Delivery of NASA's large Curiosity rover to a Martian crater next Aug. 6 will be a do-or-die test, not just of its “sky crane” landing system, but of an equally new approach to targeting distant bodies with unprecedented precision.

Curiosity will be lowered on high-tech cables from a hovering spacecraft—that looks a little like something out of a Star Wars movie—to the smallest landing zone ever. Whether it works or not, getting the rover into position for the touchdown also is stretching planetary entry and descent techniques in ways that will shape space exploration for decades.

“Overall we're fairly confident in this system, but going through the process we have recognized that there are limits to how far we can go with the mass of the landed system with this, and we have ways that we think we could develop improvements to allow even more landed mass,” says Steven W. Lee of the Jet Propulsion Laboratory, who is guidance, navigation and control manager on the MSL project. “It's the same with the precision landing.”

Targeted in the $2.5 billion Mars Science Laboratory (MSL) mission is a 20 X 25-km ellipse adjacent to a 5-km-tall (3-mi.)mound inside Gale Crater. Just south of the planet's equator, the crater measures 154 km across—smaller than the landing zones of the Mars Exploration Rovers launched in 2003.

On Dec. 10 controllers at the Jet Propulsion Laboratory will oversee the first in a series of trajectory correction maneuvers spread over the probe's 255-day cruise to Mars designed to deliver it to the proper entry corridor at the outer edge of the planet's thin atmosphere. The MSL lifted off from Cape Canaveral AFS, Fla., at 10:02 a.m. EST Nov. 26 atop a United Launch Alliance Atlas V-541, with four solid-fuel boosters and a Centaur upper stage. The Centaur's second burn, lasting almost 8 min., sent the spacecraft out of Earth orbit toward Mars.

“The launch vehicle has given us a great injection into our trajectory, and we're on our way to Mars,” states MSL Project Manager Peter Theisinger. “The spacecraft is in communication, thermally stable and power-positive.”

Initially the probe was targeted away from Mars slightly to keep the Centaur stage from following it to the planet's surface and perhaps contaminating it with terrestrial microbes. With the precise launch, the first correction maneuver could come even later than Dec. 10, Lee says, because it will only need a small 3-meter/sec. “nudge” to turn it back on course.

From there on, the cruise phase of the mission will look a lot like other Mars-landing transits—checking out the suite of instruments and making additional trajectory adjustments as the spacecraft approaches its targeted entry corridor. The differences will come at the end of the trip, when the spin-stabilized cylindrical cruise stage feeds a final position fix from its Sun sensor and star trackers to the entry capsule and then separates before reaching the atmosphere.

JPL engineers drew on expertise at Johnson Space Center and elsewhere dating back to the Apollo era to devise what will come next. After the cruise stage jettisons, springs will kick two 73-kg (161-lb.) tungsten weights away from the entry element to shift its center of mass off the aeroshell's centerline. That will allow the spacecraft's entry, descent and landing (EDL) algorithms literally to fly the capsule toward the floor of the crater.

“The airflow goes faster over the top of the aeroshell relative to the bottom of the aeroshell; that gives you a lift vector,” Lee says. “We can use that lift vector not only to keep ourselves a little higher in the atmosphere, but we can maneuver that lift vector to control the energy of entry to try to control our down range relative to the center of the landing ellipse, and also use it to control cross-track.”

A Honeywell inertial measurement unit keeps track of the capsule's position and velocity as it hurtles through the Martian atmosphere, flying “S” turns like a reentering space shuttle to bleed velocity and home in on the landing ellipse. When the capsule has slowed almost to Mach 2, the vehicle will jettison more tungsten ballast, broken into six 28-kg spring-ejected pieces for a gentler shift of the center of mass back to the center of the vehicle. That way it will be falling straight ahead when it pops its 16-meter-dia. parachute at Mach 2.

After the chute opens 11 km above the surface, plus or minus 2 km depending on landing-day conditions, the heat shield falls away, exposing the Curiosity rover and its eight-engine descent stage. A six-beam Ka-band radar—the “terminal descent sensor”—is activated at 8 km, again plus or minus 2 km, giving the EDL algorithm the information it needs on distance to the ground and rate of descent.

“The terminal descent sensor has been another lynchpin in enabling this kind of a mission, mainly because we're such a large vehicle and we need to commit to our maneuvers using the radar data relatively early in the timeline,” Lee says. “As a result, we need to have a radar that can work at higher altitudes than past radars.”

When the parachute and backshell are released at 1,800 meters above the surface, the descent stage engines begin firing to slow the descent further. Firing asymmetrically at first, they maneuver the stage away from the chute and start the final trip to the surface. The 899-kg rover is lowered toward the surface on three cables and a data umbilical in the sky crane technique (AW&ST Aug. 1, p. 38).

If all goes as planned, Curiosity sets down on its six wheels—sparing the need for more landing gear—to begin exploring the crater's sedimentary mound for evidence of past habitability.

The site has been characterized by the Mars Reconnaissance Orbiter's High-Resolution Imaging Science Experiment (HiRISE), which gives mission managers confidence that the spacecraft won't come down on a large boulder or teetering on a low cliff. That is necessary, because while the final landing sequence is complex, it does not involve active hazard avoidance to maneuver the descending spacecraft away from danger.

“Our landing site experts here could actually show you the landing ellipse and point to a little scarp there, a little mesa here,” says Lee. “I don't think there's a scarp or mesa that we don't know about, or worry that we haven't characterized the site. It's a very small—less than 1 percent, certainly, and probably much less than 1 percent—chance that we just happen to come down right in that bad spot. Those are risks that, when you're going to Mars, you just have to accept.”

Humans-in-the-loop have been actively avoiding terrain hazards while landing since Neil Armstrong piloted the Eagle to its touchdown in the Sea of Tranquility on Apollo 11. Lee says JPL is already researching active hazard avoidance all the way down for robotic missions in rougher terrain than the relatively flat target ellipse in Gale Crater.

In developing the MSL mission, Lee and his colleagues used extensive simulation and “millions” of Monte Carlo risk-reduction runs of the approach and landing, raising confidence. And despite the complexity of the EDL sequence, driven by the size of the rover and the tight landing site, there was an ongoing effort to keep it as simple as possible.

“Part of our culture has been, when we think through a problem and think through options, we always want to be able to go back to a whiteboard and start with first principles,” Lee says.