How did Apollo’s inertial navigation system work?

A Maritime Compass for the Ocean of Space

When the Apollo missions launched in the 1960s, they faced an unprecedented navigational challenge: how do you find your way to the Moon, a moving target 240,000 miles away, and then return safely to Earth? Unlike maritime explorers who could use the stars and horizon as references, or modern travelers with GPS satellites, Apollo astronauts needed a self-contained system that could determine “Where am I?” at all times, even when out of contact with Earth.

The answer was Apollo’s Primary Guidance, Navigation and Control System (PGNCS, pronounced “pings”), an engineering marvel that served as the spacecraft’s internal compass and speedometer. This inertial navigation system, installed in both the Command Module (CM) and Lunar Module (LM), allowed Apollo crews to navigate autonomously, especially when flying behind the Moon, where radio contact with Mission Control was impossible.

What made this system so revolutionary was not just its precision, but its reliability. Using 1960s technology, no GPS, no modern computers, Apollo’s guidance system successfully guided humans on a 500,000-mile round trip with accuracy measured in feet rather than miles. Let’s dive into how this remarkable system worked and why it remains one of NASA’s greatest technological achievements.

The Principles of Inertial Navigation: Knowing Your Place in Space

How Inertial Navigation Works

Inertial navigation operates on a deceptively simple principle: if you know your starting position and velocity, you can determine your current location by measuring all acceleration and rotation since your journey began. It’s like keeping a running tally of every step and turn you make while walking blindfolded.

At the heart of Apollo’s guidance system was the Inertial Measurement Unit (IMU), essentially a “stable platform” containing sensitive gyroscopes and accelerometers. This device provided a fixed, non-rotating reference frame inside the spacecraft, allowing the guidance computer to track the spacecraft’s movement through space.

The IMU used three rate-integrating gyroscopes to detect any rotation and three pendulous accelerometers to measure acceleration along orthogonal (perpendicular) axes. When the spacecraft rotated, servo motors on the gimbals would drive the platform to remain fixed in inertial space, maintaining its orientation regardless of the spacecraft’s movement.

The Apollo IMU: A Basketball-Sized Marvel

The Apollo IMU was a direct descendant of earlier missile guidance systems (evolved from the Polaris submarine-launched ballistic missile guidance), but refined for crewed spaceflight. It was a basketball-sized spherical assembly (approximately 12 inches in diameter) containing a stable platform inside.

Within this stable platform (also called the “stable member”) were the three gyros and three accelerometers, aligned perpendicular to each other. The whole assembly was mounted on three gimbals (inner, middle, and outer) that allowed the platform to rotate freely in any direction while the spacecraft moved around it.

Unlike earlier missile guidance systems, the Apollo IMU had no gear trains; the torque motors drove the gimbals directly to avoid backlash and wear. This improved reliability and allowed the platform to withstand rapid vehicle rotations, essential qualities for a system that had to work flawlessly for days on end.

The Complete PGNCS: More Than Just a Platform

The inertial platform was just one component of the broader Primary Guidance, Navigation and Control System (PGNCS). Each Command Module and Lunar Module had its own PGNCS installed, comprised of several key elements:

1. The IMU (inertial platform assembly)
2. The Apollo Guidance Computer (AGC)
3. Optical instruments for making star sightings (a sextant and scanning telescope in the CM, or an Alignment Optical Telescope in the LM)
4. A navigation base that tied these pieces together
5. Associated electronics like resolvers and power supplies

In the Command Module, the IMU was mounted under the couches on the navigation base along with the optics, ensuring they maintained a fixed geometry relative to each other. In the Lunar Module, the navigation base was likewise rigidly connected to the IMU, the optical telescope, and the rendezvous radar to a common frame.

Importantly, some functions were deliberately kept independent of the computer for reliability. For example, the Apollo Flight Director Attitude Indicator (FDAI)—the iconic “8-ball” attitude horizon instrument—was driven directly by the IMU gimbal angles, not by computer output. This meant the crew would always see the spacecraft’s true attitude as sensed by the inertial platform, even if the computer malfunctioned.

