[X] My Assistant

[dilo]

After first month on Mars, I think it's time to start a topic similar to MER one...
First odometry plot to start:

[mcaplinger]

Are you familiar with the coastline paradox? http://en.wikipedia.org/wiki/Coastline_paradox

There are so many error sources for odometry that it may not be physically meaningful to even discuss the "true" value. Even figuring out where the rover is at any given moment isn't free of ambiguity.

I think you're doing a great job with the data you have.

[jmknapp]

Are you familiar with the coastline paradox? http://en.wikipedia.org/wiki/Coastline_paradox

Yes! Benoit Mandelbrot's book had a chapter titled "How Long is the Coast of Britain" and the answer was: infinite, if you use a small enough ruler.

It's more important for me to keep the map position more or less right, because when it deviates I definitely start getting emails. And now the rover is poised to travel beyond the bounds of my initial HiRISE map tiles. I'm working to add another set of tiles--hopefully it'll be done before the rover moves much further, now just a couple hundred meters from the edge.

"There be dragons!"

[mcaplinger]

It's more important for me to keep the map position more or less right, because when it deviates I definitely start getting emails.

I don't know how the SPICE location is updated with information from outside the rover system. The rover basically can't know its absolute location in the Mars frame all by itself to good accuracy over a long time scale. The team uses map products that have the current site as the origin, so we never really use the SPICE location for anything in absolute terms as far as I know.

Then there's the question of how well the orbital mosaic was georectified in the first place. All in all, I suspect that putting the location on a map in an automated fashion using just the SPICE would be difficult, if not impossible (but I haven't really tried it, so I can't be sure.)