For most airline passengers, the final moments before landing look deceptively simple. The airplane follows the runway centerline, descends along a precise glide path, and appears to make every correction without pilot input. Yet even on a conventional Category I ILS approach, there is a specific point at which that automation typically stops being the primary means of controlling the aircraft. At about 200 feet (61 meters) above the runway, the pilots must decide whether they have enough visual reference to continue. If they do, one pilot can disconnect the autopilot and land manually. If they do not, the aircraft must execute a missed approach rather than continue toward a runway the crew cannot see. The Federal Aviation Administration (FAA) defines Category I ILS operations around a 200-foot decision height, making that altitude a fundamental boundary between instrument guidance and the visual portion of a normal landing.

That does not mean an airliner suddenly becomes dependent on human control at 200 feet, nor does it mean pilots always disconnect the autopilot at precisely that altitude. Modern aircraft can fly straight onto the runway with autopilot when the airport, airplane, crew, and weather conditions support an authorized autoland. But it requires additional aircraft systems, specialized airport infrastructure, operational approval, and advanced crew procedures. Understanding why airlines normally hand-fly the final portion of an approach, while retaining the ability to let automation land the airplane in very poor visibility, reveals how pilots balance technology with direct control.

Why 200 Feet Is Such An Important Boundary

Credit: Shutterstock

The 200-foot figure comes from the definition of a standard Category I ILS approach. An ILS provides lateral guidance through the localizer and vertical guidance through the glideslope, allowing the aircraft’s flight guidance system to maintain a highly precise path toward the runway. The approach remains an instrument procedure until the published decision altitude or decision height is reached. At that point, the crew must determine whether the required visual references are available.

The decision height is not simply a convenient altitude chosen by pilots, but rather the point at which continuing without the necessary visual information is no longer permitted under the applicable approach authorization. If the runway environment cannot be identified, the correct response is a go-around or missed approach. When the required visual references are present, the landing can continue. Depending on the airline’s standard operating procedures and the aircraft’s configuration, the pilot flying may disconnect the autopilot and control the aircraft manually. The transition is normally planned well before the decision height, so the crew knows what it intends to do rather than making a last-second choice.

The title’s 200-foot figure should not be interpreted as a universal autopilot-disconnect altitude. Some pilots disconnect earlier, while others may keep the system engaged longer. An aircraft equipped and authorized for a coupled approach can continue using its automatic flight-control system below 200 feet. The FAA’s published ILS categories make the underlying principle clear. Category I has a 200-foot decision height, Category II lowers that figure to 100 feet, while Category III operations can permit approaches with no decision height, depending on the specific authorization and category.

The Pilot Is Still Working During An Automated Approach

Credit: Stantec.com

A common misconception is that an engaged autopilot means the pilots have little left to do. The opposite is true during an instrument approach. Automation controls the aircraft’s flight path, but the crew must continuously confirm that the system is doing what it is supposed to do. The pilots monitor the localizer and glideslope indications, airspeed, altitude, vertical trend, aircraft configuration, and flight guidance modes. They also cross-check the approach against the published procedure and the aircraft’s operating limits. If the aircraft begins deviating from the expected path, the crew must recognize the discrepancy and decide whether to correct it, disconnect the automation, or discontinue the approach.

This monitoring becomes particularly important because the ILS itself is not an infallible source of guidance. The FAA warns that disturbances to ILS signals can affect coupled or autoland operations during the final stages of an approach. Aircraft conducting those operations are expected to monitor the automatic flight-control system closely and remain prepared to intervene. The pilot’s role therefore changes rather than disappears. Instead of continuously moving the controls to maintain the flight path, the crew supervises the system, making those control inputs electronically.

When the pilot disconnects the autopilot in good conditions, the aircraft is not being transferred from an unattended machine to an unprepared human. The crew has been actively monitoring the approach throughout its final stages. The manual landing is the next phase of a process that has already involved constant assessment. For passengers, this handover can be almost invisible. A small click or change in the cockpit’s flight guidance indications may be the only evidence that the pilot has taken direct control.