Component	Function	Location
Inertial Measurement Unit (IMU)	Provided space-fixed reference coordinates and measured accelerations along three axes	Both CM and LM
Apollo Guidance Computer (AGC)	Processed data from IMU and other sensors, calculated navigation solutions	Both CM and LM
Alignment Optical Telescope	Allowed astronauts to sight stars for platform alignment	LM only
Sextant and Scanning Telescope	More advanced optics for star sightings and celestial navigation	CM only
Flight Director Attitude Indicator (FDAI)	Displayed spacecraft’s attitude based on IMU data	Both CM and LM

The Brains of the Operation: Apollo Guidance Computer

All the inertial sensing hardware would have been useless without a way to process the data. The Apollo Guidance Computer (AGC) was the digital computer that tied the whole guidance system together, reading the IMU sensors, calculating the navigation solutions, and issuing commands to control the spacecraft’s engines and thrusters.

The AGC was a 15-bit word machine with about 36K words of fixed memory and 2K words of erasable memory—a remarkable feat of compact computing for its time. It continuously integrated the accelerometer outputs to update the spacecraft’s state vector (position and velocity in three dimensions) in near real-time.

The guidance equations ran in the background so that acceleration pulses from the IMU’s accelerometers were summed into velocity change registers (often called ΔV counters) and added to the running velocity estimate. Meanwhile, the gyros’ information (the gimbal angles) was used to compute the spacecraft’s attitude and rotate the measured acceleration vectors into the proper reference frame for navigation calculations.

During a powered engine burn, the AGC would use the IMU data to determine when the desired velocity change had been achieved, and then send a command to cut off the engine. It could also drive steering: the famous lunar landing guidance software in the LM’s AGC used the IMU’s acceleration and attitude data along with radar measurements to guide the LM down to the Moon’s surface, issuing thrust and attitude commands.

During coast phases (no engine firing), the AGC propagated the spacecraft’s trajectory purely by dead reckoning of the inertial data, occasionally corrected by new measurements like star sightings or radar fixes.

Learn more about the remarkable AGC in my detailed article on the Apollo Guidance Computer and how a 32KB computer saved the moon landing.

Astronaut Interaction: The DSKY and Optics

The crew interfaced with the AGC through the DSKY (Display and Keyboard unit), a panel with numeric displays and a calculator-style keyboard. Using two-digit codes for verbs and nouns, astronauts could request information or input commands to the AGC.

For instance:

• Verb 16 Noun 62 would display the IMU gimbal angles (spacecraft attitude)
• Verb 06 Noun 20 would display the current velocity components

One critically important routine was Program 52 (P52), the IMU realignment program. Because the inertial platform would drift over time, astronauts performed P52 periodically (typically every 8 to 12 hours, or before any critical maneuver) to recalibrate the platform’s orientation.

In a P52 alignment, the AGC would prompt the crew to sight two specific stars through the optical telescope. The crew would manually align the telescope on the first star and mark it, then do the same with the second star, allowing the computer to compare the observed star angles to its internally stored star catalog.

From the difference, the AGC would compute how much the platform had drifted and drive the platform’s gyro torquers to correct the orientation error, effectively rotating the stable platform back to perfect alignment with the chosen reference axes.

After a successful P52, the “Star angle difference” would be nearly zero (often just a few hundredths of a degree or less), and the IMU drift rate could also be estimated from how much correction was needed (typically on the order of 0.01–0.05°/hr, matching the approximately 1 milliradian/hr specification).

If you’re curious about the challenges of maintaining proper orientation in space, check out my article on Apollo guidance and navigation: understanding the challenges of IMU gimbal lock.

Reference Frames and REFSMMATs

Importantly, the astronauts could choose different reference frames for alignment depending on the mission phase. These reference orientations (called REFSMMATs- Reference Stable Member Matrix) were pre-defined attitude frames that optimized the gimbal orientations for different tasks.

For example:

• One REFSMMAT might align the platform with respect to Earth’s axis for a translunar cruise
• Another would align with respect to the lunar landing site’s local vertical for descent
• A third might align with respect to the reentry corridor for Earth return

By realigning the platform to a new REFSMMAT at appropriate times, the crew ensured the gimbals had plenty of range of motion around the upcoming maneuver attitudes (avoiding gimbal lock) and that the FDAI attitude display showed something useful (e.g., a level horizon) for that phase.

Coping with Drift: Keeping the System in Tune

Because inertial systems inherently drift over time, Apollo crews developed a rhythm of alignments and checks. A typical translunar coast saw alignments every 6 to 8 hours. Apollo 8’s crew, for instance, performed ten P52 alignments on the way to the Moon.

These frequent calibrations kept their guidance system “in tune”; mission reports show the platform drift was usually only a few arcminutes between alignments. If any anomaly was suspected, they could do additional sightings.

