds the route, from its database and the crew's entries; it works out where the aircraft is, by blending inertial, satellite and radio positions; and it guides the aircraft along the route laterally, vertically and, if asked, to a time. Each step has its own ways of going wrong, and most of the crew's FMS monitoring is aimed at catching them: a route that is not the one cleared, a position that has drifted, a vertical path built on wrong winds.
The principles are those of area navigation, and the accuracy the aircraft must achieve is set by performance-based navigation. This article covers how an airline FMS applies them, with the Boeing 737 and the Airbus A320 as examples.
Route building and discontinuities
The route is assembled from the navigation database: a company route, or origin, destination and airways, then the departure, arrival and approach selected by name from the departure and arrival pages. An identifier that is not stored produces NOT IN DATA BASE; the point can then still be entered by latitude and longitude or as a place/bearing/distance. Because the database encodes each leg's path terminator, fly-over points and speed and altitude constraints, the AIM requires approach procedures to be retrieved by name, never typed in as waypoints.
A route discontinuity, called a flight plan discontinuity on Airbus, is a gap in the plan where the FMS has no defined path from one point to the next. It typically follows a leg ending in manual termination, such as "fly heading 090, expect vectors", or appears where an arrival and an approach do not join. Reaching one ends lateral guidance: on the 737 LNAV disconnects, and on the A320 NAV reverts to heading or track, and a managed descent reverts to vertical speed with it.
A discontinuity is a question the FMS asks the crew, not a fault. It should be removed only once the intended routing is known, not by habit: closing it links the points either side into one continuous route, which may not be the path ATC expects. When ATC vectors the aircraft off a procedure, the AIM advises against deleting waypoints from the active legs page too early, so that the aircraft can rejoin the procedure.
Direct-to and pilot-defined waypoints
Direct to (DIR TO) replaces the legs before a chosen waypoint with a single leg from present position straight to it, much like the direct to fix (DF) leg type of RNAV procedure design. On the A320 it is a DIR TO revision made on the MCDU; on the 737 it is made on the LEGS page, where lateral route data is modified, and executed like any other change. Direct clearances are limited near approaches: under the AIM, an RNAV aircraft may be cleared direct to an initial or intermediate fix with an intercept angle of no more than 90°, and to a fix between the intermediate fix and the final approach fix with no more than 30°. ATC does not clear aircraft direct to a waypoint that begins, or lies within, a radius-to-fix leg.
A pilot-defined waypoint is a point the crew creates. The usual forms are:
- Place/bearing/distance (PBD): a stored point, a bearing from it and a distance. TRN250.0/76 is 76 NM from TRN on the 250° bearing, and the FMS names it TRN01 if it is the first created from TRN.
- Latitude and longitude, entered in the exact format the FMS expects, including leading zeros.
- Along-track waypoints, placed a set distance before or after a route point.
The A320 FMGC stores up to 20 pilot-defined waypoints; the 737 FMC keeps crew-defined items in supplemental and temporary databases. Their use on published procedures is restricted. The AIM forbids changing a database waypoint from fly-by to fly-over or the reverse, and allows no user-defined waypoints on published RNAV or RNP procedures except to change an altitude or speed constraint as ATC instructs, or to insert an along-track waypoint to comply with an ATC crossing restriction. Manual entry by latitude and longitude or place/bearing is not permitted for approaches.
Position updating sources
FMS position updating, or FMC position updating in Boeing terms, starts from inertial data. An unaided inertial system drifts, EASA texts say, by about 1 to 2 NM per hour; the FMC removes most of this by blending the inertial position with external fixes, weighting each by its expected error in a Kalman filter. With GPS input, EASA training texts quote a radial error rate below 0.05 NM per hour as not uncommon.
The 737 FMC uses the sources in this order of priority:
- GPS;
- two or more DME stations (a DME/DME fix);
- one VOR with a collocated DME;
- one localiser and collocated DME;
- one localiser.
