Location Is Everything
Summary (Jan 09, 2004): To tie down Spirit's location, three mission teams have agreed to within a quarter-mile. But before driving their rover around, the teams want to agree precisely on terrain and features in an otherwise flat landscape.
Display Options: 
Location Is Everything
Pasadena, Spirit Mission Sol 6
By: Henry Bortman Within 24 hours after Pathfinder landed on Mars in 1997, NASA scientists had pinpointed its landing site.
Spirit's story is a bit different. Spirit landed six days ago, but scientists are still struggling to figure out exactly where.
 |
Flat landform gives few clues from orbit. Click for larger view.
Credit: NASA/JPL |
According to
Matt Golombek, who is leading the effort to tie down Spirit's location, different groups of scientists working on the problem agree to within about 500 meters (about a quarter of a mile). But, says Golombek, "that's really not good enough. I want to get it down to 50 to 100 meters" before Spirit starts driving across the landscape.
One problem is that the Spirit landing site is flat, flat, flat. That's a good thing for landing, and a great thing for driving the rover long distances. But there's a downside: The science team doesn't have many identifiable landmarks to help them figure out exactly where the rover is.
 |
Five crater pattern shown by arrow seen in descent images and from orbit.
Credit: NASA/MOC |
What they do have are four distinct sets of information, which they are working to correlate.
The first is a set of 3 DIMES (Descent Image Motion Estimation System) images, taken during Spirit's descent through the Martian atmosphere. The DIMES images are pretty fuzzy, but they clearly show some of the most prominent features of the landscape, notably a set of 3 closely spaced craters.
By comparing the DIMES images with archival images taken by the Mars Orbital Camera (MOC), scientists were quickly able to figure out where Spirit first hit the ground. The MOC images constitute the second set of data being used to figure out where Spirit is. MOC is one of the instruments onboard the Mars Global Surveyor (MGS) orbiter. One of its tasks has been to build up a database of photographs that cover the entire Martian globe.
Knowing where Spirit first slammed into Mars is handy, but the spacecraft didn't land just once. It bounced, and bounced some more, and then rolled a bit - traveling perhaps a kilometer (half a mile) or more before coming to a halt. Which way it bounced and how far are questions no-one has precise answers to.
That's where the third set of data, known as inertial data, comes in. Inertial data provides a way of locating Spirit in a vast imaginary 3-dimensional space whose center is at the center of the Earth. Scientists can figure out where Spirit is in "inertial space" quite precisely. Every time Spirit communicates by radio, whether it be directly to Earth or to one of the spacecraft orbiting Mars, they can track the location in inertial space from which it is broadcasting.
 |
Complex descent steps, used to build inertial map. Click to enlarge.
Credit:JPL |
The problem is that no-one knows exactly how to correlate inertial space, in which Spirit can be located very precisely, with the MOC images of the landing site. There's simply no data available that matches up the two sets of information.
"You don't have GPS stations sitting on the surface [of Mars] looking at the stars," as you do on Earth, says Golombek. "You only have one place where you know for sure where it is on [MOC images of] Mars, and that's the Pathfinder landing site."
So now the problem comes full circle. The fourth set of information is visual clues from the surface. And they're slim pickings.
 |
| (Click image to view details) Gusev hills Credit: JPL/NASA |
What made it easy to find Pathfinder was the presence of large, prominent and distinctly shaped landforms at its landing site. Even though MOC images were'nt available - MOC hadn't yet been sent to Mars, so Viking images were used to locate Pathfinder - it was easy to match up what scientists saw in the pictures taken from orbit with the features Pathfinder saw on the ground.
Not so at the Spirit site. There are only a handful of features that rise above the vast plain of the Gusev landing site. Moreover, they're not that big, and they're not that distinctive. Even the highest resolution images from MOC don't show every little bump and dip on the Martian surface. So it's hard to tell whether the hills to the east of the Spirit lander is this tiny dark spot on a MOC image, or that one.
You might wonder what all the fuss is about. Does it really make all that much difference whether Spirit is half a kilometer this way or that? Or is it just a matter of pride?
It matters. The MER scientists need to know where Spirit is so they can figure out where to send it. Even when it spends an entire day driving, the rover can cover only about 50 meters. So if the science team wants to send it on a long drive, they want to know whether their target is a kilometer away or a kilometer and a half. The extra half a kilometer means an extra 10 days of driving. That's more than 10 percent of the total mission.
So when will they know? The answer to that question isn't all that precise, either. "Soon," is all anyone will commit to.
Related Web Pages
Rover Science Team Eyes Sleepy Hollow There's History in Them Thar HillsPancam: Surveying the Scene - Martian StyleInterview with Nathalie Cabrol - Pasadena, Spirit Mission Sol 4Cabrol Presentation, Testable Hypotheses at GusevNote:
Mapping
Display Options:
Friday, January 09, 2004
