After a month of spinning their wheels in a sandbox, NASA engineers have settled on a strategy to free the rover Spirit from a sand trap on Mars: just wing it.
The geriatric rover slid into soft terrain in early May, burying its wheels halfway into the flour-like soil. To test strategies for getting out without risking a catastrophic misstep, engineers rolled a duplicate rover into a sandbox at NASAās Jet Propulsion Lab in Pasadena, California, that mimics the Martian morass.
After testing dozens of individual moves, from simply rolling forwards and backwards to ācrab-walkingā sideways, engineers decided to throw the kitchen sink at the problem.
Advertisement
āEach manoeuvre shows progress up to a point, and then by changing the manoeuvre we can continue the progress,ā says project manager John Callas. āThat right now looks like the emerging strategy ā changing the manoeuvres each time to continue to make progress.ā
The team will hash out the final choreography in a meeting on Thursday, and will begin moving the rover on Monday, Callas says.
Extra challenges
Spiritās twin, Opportunity, had a similar dilemma in 2005, when its wheels sunk into a sand dune scientists later named āPurgatoryā. It finally escaped by slowly backing out until its wheels found traction again.
But Spirit faces extra challenges. Its right-front wheel stopped working three years ago, and since then scientists have been driving backwards using the five remaining wheels, dragging the crippled wheel along behind it.
It has also been plagued by rocks. A rock that is poking the rover in the belly might keep the wheels from getting traction if the rock is bearing any of the roverās weight, though a similarly placed rock has not been an issue for the test rover in the JPL sandbox.
āItās a complication, but not a serious complication,ā Callas told New ŠÓ°ÉŌ““. The rock seems to be just grazing the roverās undercarriage, not holding it up. The rock also appears to be āfloatingā on the planetās surface, so it would sink into the sand if Spirit put any pressure on it, says Callas.
ŠÓ°ÉŌ““s also worried that Spiritās left-centre wheel had a rock stuck in it because early on the wheel had trouble spinning. But the wheel rotated fully in later tests. āWe donāt know if whatever caused the stall is something that could recur,ā Callas says. āItās possible. The first movement out of the gate could be a stall.ā
āThin iceā
To make matters worse, Spirit is also on a steeper slope than Opportunity was, and an even deeper sandpit lies directly downhill.
āIt took us almost a month to get out of Purgatory dune, and that was with a six-wheeled rover that didnāt have the sloping problems,ā Callas says. āThis is a highly complex problem with a compromised rover.ā
The type of soil Spirit is in also presents a problem. The region is covered with a ācamouflaging crust.ā Spirit drove safely over the crust for an unknown amount of time before breaking through and embedding its wheels in the fine, loose soil underneath. ŠÓ°ÉŌ““s donāt know how far the crust extends in any direction, including the direction from which Spirit came.
āThe region we were safely over before could be a challenge now,ā Callas says. āItās like skating on thin ice.ā
Time delay
So instead of backing up, scientists plan to tell Spirit to push forward. The first move will probably be straightening the wheels, and then a small step straight ahead.
After that, the plan is murky. If all goes well, mission managers may order the rover to take another step; if not, theyāll turn all four corner wheels to the right and crab-step uphill, with the wheels pointed 20Ā° to the right.
Other options include hard turns with the wheels angled 60Ā° degrees to the right and backwards drives if necessary. The rover will hopefully tiptoe forward, stopping after each step to assess its progress.
One reason for the slow progress is that the amount of time it takes for messages to get between Earth and Mars limits engineers to one driving command per day. If Spirit attempted too complicated a manoeuvre and got buried even deeper, it wouldnāt get a chance to fix the problem until the next day. The safest move is therefore baby steps.
āWeāll take small steps to begin with, and weāre expecting small results,ā Callas says. āOur expectation is to move forward, but weāre doing it cautiously.ā
Different worlds
Despite all the tests, engineers canāt plan a more specific series of moves. This is partially thanks to the inherent differences between Earth and Mars. Martian gravity is only three-eighths of Earthās, and it has a thin atmosphere with no moisture. Earth-bound tests are āonly an approximation,ā Callas says. āNo one can tell you how the roverās going to respond on Mars until we start moving.ā
āThe public has been offering lots of suggestions to us, but theyāre all based on Earth analogues, people getting stuck in the mud or the snow,ā he adds. āThose are not applicable ā thereās no mud and thereās no snow on Mars. Theyāre two different worlds.ā
The benefit of the sandbox tests was to build up an arsenal of movements that could help extricate the rover, Callas says. And heās guardedly optimistic. āItās going to take us a long time to get out,ā he told New ŠÓ°ÉŌ““. āBut itās encouraging that the rover in the sandbox is climbing; it has enough capability to fight Earth gravity to get altitude. Weāre expecting that on Mars, where thereās only three-eighths gravity, that the rover may actually do better.ā