With no external updating, the FMC uses the IRS position with a correction learned while updating was available. On the ground, radio updating is inhibited; the 737 uses GPS there, or the IRS position if GPS is unavailable. With GPS updating switched off, pushing TO/GA for take-off updates the position to the take-off runway threshold, so an intersection take-off must be entered on the take-off reference page. For RNAV 1 departures flown with DME/DME/IRU rather than GPS, the AIM requires the position to be confirmed within 1,000 ft at the start of the take-off roll, by an automatic or manual runway update.
On the A320 the two GPS receivers, either separate GPS sensor units (GPSSUs) or part of the multi-mode receivers, send their data to the ADIRUs, which compute a hybrid GPS-inertial position used by the FMGCs. The crew sees GPS PRIMARY when that position is in use, and GPS PRIMARY LOST when it is not.
Wrong positions still happen, from inertial drift without updating, a wrong runway entry or GNSS interference. The defence against a map shift is to cross-check the FMS position with raw data from the ILS, VOR, DME and ADF (see GNSS).
Navaid autotuning
For radio position updating the FMS must tune the right beacons itself. In normal operation each A320 FMGC tunes its own navigation receivers; if both FMGCs fail, the crew tunes navaids through the radio management panels. On the 737 the primary FMC allocates navaid tuning and updating between the two FMCs.
Autotuning depends on the FMS position being right. The AIM warns that an aircraft spoofed off its intended path may find that autotuning no longer selects the nearby navaid, one of several effects of GNSS interference on FMS-equipped aircraft. Raw data from a navaid used for a cross-check should be identified by its Morse code.

ANP and EPU
Every FMS estimates the quality of its own position. Boeing calls this actual navigation performance (ANP): the radius of a circle around the FMC position within which the aircraft lies with 95% probability, shown on the RTE LEGS and position pages. Embraer calls it estimated position uncertainty (EPU) and Airbus estimated position error. The lower the value, the more confident the FMS.
The estimate is compared with the required navigation performance (RNP) for the phase of flight. On the 737, when ANP exceeds RNP for longer than the time to alert, UNABLE REQD NAV PERF-RNP appears on the CDU and the map, and the amber FMC lights come on. Airbus shows NAV ACCUR DOWNGRAD when its estimate exceeds the required accuracy; on an A320 approach, no separate navigation accuracy check is needed while GPS PRIMARY is displayed.
The 737 applies the idea vertically too. Vertical ANP (VANP) estimates the maximum altitude error with 99.7% probability, computed from barometric altitude, so it is valid only with the altimeter setting from the ATIS or the approach clearance; the default vertical RNP in the oceanic, en-route and terminal phases is 400 ft.
VNAV path and speed modes
Vertical navigation (VNAV) flies the vertical profile computed by the FMS. In the climb it uses climb thrust, respects the speed and altitude constraints of the departure, and then climbs at the economy speed; if the climb speed profile cannot meet an altitude restriction, UNABLE NEXT ALTITUDE appears. On the 737 VNAV arms on the ground only with a valid flight plan and performance data entered and executed, and both flight director switches on.
A descent can be flown two ways:
- In a VNAV PATH descent the FMS uses pitch to hold a computed vertical path, like a glideslope in three dimensions, with thrust at idle. If the groundspeed becomes too low to hold the path, the 737 autothrottle can leave idle and hold speed.
- In a VNAV SPD descent the FMS holds a target speed with pitch at idle thrust, like a level change descent, and the path is not flown.
The two combine. On the 737 a descent started early, before the top of descent, is flown at 1,000 ft/min until the idle path is intercepted; speed intervention during a late descent changes VNAV to speed guidance until the path is regained. If the descent target speed is within about 6 kt of VMO or MMO at the top of descent, VNAV may revert to LVL CHG to prevent an overspeed. Without an end of descent point in the flight plan there is no VNAV path descent. The A320's managed descent, DES, is available only with NAV engaged; its top of descent is computed backwards from a point 1,000 ft above the aerodrome on final at the approach speed, VAPP, respecting the constraints.
Idle and geometric descent paths
The FMS places the top of descent so that a descent at idle thrust from cruise meets the approach altitude and speed without level-offs or added thrust, the most fuel-efficient profile. That is the idle path. Where altitude constraints do not fit it, the FMS joins them with a geometric path, a fixed angle between constraints that may need more than idle thrust.
