PDA

View Full Version : Spirit Looks Back



harlequin
2004-Mar-11, 10:31 PM
Spirit look down on its previous locations (http://qt.exploratorium.edu/mars/spirit/pancam/2004-03-11/2P132219500EFF1600P2283L6M1.JPG)

It is the first time I have seen the lander in a Spirit photo is some time.

The directory has load and loads of rocks. It is definately more rocks near the rim of Bonneville.

Manchurian Taikonaut
2004-Mar-11, 10:32 PM
wonderful picture, all we need now is a little color and that martian sunset

Amadeus
2004-Mar-11, 10:46 PM
I love the way the tracks zig-zag! It reminds of getting home from a late night in the pub :D

tlbs101
2004-Mar-11, 10:48 PM
Considering that the last half of the drive (most of it anyway) was done autonomously, and that the last half is the worst half, obsticle-wise; I am very impressed with the JPL AI software team.

Look at the path Spirit took to avoid certain tall rocks, and yet plow right over other smaller ones.

Good job guys!

=D>

ToSeek
2004-Mar-12, 12:21 AM
Considering that the last half of the drive (most of it anyway) was done autonomously, and that the last half is the worst half, obsticle-wise; I am very impressed with the JPL AI software team.

Look at the path Spirit took to avoid certain tall rocks, and yet plow right over other smaller ones.

Good job guys!

=D>

Actually, my impression is that it was kind of a mix: the mission control team would plan out the route as far as they could see it, then let the rover continue on for another 5-10 meters after that autonomously.

I'm just waiting for them to do the same with Opportunity. I think it's going to be something like "Just keep going in this direction until you run into Endurance Crater. Then stop."

harlequin
2004-Mar-13, 04:13 AM
Another look at the lander (http://qt.exploratorium.edu/mars/spirit/pancam/2004-03-12/2P131773257EFF1159P2214L5M1.JPG) from March 12. The shiny thing below and to the right appears in multiple images.

Maksutov
2004-Mar-13, 06:30 AM
Spirit look down on its previous locations (http://qt.exploratorium.edu/mars/spirit/pancam/2004-03-11/2P132219500EFF1600P2283L6M1.JPG)

It is the first time I have seen the lander in a Spirit photo is some time.

The directory has load and loads of rocks. It is definately more rocks near the rim of Bonneville.

Interesting to look at the landscape above and beyond the lander and note all the apparent layering. :-k

Crazieman
2004-Mar-13, 03:55 PM
Another look at the lander (http://qt.exploratorium.edu/mars/spirit/pancam/2004-03-12/2P131773257EFF1159P2214L5M1.JPG) from March 12. The shiny thing below and to the right appears in multiple images.

Can't wait for the Hoagland explanation(tm) of that odd shiny thing.

Manchurian Taikonaut
2004-Mar-15, 12:56 PM
Great view of sky and image looking back at the tracks made through the martian soil

http://qt.exploratorium.edu/mars/spirit/navcam/2004-03-14/2N132580528EFF1800P1948L0M1.JPG

http://www.spacedaily.com/images/mars-mera-tracks-bg.jpg

Rift
2004-Mar-15, 03:24 PM
I love the way the tracks zig-zag!

Well, it had all those other martian bunnies to run over before Hoagland saw them... :P

Manchurian Taikonaut
2004-Mar-17, 01:10 AM
the real views of spirit

http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040113a/Horizon_hills_azimuths-A11R1_br2.jpg

http://qt.exploratorium.edu/mars/spirit/forward_hazcam/2004-03-16/2F132758240EFF1957P1201L0M1.JPG
(getting stuck in sandy soil?)

http://qt.exploratorium.edu/mars/spirit/micro_imager/2004-03-16/2M132663697EFF1800P2973M2M1.JPG
great image

tlbs101
2004-Mar-17, 06:11 PM
tlbs101 wrote:
Considering that the last half of the drive (most of it anyway) was done autonomously, and that the last half is the worst half, obsticle-wise; I am very impressed with the JPL AI software team.

Look at the path Spirit took to avoid certain tall rocks, and yet plow right over other smaller ones.

Good job guys!



Actually, my impression is that it was kind of a mix: the mission control team would plan out the route as far as they could see it, then let the rover continue on for another 5-10 meters after that autonomously.

BTW: How does one do a nested quote on this board?



I read that the rover driver team had been doing what you describe, but since one of the hollows before "middle ground", they have let Spirit do most of the driving itself.

My impression is that (since that date) Spirit has autonomously found its own way for 15-20 meters, then come upon some obstacle that it can't figure out, or the odometer is correct, but Spirit is short of where it actually is; and the driving team has to take over manually for the last few meters.

ToSeek
2004-Mar-17, 07:29 PM
Actually, my impression is that it was kind of a mix: the mission control team would plan out the route as far as they could see it, then let the rover continue on for another 5-10 meters after that autonomously.



I read that the rover driver team had been doing what you describe, but since one of the hollows before "middle ground", they have let Spirit do most of the driving itself.

My impression is that (since that date) Spirit has autonomously found its own way for 15-20 meters, then come upon some obstacle that it can't figure out, or the odometer is correct, but Spirit is short of where it actually is; and the driving team has to take over manually for the last few meters.

Well, let's look at the Status reports (http://marsrovers.jpl.nasa.gov/mission/status_spiritAll.html#sol71):

Sol 71: 15-meter blind drive followed by 3 meters of autonomous navigation
Sol 67: direct drive followed by automatic
Sol 66: directed drive followed by auto-navigation
Sol 65: directed drive followed by auto-navigation
Sol 64: direct drive followed by autonomous navigation
Sol 62: not specified
Sol 61: not specified
Sol 54: ground-controlled
Sol 53: not specified
Sol 52: ground-controlled
Sol 51: first 18 meters direct, last 12 auto-navigation
Sol 50: first 16 meters direct, last 2.8 autonomous

I don't think the planning cycle allows for the approach you indicate. For the ground controllers to direct the vehicle, they need to have images taken, downlink the images, analyze the images, create a command package to control the driving, and uplink the package. That's not something that's going to be done in real-time. My impression is that the cycle is more like:

Sol X (near end of day):
- Rover stops moving
- Rover takes images
- Rover downlinks images
- Rover goes to bed
- Scientists look at images and select destination
- Drivers look at images and select route
- Drivers build command package to take route as close to destination as imaging will allow
Sol X+1 (the next morning):
- Rover wakes up
- Commands uploaded
- Rover drives to end of pre-programmed route
- If not at destination, rover directed to navigate autonomously toward destination
- Rover gets as far as it can based on terrain and power
- Rover takes images
... and the cycle begins again.

I would assume that ground-directed driving is faster because the rover doesn't have to keep stopping to analyze the terrain, which is why driving is commanded whenever possible.

ToSeek
2004-Mar-18, 01:26 AM
Spirit looks back - shows Spirits travels so far (http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040317a/Bonneville_Crater_Traverse_Sol65_Color-A073R1_br2.jpg)

507k version (http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040317a/Bonneville_Crater_Traverse_Sol65_Color-A073R1.jpg)