A [nav]cam pointed straight down, (for visual odometry) Options

[stevesliva]

A brief mention of the time it takes for visual odometry in the Spirit route thread got me thinking off on digital signal processing tangents--

The optical mouse you're using right now has no issues with visual odometry, unless you use it on too plain a surface, and the basic principle is a camera taking lots of tiny photos of the surface connected to a DSP.

Have any rover technologies looked at analogous techniques for really precise odometry? Or perhaps navigation is a better term than odometry because it can handle arbitrary vectors. Is it too tough to do this with an eyeball farther from the ground? Without a light source that can flatten the light for a larger area? Isn't this easier than doing odometry with cameras looking any other direction?

Does the rover visual odometry use lots of successive photos of the most immediate objects, or does it try to use big objects farther away as waypoints?

[Geert]

Have any rover technologies looked at analogous techniques for really precise odometry? Or perhaps navigation is a better term than odometry because it can handle arbitrary vectors. Is it too tough to do this with an eyeball farther from the ground? Without a light source that can flatten the light for a larger area? Isn't this easier than doing odometry with cameras looking any other direction?

The old Lunochods just carried a 9th (unpowered) wheel center aft which was used to measure odometry, that's probably the most simple solution although offcourse you will have to correct for drag, etc, etc.

I don't think you will need a camera, you can use a small laserbeam, pointing down, exactly similar to the technique an optical mouse is using, that will give you very exact data, but like everything else it will have a price in extra weight, extra complexity, and extra costs. We can dream up a 'perfect' rover but nobody will have the budget to pay for it...

[lyford]

MARDI might be up for this job on MSL if I understand correctly.

Once on the ground images will cover an area about 1 by 0.75 m across at a resolution of about 1.5 mm per pixel pair.

Though I don't know if any plans exist for such use once landed.

EDIT - Oops! Shoulda read the whole thing here:

Although not an original requirement or objective of the investigation, additional images may be taken during rover traverses for visual odometry and geologic mapping.

Yeeeehaw!

[CosmicRocker]

That's an interesting idea, Steve.

... the whole thing here:...

I suppose the Swiss Army knife in the image on that page may have been meant to emphasize the instrument's myriad potential uses.

[stevesliva]

Neat! Thanks, lyford. If you've got the framerate (they do) and the processing power, I'd imagine it'd be harder to find a better direction to point for visual odometry.