The thrust mode shows which is being flown. On the A320 the autothrust is in THR IDLE on an idle segment and in SPEED on a geometric segment; on the 737, speed intervention on an idle segment keeps VNAV PATH and hands the speed to the autothrottle. Because the idle path assumes the entered winds, a tailwind moves the top of descent further out, and missing wind entries give time and fuel errors (see climb and descent planning).
Required time of arrival
A four-dimensional FMS adds time. The required time of arrival (RTA) function lets the crew specify when the aircraft must cross a waypoint; on the 737, VNAV then controls cruise speed to meet it, and the FMC computes a recommended take-off time, the speeds needed and progress information. If the time cannot be met, the scratchpad shows RTA UNACHIEVABLE. EASA training texts cite time-managed arrivals and noise-abatement procedures as uses.
Exam tip: 2D RNAV is lateral only, 3D adds the vertical profile, and 4D adds time through the RTA function. A modern airline FMS is a 4D system.
Frequently asked questions
How does an FMS update its position?
It blends several sources. The inertial reference systems give a smooth position that drifts slowly; GNSS and radio fixes do not drift but are noisier and can be lost. A Kalman filter weights each source by its expected error. The Boeing 737 FMC uses, in order of priority, GPS, two or more DMEs, a VOR with collocated DME, a localiser with collocated DME, and a localiser. Without external updating it relies on the inertial position.
What is actual navigation performance (ANP)?
ANP is the flight management computer's estimate of the quality of its own position, given as the radius of a circle around the FMC position within which the aircraft lies with 95 per cent probability. The smaller the ANP, the more confident the FMC is. It is compared with the required navigation performance, and an alert is given when ANP exceeds RNP. Other manufacturers call it estimated position uncertainty (EPU) or estimated position error (EPE).
What is a route discontinuity in an FMS?
A discontinuity is a break in the flight plan where the FMS has no defined path from one waypoint to the next, for example after a leg ending in radar vectors. When the aircraft reaches it, lateral navigation stops: on the Boeing 737 LNAV disconnects, and on the A320 NAV reverts to heading or track mode, taking a managed descent with it. The crew closes the gap only once the intended routing is clear.
What is the difference between VNAV PATH and VNAV SPD?
In a VNAV path descent the FMS holds a computed vertical path with pitch, much like a three-dimensional glideslope, with thrust normally at idle. In a speed descent it holds a target speed with pitch at idle thrust, like a level change descent, and the path is not held. On the 737, speed intervention during a late path descent makes VNAV change to speed guidance until the path is intercepted.
What is required time of arrival (RTA) in an FMS?
RTA is the time function of a four-dimensional FMS. The crew enters a waypoint and the time at which the aircraft must cross it, and VNAV adjusts the cruise speed to meet it. The Boeing 737 FMC also computes a recommended take-off time and progress information. If the time cannot be met, the scratchpad shows RTA UNACHIEVABLE.
Test yourself on FMS Navigation and Position Updating
The v1prep banks cover this topic in Instrumentation (022), with a worked explanation for every answer. EASA ATPL, PPL, IR and CPL, the FAA written tests and A320/B737 type ratings.
Start practising →Sources and further reading
- FAA Aeronautical Information Manual, Chapter 1 Section 2 (1-2-1 RNAV and FMS, 1-2-4 GNSS interference)
- FAA Aeronautical Information Manual, Chapter 1 Section 1 (1-1-17, GPS database and waypoint rules)
- FAA Aeronautical Information Manual, Chapter 5 Section 4 (5-4-6, Approach Clearance, direct clearances to approach fixes)
- FAA Aeronautical Information Manual, Chapter 5 Section 5 (5-5-16, RNAV and RNP Operations)
- EASA, Explanatory Note to ED Decision 2019/011/R (CS-ACNS Issue 2, PBN)
- FAA Instrument Flying Handbook (FAA-H-8083-15B), Chapter 9, Navigation Systems
Library articles are written for study and exam preparation. They do not replace your aircraft's approved documentation, your operator's procedures or the regulations themselves.