15
$\begingroup$

On Earth, wild life cameras are sometimes equipped with motion detectors to record rare species automatically.

Could they not be applied in the case of Mars too ?

Is Curiosity not equipped with motion detectors to be able to record moving objects like boulders or dust devils automatically ?
Even shadows and meteorites ?

And if not, would that not be a good idea for the Mars 2020 mission ?

$\endgroup$
  • 4
    $\begingroup$ Sure, it could, but why? What might be the scientific benefit of such a sensor? Remember: every ounce of a payload is precious and anything that doesn’t bring significant benefit is a waste. $\endgroup$ – Paul Dec 10 '18 at 14:40
  • $\begingroup$ @Paul With such sensors it could record (unexpected) events or objects like meteorites, falling rock, dust devils and maybe things we would never expect beforehand. $\endgroup$ – Conelisinspace Dec 10 '18 at 15:06
  • 1
    $\begingroup$ If the Opportunity rover had motion detectors we might know a bit more about how this rock appeared "from nowhere" 12 days after a picture was taken of the same area. $\endgroup$ – Fred Dec 11 '18 at 16:29
  • $\begingroup$ @Fred "Mysterious doughnut" ! NASA thinks it may have been flipped upside down when a wheel dislodged it. Good to know it has high concentrations of sulfur and magnesium ! $\endgroup$ – Conelisinspace Dec 11 '18 at 18:41
24
$\begingroup$

Most conventional motion detectors wouldn't work well on Mars.

  • PIR: Usually only works well with warm bodies such as humans and animals. Would probably never trigger on Mars

  • Ultrasonic: Due to the extremely low pressure on Mars any acoustic sensing through the atmosphere would be greatly hindered. These sensors usually only pick up on very large, perpendicularly reflective objects. Such a sensor also has a very short range and would also never trigger on Mars (unless the rovers arm swings past the sensor)

A radio-based motion detector would work in principle but there's basically nothing that moves on Mars with any real mass. Rocks only really move on geological timescales and the only thing that would ever move on the surface near the rover is a dust devil or other wind activity (the rover would want to be far away from regions where landslides are possibilities).

This however leaves us with one sensor on the list, "video camera software". The Curiosity rover currently has 17 cameras and it takes lots of pictures that can be compared with each other to detect motion. For example, here is a "video" from Curiosity recording a dust devil moving across the surface:

Dust devil on mars

Summary: Motion detection is a software feature which doesn't require special hardware to work, it just needs the already existing cameras.

$\endgroup$
  • 3
    $\begingroup$ very nice answer! "...there's basically nothing that moves on Mars with any real mass. Rocks only really move on geological timescales..." check Tall boulder rolls down martian hill, lands upright and also How to use this Google Mars viewer to find this Martian landslide event? but yes, these are pretty rare ;-) $\endgroup$ – uhoh Dec 10 '18 at 15:43
  • 4
    $\begingroup$ "nothing that moves on Mars with any real mass." -- that we know of :) $\endgroup$ – Barmar Dec 10 '18 at 19:37
  • 1
    $\begingroup$ Well, the whole point of such a scientific instrument would be to confirm or deny your assumption that "Rocks only really move on geological timescales and the only thing that would ever move on the surface near the rover is a dust devil or other wind activity (the rover would want to be far away from regions where landslides are possibilities)". But I do of course agree that such a mission would be neither economical nor particularly useful in all likelihood. $\endgroup$ – Lightness Races with Monica Dec 11 '18 at 10:54
  • 2
    $\begingroup$ Done by software on board. That's how they catch the dust devils. Otherwise it would just be incredible luck. This software was first deployed on the Mars Exploration Rovers in 2007. Also used to catch clouds. $\endgroup$ – Mark Adler Dec 12 '18 at 5:41
  • 1
    $\begingroup$ InSight adds "seismic detector" to the mix. Once it's deployed, the seismometer will detect anything with any serious mass that does move (like a landslide, a Martian walking past or a meteorite impact ). $\endgroup$ – Steve Linton Dec 12 '18 at 11:19
1
$\begingroup$

So, the other answer establishes the least-disruptive motion sensor is using motion analysis software on the camera feeds.

Doing the processing on the ground is easy enough, but we don't receive enough images for this. Curiosity does not repeatedly image the same scene. If you want to do that, the amount of transmitted data goes up, and in another question we've seen there's not much room to do that.

So the question becomes, can this be done in a rover's onboard systems?

Apart from processing power, there is an electrical power issue. The rover's systems draw more power than the RTG supplies, so the rover has to hibernate much of the time to recharge its batteries. This limits motion detection to the times the rover is active anyway, and that time is spent traveling or operating the instruments. It may be possible to take some extra pictures during instrument operations and run a comparison on them.

$\endgroup$
  • $\begingroup$ With a CCD electrical charges are converted into digital values in memory. This CCD could be in an active state all day (and night?) Could not software be designed to registrate an image in memory only if one or a few pixels from the CCD differ from the last stored image ? That could save a lot of memory and power ? $\endgroup$ – Conelisinspace Dec 12 '18 at 11:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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