1
$\begingroup$

This answer proposes that light bars would be better than a single headlight for night time navigation on the Moon because a single headlight will generate hard shadows. Considering this is exactly what sunlight does to the appearance of the skylight-less Moon's terrain for every rover that's been there as well as for the astronauts. (note that humans and some rovers have used binocular vision as well)

But hard shadows generate a lot of potentially helpful terrain information in addition to the loss of information in the dark areas, and this could be quite helpful.

Question: Have lunar (or Martian) rovers exploited shadows, perhaps for terrain comprehension or obstacle avoidance? Has there been any work on this technique or proposed systems that would exploit hard shadows produced either by sunlight (especially on bodies without a sky for fill-illumination)?

$\endgroup$
3
  • 1
    $\begingroup$ I don't understand the question. Shadows are the only way to see any structure on an uniformly colored surface like on the Moon. $\endgroup$
    – asdfex
    Oct 14 '19 at 10:01
  • $\begingroup$ @asdfex I think you do indeed understand the question and probably have the basis of a good answer! (i.e. on the Moon, shadows are our friends, and getting rid of them (e.g. by using a light bar) might not necessarily be such a good idea.) Here's a photo demonstrating just how bland the moon looks to a rover looking at the antisolar point for example. $\endgroup$
    – uhoh
    Oct 14 '19 at 10:22
  • 1
    $\begingroup$ I seem to remember some near escapes where Apollo LRV drivers nearly ended up in a crater because they misread the terrain (due to the strange lighting conditions and lack of visual references for scale and distance). $\endgroup$
    – Hobbes
    Oct 14 '19 at 14:34
1
$\begingroup$

This NASA article describes the autonomous driving of the Perseverance rover. In particular, Perseverance is the first rover to use vision for autonomous driving.

But increasingly, the rover will take charge of the drive by itself, using a powerful auto-navigation system. Called AutoNav, this enhanced system makes 3D maps of the terrain ahead, identifies hazards, and plans a route around any obstacles without additional direction from controllers back on Earth.

and

Unlike its predecessors, Perseverance can employ one of its computers just for navigation on the surface; its main computer can devote itself to the many other tasks that keep the rover healthy and active.

This Vision Compute Element, or VCE, guided Perseverance to the Martian surface during its entry, descent, and landing in February. Now it’s being used full-time to map out the rover’s journey while helping it avoid trouble along the way.

The rover compares stereoscopic images from two cameras:

[photo caption:] Perseverance relies on left and right navigation cameras. The view seen here combines the perspective of two cameras rover during the rover’s first drive using AutoNav, it’s auto-navigation function.

It can also detect changes between images from the same camera after it has moved:

The rover also keeps track of how far it’s moved from one spot to another using a system called “visual odometry.” Perseverance periodically captures images as it moves, comparing one position to the next to see if it moved the expected distance.

The article does not explicitly mention shadows or lighting. However, shadows certainly increase the differences between images that the rover relies on in its vision processing.

$\endgroup$

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.