I used NASA's eyes tool to see what's going at the time of New Horizon's closest approach to Pluto.

I realized that in the programmed sequence, New Horizons is rotated on its axis by more than 90° in just a few seconds, and then immediately maintains nadir fixed attitude with respect to Pluto to make the next observation.

How is such precise attitude control done mechanically?


3 Answers 3


I will try to answer your general question with application to NASA's New Horizons spacecraft.

The AOCS/GNC (Attitude and Orbital Control System / Guidance Navigation and Control) subsystem takes care that a spacecraft points to a specific point in space, determines the s/c attitude, and does trajectory corrections.

So in simple terms, an AOCS system is built up by:

  • sensors (star-trackers, Earth sensor, magnetometer IMU, etc.)
  • actuators (the different thrusters, reaction wheels, magnetorquers)
  • control algorithm (for example C code).

Now, to put this into perspective, you need a basic diagram to understand the very basics of control engineering:

             enter image description here

A typical, single-input, single-output feedback loop with descriptions for its various parts. Image: Wikimedia Commons

I won't get into too much detail. In essence: you are trying to minimize the error based on sensory feedback via controlling the s/c actuators.

According to Johns Hopkins Applied Physics Laboratory's (APL's) mission description, Spacecraft Systems and Components page, New Horizons has the following GNC system:

Attitude determination – knowing which direction New Horizons is facing – is performed using star-tracking cameras, Inertial Measurement Units (containing sophisticated gyroscopes and accelerometers that measure rotation and horizontal/vertical motion), and digital Sun sensors. Attitude control for the spacecraft – whether in a steady, three-axis pointing mode or in a spin-stabilized mode – is accomplished using thrusters.

The IMUs and star trackers provide constant positional information to the spacecraft’s Guidance and Control processor, which like the Command and Data Handling processor is a 12-MHz Mongoose V. New Horizons carries two copies of each of these units for redundancy. The star-tracking cameras store a map of about 3,000 stars; 10 times per second one of the cameras snaps a wide-angle picture of space, compares the locations of the stars to its onboard map, and calculates the spacecraft’s orientation. The IMU feeds motion information 100 times a second. If data shows New Horizons is outside a predetermined position, small hydrazine thrusters will fire to re-orient the spacecraft. The Sun sensors back up the star trackers; they would find and point New Horizons toward the Sun (with Earth nearby) if the other sensors couldn’t find home in an emergency.

Operators use thrusters to maneuver the spacecraft, which has no internal reaction wheels. Its smaller thrusters are used for fine pointing; thrusters that are approximately five times more powerful are used during the trajectory course maneuvers that guide New Horizons toward its targets. New Horizons spins – typically at 5 revolutions per minute (RPM) – during trajectory-correction maneuvers and long radio contacts with Earth, and while it “hibernated” during long cruise periods. Operators steady and point the spacecraft during science observations and instrument-system checkouts.

  • $\begingroup$ Thank you for this great answer. I know what losed control loops are, but my actual question was what kind of sensors and actors are used for that - because I had no idea how this could be done in space at this speed and the required accuracy. Sorry for beeing unclear in my question. $\endgroup$
    – joe
    Jul 21, 2015 at 17:37

According to the NASA New Horizon's site

New Horizons has operated mostly in a spin-stabilized mode while cruising between planets, and also in a three-axis “pointing” mode that allows for pointing or scanning instruments during calibrations and planetary encounters (like the Jupiter flyby and, of course, at Pluto). There are no reaction wheels on the spacecraft; small thrusters in the propulsion system handle pointing, spinning and course corrections.

  • $\begingroup$ This addresses how NH changes its attitude, but not how it determines its current attitude, nor Pluto's relative position. $\endgroup$ Jul 16, 2015 at 0:26

Attitude measurement is done by an inertial navigation unit (gyroscopes) and by modified Galileo Avionica A-STR star trackers.

Pluto's position is not measured by the spacecraft. During the encounter, the spacecraft is preprogrammed to point in direction X at time Y and they have to hope Pluto is in frame. (source: the National Geographic documentary "Mission Pluto")

From a recent Washington Post article:

I asked Project Manager Glen Fountain if it was possible the spacecraft would turn in the wrong direction and take photographs of empty space, and he said no — not unless the ephemeris data was way off.

Emphasis mine: they use ephemeris data collected on Earth, plus navigation images made by NH's main cameras which are interpreted on Earth.

For attitude control, NH uses 12 Aerojet MR-103H hydrazine thrusters, which were chosen because they can be run for intervals as short as 4 ms. A single solenoid valve controls the amount of fuel injected. Because this is a monopropellant, there's no uncertainty about mixing propellant and oxidiser, and thrust is predictable even at these short bursts. The result is that the craft can be pointed accurately enough using thrusters alone.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.