In early Apollo flights, the inertial system’s performance validated the design: Apollo 8 (the first lunar orbit mission) carried out ten separate platform alignments during its 6-day flight and found only minor drift each time, on the order of a few thousandths of a degree per hour, well within expected limits.

Astronaut Jim Lovell commented that almost without exception, Apollo 8’s spacecraft systems (including guidance) operated as intended on the trip to the Moon and back, a testament to the PGNCS design.

Mission	Number of Platform Alignments	Typical Drift Rate	Notable Navigation Events
Apollo 8	10 during 6-day mission	0.01-0.05°/hr	First lunar orbit navigation; platform performance exceeded expectations
Apollo 11	Multiple (standard schedule)	Within specifications	Successful landing within meters of target; brief gimbal lock during LM jettison
Apollo 13	Emergency alignments using Sun and Earth	Challenging conditions	IMU survived powered-down state; manual burns guided by LM platform
Apollo 16	Regular alignments	0.03-0.1°/hr	Delay due to CM guidance concern; AGS and PGNCS cross-checked during descent

Complementary Systems: Beyond the IMU

To complement the inertial system, Apollo’s guidance also integrated other sensors. The Command Module had a sextant for celestial navigation, astronauts could measure angles between stars and Earth’s horizon or the Moon to update their knowledge of position.

The Lunar Module had a rendezvous radar and a landing radar. The AGC could blend these measurements with the inertial solution. For example, during LM descent, the AGC used the landing radar’s measured altitude and velocity to update the state vector and null out any accumulated IMU error in altitude.

In lunar orbit, crews sometimes performed External Delta-V maneuvers using the Service Module’s reaction control thrusters and then updated their state vector via sightings. But if communications were lost, the crew could navigate independently with IMU and optics.

In fact, Apollo 8 famously navigated out to the Moon, orbited, and returned with minimal corrections; the onboard state vector was within a few nautical miles of the actual position after hours of coasting, demonstrating the accuracy of the inertial system aided by occasional star updates.

To learn more about the incredible people who monitored these systems from Earth, read my article about the Maestros of the MOCR: Meet the Apollo Flight Directors.

The Lunar Module’s Abort Guidance System: A Critical Backup

While the Command Module had only the PGNCS for guidance, the Lunar Module carried a secondary, independent inertial system: the Abort Guidance System (AGS). The AGS was developed by TRW as a simplified backup in case the LM’s primary PGNCS failed during the descent or ascent from the Moon.

It was the first ever use of a strapdown inertial system in crewed spaceflight, unlike the PGNCS’s gimbaled stable platform, the AGS’s Abort Sensor Assembly (ASA) used strapdown gyros fixed to the spacecraft structure. Without gimbals, the AGS relied entirely on software to integrate the gyro outputs to track attitude, a computationally intensive approach for the era.

The ASA also included accelerometers to measure acceleration in a couple of axes (enough to estimate velocity change during an abort ascent). To process this data, AGS had its own small computer, the Abort Electronics Assembly (AEA), with 4,096 words of core memory. The astronaut interface to the AGS was a simple Data Entry and Display Assembly (DEDA), essentially a numeric keypad and a small readout, more rudimentary than the DSKY.

The AGS was powered off during normal operations to conserve LM battery power. However, during critical phases like lunar descent, the crew would power it up and keep it running in standby. If needed, the AGS could take over attitude control and guide the LM through an abort ascent and rendezvous with the Command Module.

Throughout Apollo, the AGS thankfully never had to rescue a mission in a real abort scenario. But it was tested. Apollo 10 exercised the AGS during the LM’s ascent from lunar orbit, the crew switched to AGS control briefly and verified it could hold attitude and manage the abort profile.

For more context on how the Apollo guidance software was developed and tested, see my article on the Apollo 11 guidance software engineering humanity’s path to the Moon.

Dramatic Moments: When Navigation Saved the Day

Apollo 11’s Gimbal Lock Incident

A dramatic example of using backup methods occurred on Apollo 11 after the lunar landing. When Armstrong and Aldrin lifted off from the Moon in the LM, they had to rendezvous with Collins in the Command Module. After docking and transferring into the CM, Armstrong decided to jettison the LM ascent stage with an unusual maneuver, and in doing so, he unknowingly put the LM’s platform into gimbal lock.

