The LM Digital Autopilot

Nothing about the LM itself would be intuitive to fly. It had sixteen RCS (reaction control system) thrusters (four clusters of four) for attitude control, but the large descent engine on the bottom pivoted on gimbals and could also control attitude and hence steer the spacecraft. Failure of a single RCS thruster, or a cluster of them, could send the vehicle spinning. The center of gravity moved as the engines consumed fuel, and sloshing of the propellants could make the spacecraft difficult to handle. These and any number of other complexities meant the LM had no inherent or natural match to a human pilot. Only the LM's computer, a set of software routines combined with a host of data, sensors, and actuators, could give the astronauts the feeling that they were ''flying'' the vehicle. These routines comprised the digital autopilot.

The autopilot took over the computer every tenth of a second for its calculations, which took about a fortieth of that time (twenty-five milliseconds) to complete. In addition, during powered flight every two seconds another routine would intervene to adjust the autopilot and its parameters. The digital autopilot maintained a virtual model of the vehicle inside its ''state estimator'' (equations similar to Battin's recursive techniques) that kept track of the various forces generated by and acting on the LM. When the thrusters fired, it estimated their effects on the LM's attitude and incorporated them into a new estimate even before their effects showed up on the accelero-meters. ''Jet selection logic'' automatically determined which of the sixteen thrusters would fire in response to a command. If a thruster failed, or a cluster of them failed, the selection logic would automatically sense the failure and compensate with other actuators. The state estimator calculations automatically compensated for the changes in vehicle mass as the engines consumed fuel (a less massive vehicle accelerates more in response to a given thrust).

One set of digital autopilot routines related to ''coasting flight,'' when the spacecraft was in orbit but not changing its velocity (such as during the visual inspection after undocking). A program called KALCMANU (for ''calculate maneuver'') could rotate the spacecraft to any specified orientation in the most efficient manner, gently leading the spacecraft's attitude servos for a smooth motion and avoiding gimbal lock.

Another set of routines controlled powered flight, when the descent or ascent engines were firing, changing the spacecraft's velocity. Powered descent to the lunar surface represented the most complex case. Ten times per second, the computer read the changes in velocity and attitude from the inertial platform as the engine slowed the spacecraft out of orbit. The guidance equations (running twice per second) then extrapolated that data to get the spacecraft's new position and velocity, and determined new thrusting commands accordingly. These routines responded to higher-level guidance routines that determined the desired position and velocity of the spacecraft at any given point and sequenced through the three landing phases.20

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment