For a drone to be autonomous, it needs something like a GPS to be able to navigate. If it doesn't have GPS, you'd have to manually control it.

Do the orbiters provide something like GPS then? I thought the orbiters functioning now were too old to have those technologies.

  • 81
    $\begingroup$ A roomba can navigate autonomously without using GPS $\endgroup$
    – user20636
    Commented Oct 24, 2018 at 16:20
  • 41
    $\begingroup$ @JCRM Now you have me imagining a drone that navigates by crashing into every possible surface :P $\endgroup$
    – Shadow
    Commented Oct 25, 2018 at 3:50
  • 34
    $\begingroup$ People have been navigating autonomously for millenia before the existence of GPS. Why do you think a drone could not use those exact same techniques? $\endgroup$ Commented Oct 25, 2018 at 8:10
  • 16
    $\begingroup$ @LightnessRacesinOrbit - If a Martian GPS-type systems were being spoofed, NASA has some explaining to do... $\endgroup$
    – Jon Custer
    Commented Oct 25, 2018 at 12:54
  • 2
    $\begingroup$ @JonCuster More generally addressing the false assumption that GPS is not only something you need to use for this kind of task but something that you should solely use in the first place! But yes lol $\endgroup$ Commented Oct 25, 2018 at 12:57

6 Answers 6


Edit: The JPL Mars Helicopter Scout will use inertial navigation:

The inconsistent Mars magnetic field precludes the use of a compass for navigation, so it will use a solar tracker camera integrated to JPL's visual inertial navigation system. Some additional inputs might include gyros, visual odometry, tilt sensors, altimeter, and hazard detectors.[15]

There are other ways to navigate than GPS.

  • Inertial navigation uses accelerometers to detect in which direction you're moving, and speed or distance sensors to detect how far you've gone. Used e.g. by submarines to navigate when they're underwater and out of GPS reach. Also used by current Mars rovers.
  • You can use radio direction finding. With 2 transmitters in known locations, you can triangulate your position. With 1 transmitter navigation is more limited but you can still get back to the transmitter location. Used in WW2 by aircraft.

In addition to knowing your position, you need to know something about the terrain around you. You need to know which spots can be traversed, and which are too steep/rocky/soft for the vehicle to traverse. For an aircraft, you need to know where the mountains are and where it's safe to land.

You can use a map built by someone else as your reference (this is what satnav in your car does), or you can build your own map as you go along.

The current Mars rovers have demonstrated autonomy: they take pictures of the terrain ahead, and they can choose a safe path. GPS is not necessary for this.

  • 5
    $\begingroup$ 2 transmitters is not enough; 2 circles intersect at 2 points, so using only 2 transmitters leaves 2 possible solutions for positioning (unless both transmitters are in exactly opposite directions), so you still need an additional piece of evidence to choose between the 2. $\endgroup$
    – gerrit
    Commented Oct 24, 2018 at 14:56
  • 22
    $\begingroup$ @gerrit: 2 transmitters are enough. Antennas are directional and can be used to determine the direction of a radio source. If you use transmeitters A and B, the two solutions would have A on the left and B on the right, or vice versa, so they can be distinguished. $\endgroup$ Commented Oct 24, 2018 at 16:11
  • 4
    $\begingroup$ @Uwe: Right, but that’s a problem no matter how many antennas you have, and it’s a problem with GPS—it’s possible to have poor reception. Two beacons is just a starting point. The point is that you can make the receivers directional. $\endgroup$ Commented Oct 24, 2018 at 16:46
  • 8
    $\begingroup$ @DietrichEpp: Alternatively, one could use radio lighthouses. Use a rotating antenna which produces a null at a particular angle, and continuously transmit a signal indicating which way it's pointing. The signal strength will drop off enormously once per rotation, and by figuring out what the angle is each time that happens will allow one to know one's bearing from the transmitter, without the receiver needing a directional antenna. $\endgroup$
    – supercat
    Commented Oct 24, 2018 at 17:01
  • 16
    $\begingroup$ @gerrit That's not a problem since you generally know roughly where you are from other sources. The old Omega system that planes and ships used before GPS only used two measurements, and gave you possible 4 locations, but pilots generally knew hemisphere of the Earth they were currently in. $\endgroup$
    – user71659
    Commented Oct 24, 2018 at 17:33

GPS is one of several possible technologies available for assisted navigation. It's commonly used on commercial drones because the framework is in place and GPS signal is usually available on Earth.

It requires a flotilla of satellites around our planet to work, though - something we don't have (yet) around other celestial bodies. In that case we need different approaches, such as:

  • Solar tracking, together with Inertial navigation (as mentioned in Hobbes' answer) will be the navigational technology of choice for the Mars Helicopter Scout since Mar's weak magnetosphere can't be relied on:
    KVH’s 1750 IMU KVH’s 1750 inertial measurement unit(IMU), primary navigation sensor on AERO. Link

  • Astrometry - Δ-DOR, or Delta-Differential One-Way Ranging can use quasars as calibrators in order to provide positional resolution under a few hundred meters on Mars to an orbiting vessel than can, in turn, share the information with surface vehicles:
    Delta-DOR definition ESA Delta DOR: from implementation to operation R. Maddè, T. Morley ESOC, 16 th March 2007. Link

  • Surface mapping - a drone may store 3D surface maps of its surroundings, and then use LiDAR/RADAR sensing (plus altimeter, gyros and inclinometers) together with matching algorithms to determine its current position: TERCOM - process, Golden 1980 TERCOM - process, Golden 1980 (ieee.org) - Link

  • Visual Odometry - Image processing can determine how fast a vehicle is by measuring the differences between two still images if the time lapse between them is known;

Several techniques can be combined in order to provide better accuracy.

  • 1
    $\begingroup$ @Uwe You're completely right. In my defense, I wanted to keep it short and to the point. Adding to your comment, even general relativity is used to compensate for time dilation-related error analysis by the ground stations, as described here: en.wikipedia.org/wiki/… $\endgroup$
    – OnoSendai
    Commented Oct 24, 2018 at 15:35
  • 5
    $\begingroup$ Surface mapping is, incidentally, how cruise missiles used to operate. It's well tested, and very functional, as various US-targeted regimes can attest to. (GPS has supplanted it, but it'd likely work quite well for Mars.) $\endgroup$
    – ceejayoz
    Commented Oct 24, 2018 at 16:47
  • 2
    $\begingroup$ @ceejayoz Precisely - added a reference image for TERCOM, thanks for bringing that up. $\endgroup$
    – OnoSendai
    Commented Oct 24, 2018 at 18:06
  • 1
    $\begingroup$ There were a number of accurate radio-navigation techniques in use pre-GPS. These included: active radio ranging (measure distance from vehicle to fixed point), radio bearing (measure solid angle from vehicle to fixed point), passive hyperbolic (measure time difference between signal from two stations), doppler (measure velocity of a single satellite). These could be used in conjunction [although as below, they've gone with optical terrain based]. $\endgroup$
    – Rich
    Commented Oct 24, 2018 at 22:18
  • 1
    $\begingroup$ @uhoh It wouldn't be much use for real-time navigation, but if the helicopter makes short, planned hops it'd be good for fixing current position before embarking on one. Might be a handy supplement for inertial/vision-based real-time guidance. $\endgroup$
    – ceejayoz
    Commented Oct 25, 2018 at 0:32

I've reused some material from this answer here to show that the MARS 2020 rover will land on Mars using optical navigation in part.

A helicopter can use similar environmental learning techniques developed for robots on Earth. This isn't a perfect example, but it gives the idea that the robot builds up a map over time. At each point you can identify the location of a new landmark with respect to the previous landmarks.

From Entry, Descent, and Landing Technologies:


The key to the new precision landing technique is choosing the right moment to pull the "trigger" that releases the spacecraft's parachute. "Range Trigger" is the name of the technique that Mars 2020 uses to time the parachute's deployment. Earlier missions deployed their parachutes as early as possible after the spacecraft reached a desired velocity. Instead of deploying as early as possible, Mars 2020's Range Trigger deploys the parachute based on the spacecraft's position relative to the desired landing target.

enter image description here

above: Illustration of the Range Trigger concept - using the historical /Mars Science Laboratory(MSL)/Curiosity landing site as an example.

Also, during the descent stage, images will be compared in real time to stored terrain maps to improve the accuracy of the landing location:


Terrain-Relative Navigation significantly improves estimates of the rover's position relative to the ground. Improvements in accuracy have a lot to do with when the estimates are made.

In prior missions, the spacecraft carrying the rover estimated its location relative to the ground before entering the Martian atmosphere, as well as during entry, based on an initial guess from radiometric data provided through the Deep Space Network. That technique had an estimation error prior to EDL of about 0.6 - 1.2 miles (about 1-2 kilometers), which grows to about (2 - 3 kilometers) during entry.

Using Terrain-Relative Navigation, the Mars 2020 rover will estimates its location while descending through the Martian atmosphere on its parachute. That allows the rover to determine its position relative to the ground with an accuracy of about 200 feet (60 meters) or less.

It takes two things to reduce the risks of entry, descent, and landing: accurately knowing where the rover is headed and an ability to divert to a safer place when headed toward tricky terrain.

enter image description here

above: Illustration of Terrain-Relative Navigation. "Terrain-Relative Navigation helps us land safely on Mars - especially when the land below is full of hazards like steep slopes and large rocks! From here.