Why Autoland Requires Much More Than An Autopilot

Credit: Wikimedia Commons

If an aircraft can follow an ILS accurately, it might seem logical to let it land automatically every time. The obstacle is that an autoland system must remain reliable much closer to the ground, where there is less time to recover from a malfunction. CAT II and CAT III operations therefore require additional approval and equipment. The FAA maintains specific lists of US and foreign runways approved for Category II and Category III operations, and operators must satisfy authorization requirements before conducting these approaches.

The ground infrastructure is part of that equation. These advanced approaches depend on carefully maintained ILS facilities, runway lighting, protected areas around the navigation equipment, and runway visual range information. The aircraft also needs appropriate redundancy. Autoland-capable airliners use multiple information sources and flight-control capability so a single fault does not necessarily eliminate the ability to complete the landing automatically. The precise architecture varies by aircraft type, but the principle is consistent: the system must detect failures and determine whether it can safely continue. A fail-passive system is designed so that a failure does not result in an unsafe flight-control command, but the autopilot may disconnect when a significant disagreement occurs.

The FAA’s AC 120-118 establishes the framework for operators seeking authorization for CAT I, CAT II, and CAT III landing operations. It covers the approval process and associated operational requirements rather than simply declaring that an aircraft with an autopilot can perform an autoland.

What Happens When The Autopilot Does Land The Aircraft

Credit: Airbus

During an authorized autoland, the 200-foot boundary no longer represents the end of automatic flight control. Instead, the aircraft continues descending through the lower portions of the approach while the crew monitors its performance. On aircraft such as the Boeing 737, the system’s approach modes progressively establish the aircraft on the ILS, prepare for the flare, and maintain the runway centerline after touchdown. The exact mode names and activation points depend on the aircraft type and airline procedures, but the concept is consistent.

Near touchdown, the aircraft must reduce its descent rate and establish the appropriate landing attitude. After touchdown, the automatic system can continue providing directional control during rollout when the aircraft and approach are authorized for that capability. This is where autoland demonstrates its real purpose. It is not primarily intended to make an ordinary clear-weather landing easier. Instead, its major advantage appears when pilots cannot reliably acquire the runway early enough to complete the landing manually.

The FAA’s approach categories illustrate how dramatically the operating environment changes as visibility deteriorates. Category II allows a 100-foot decision height under specified conditions, while Category IIIA can operate with a decision height below 100 feet or without a decision height, subject to its RVR requirement. Category IIIB can operate with RVR as low as 150 feet under the applicable authorization. At those visibility levels, pilots may have little practical opportunity to look outside, identify the runway, and safely take over.

Why Pilots Still Prefer To Hand-Fly In Good Conditions

Credit: Shutterstock

Autoland does not make it the preferred choice for every approach. When visibility is good, manual landing can be operationally simpler and gives pilots direct control over the final phase of flight. Automatic landing systems also have limitations, and those limits can be more restrictive than the aircraft’s broader manual-landing capabilities. Crosswind and runway contamination are two examples of autoland limitations.

There is also a human-factors consideration. Monitoring an automatic landing is itself a demanding task. In favorable conditions, manually flying the final portion may provide a more straightforward division of responsibilities. Automation is available when it provides a meaningful safety advantage, particularly when visibility becomes too poor for conventional visual landing. When conditions are comfortably above those limits, pilots can use their manual flying skills rather than use a complex automatic capability simply because it exists.

An Evolving Relationship With Cockpit Automation

Credit: Shutterstock

The most important insight about the 200-foot point is that it does not represent a technological limit. Modern airliners can follow an ILS far below that altitude and, when properly equipped and authorized, complete an automatic landing. Instead, 200 feet marks the decision point for a conventional Category I operation, where the crew transitions from reliance on instrument guidance to the visual portion of the landing.

That balance is likely to remain important as flight decks become increasingly automated. Future aircraft will continue expanding automatic capabilities, but greater automation will not eliminate the need for pilots to understand when automation is appropriate, when it should be monitored, and when direct control is preferable. The trajectory of landing technology therefore points toward a more nuanced cockpit, rather than a pilotless one. In clear weather, the most efficient choice may remain a pilot manually completing the final moments of flight. In dense fog, the same aircraft may rely on redundant automatic systems to reach the runway when human vision cannot. The future of airline operations is not about choosing between pilots and automation. It is about giving crews enough capability to select the safest level of automation for the conditions in front of them.