As a result, the LM’s primary guidance (PGNCS) lost orientation. In an instant of quick thinking, Armstrong switched to the LM’s Abort Guidance System (which had been left powered on as a backup) to get an attitude reference and control the vehicle. He then redocked briefly to the Command Module to allow Aldrin to realign the LM’s inertial platform using the AGS data.

This incident, though minor in the mission timeline, showcased both the reality of gimbal lock and the value of having a redundant inertial reference. It was also the only time an Apollo crew had to rely on the AGS in flight due to a pilot-induced gimbal issue.

Apollo 13’s Manual Burns

Another noteworthy in-flight use of the inertial system was during Apollo 13’s crisis. With the Command Module powered down, Lovell and Haise used the LM’s PGNCS to guide critical burns such as the PC+2 burn (a propulsion correction burn done two hours after pericynthion) to refine their trajectory home.

In one case, Lovell famously had to manually control the spacecraft’s attitude while burning the LM’s descent engine, using the Earth’s terminator line in the window as a visual reference because they didn’t fully trust the platform alignment at that point. Even so, the IMU was running and the AGC provided timing—the burn was executed within a few feet per second of the intended change.

Apollo 13’s crew even had to improvise an alignment of the inertial platform by using the Sun and Earth’s terminator as reference objects (since a cloud of debris made star sightings impossible). They successfully aligned the LM’s IMU in this unorthodox way, allowing accurate burns to correct their course back to Earth.

For incredible stories about space medicine and the human body in space, check out Space Medicine 101: What Project Mercury Taught Us About the Human Body in Orbit.

Performance and Legacy: Engineering Excellence

In operational use, Apollo’s inertial navigation system met or exceeded all expectations. The accuracy achieved was remarkable. Apollo 8’s guidance ensured the spacecraft entered lunar orbit within mere miles of the planned trajectory, and Apollo 11’s landing point was only a few hundred meters from the targeted location (the last-minute manual flying by Armstrong was due to unexpected boulders, not a guidance error).

The inertial platforms on Apollo could detect rotations as small as a few thousandths of a degree and accelerations down to 0.0001 g. The drift rates were low, about 0.03 to 0.1° per hour in practice, and the frequent star sightings kept cumulative errors down to negligible levels for each mission phase.

Perhaps more impressive than pure accuracy was the system’s reliability and resilience. The combination of the hardened IMU hardware and the robust AGC software resulted in zero guidance-caused mission failures. Even when Apollo 14 experienced an erroneous abort signal just before powered descent (due to a faulty switch), the software was quickly patched to ignore it, and the inertial guidance continued smoothly to a landing.

The inertial systems also withstood the harsh conditions of space: temperature swings, vibration, etc. Apollo 13’s IMUs survived the chill of a powered-down transit and restarted without issue. During Command Module reentries, the IMU had to function through about 8 g of deceleration to provide attitude data to the auto-reactivation of the stabilization system upon exit from radio blackout, and they did so flawlessly.

If you’re interested in exploring how all this incredible navigation technology was documented, read about beyond the source code: exploring the saved documentation of the Apollo Guidance Computer.

Conclusion: Guiding Humanity’s Greatest Journey

Apollo’s inertial guidance system, the PGNCS with its stable platform and the supporting AGC, plus the backup AGS, was an engineering triumph. It gave the astronauts the equivalent of a space sextant, compass, and speedometer all in one, entirely onboard. With it, they could navigate to another world and back with an autonomy unprecedented at that time.

The system elegantly blended mechanical, electrical, and software ingenuity: from the floated gyros and their pickup coils to the core-rope memory of the computer encoding sophisticated guidance equations. And it was human-centric, astronauts became proficient in its operation, whether aligning on stars like Navi and Denebola (two favorites in the Apollo star list) or trusting the steady acceleration readouts during a critical burn.

The success of Apollo’s inertial navigation not only contributed directly to the success of lunar missions like Apollo 8 and Apollo 11, but it also paved the way for all modern spacecraft that sail the void between worlds using the same principles, merely updated for a new generation.

Today’s space enthusiasts can appreciate this technology as one of the key innovations that made the Moon landings possible. Want to explore the wonders of space yourself? Check out my guide to the best telescopes for amateur astronomers. Or learn about the top 10 space agencies in the world, continuing the legacy of space exploration today.

For more fascinating content about Apollo and space exploration, subscribe to my YouTube channel, where I regularly post videos diving deep into space history, technology, and the future of human exploration.

Best Telescopes 2025