In December 2014 the vision system was tested in the Mojave Desert. :

enter image description here

above: "A prototype of the Lander Vision System for NASA's Mars 2020 mission was tested in this Dec. 9, 2014, flight of a Masten Space Systems "Xombie" vehicle at Mojave Air and Space Port in California. Credit: NASA Photo/Tom Tschida" From here.

  • $\begingroup$ Of course, given the track record of Mars exploration, it'll probably fail to deploy the parachute at all... $\endgroup$
    – Vikki
    Commented Oct 25, 2018 at 1:04
  • 2
    $\begingroup$ @jwenting my next question How can the Martians distinguish the space agency of an approaching spacecraft in order to decide if they'll let it land? $\endgroup$
    – uhoh
    Commented Oct 25, 2018 at 12:04
  • 2
    $\begingroup$ @uhoh does it try to pay its landing fees in Euros or Dollars? ;) $\endgroup$
    – jwenting
    Commented Oct 25, 2018 at 12:14
  • 4
    $\begingroup$ @jwenting: Actually, it isn't just ESA missions. Far from it. $\endgroup$
    – Vikki
    Commented Oct 26, 2018 at 2:52
  • 1
    $\begingroup$ @jwenting that's right. There are NASA failures there too. That was kind-of my point; the answer to my "proposed next question" would be "they don't" or "they roll 42-sided Martian dice like everyone else". $\endgroup$
    – uhoh
    Commented Oct 26, 2018 at 2:56

Inertial navigation with occasional "fixes" to reset position to within desired accuracy limits.

Fixes can be photos of ground, positions of stars, radio triangulation or input from external tracking stations, not just satellites. An inertial platform will have gyros for rigidity, accelerometers to sense changes in velocity (acceleration) and gimbals to detect angular changes. Spin up the gyros, enter a start position and direction. And have a computer sense any changes from the start to update latitude and longitude. External fixes are only needed for corrections to errors that propagates over time.


Radar, ground mapping techniques, accelerometers and many other ways to get a relative idea of where you are.


An Austrian university is working on that right now. They use an onboard camera for the navigation (see below).

Also, a recent drone on Kickstarter uses odemetry for positioning. The project did not end that well, but on windless days, if there are no problems, the drone works: See here

From Alpen-Adria-Universität Klagenfurt's Researchers from Klagenfurt participate in an Analog Mars Mission in Oman:

“Aircraft are fast and very maneuverable. They can explore unknown environments, for instance on Mars, more rapidly than ground-based robots”, Stephan Weiss (Department of Smart Systems Technologies at the AAU) explains. However, he says: “The problem we have is the navigation of these types of drones in situations where there is no GPS to rely on.” To address this issue, he and his team have been working on further developing camera-based methods for autonomous navigation designed, among others, to facilitate the deployment of helicopter aircraft during Mars Missions.

The drone to be tested is a small multicopter aircraft equipped with a camera and inertial measurement unit as well as a data processing system. Stephan Weiss explains: “The navigation algorithms on the drone have been designed to navigate over the terrain exclusively based on images recorded by its own on-board camera and on input from the system’s own inertial measurements.“ In the case of the mission in Oman, following the deployment of the aircraft, the images recorded by the on-board camera will be synchronised with inertial data and GPS. The goal is to verify whether the visual position sensing of the aircraft functions properly. “We plan to deploy the multicopter at different times of the day and over different landscapes. Subsequently, the gathered results will be compared”, Weiss explains.

The experiments will be conducted within the scope of the AMADEE-18 Mission of the Austrian Space Forum. Mission Control is located at the Mission Support Center in Innsbruck. This team matches Earth’s Ground Control in size and will communicate with the 15 members of the field crew in Oman, comprising members from nine different nations, including analogue astronauts. They will be conducting experiments in the fields of engineering, manned exploration of planetary surfaces, astrobiology, geophysics/geology, life sciences and more. The mission is scheduled to take place from the 1st to the 28th of February 2018.

  • 1
    $\begingroup$ This is a link-only answer. Can you edit it and include the essential parts that answer the question here? $\endgroup$
    – user10509
    Commented Oct 25, 2018 at 8:47
  • $\begingroup$ the tl;dr explanation is that links break over time and the answer becomes useless to future readers unless you capture some key points here, either by summarizing the important information or citing it. I've gone ahead and shown an example of how that might work, but please feel free to rollback or edit further. Welcome to Space! $\endgroup$
    – uhoh
    Commented Oct 25, 2018 at 9:29
  • $\begingroup$ slightly related: What is the Austrian Space Forum doing in Oman? $\endgroup$
    – uhoh
    Commented Oct 25, 2018 at 9:39

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.