Beware of the Real Galactic Ghoul

I studied the faces of the flight team, anxious and intent, as they waited for signal-lock. Their spacecraft, the Mars Polar Lander, had plunged into the Martian atmosphere 23 minutes earlier, and the Deep Space Network antennas were straining to pick up the spacecraft’s signal. I could empathize with the controllers because, two years earlier, I was in their position as a flight software engineer for the Mars Pathfinder spacecraft. So, when the Mars Polar Lander failed to respond, they were living my nightmare.

With the loss of the Mars Polar Lander and its sister craft, the Mars Climate Orbiter, it appears that NASA has spent its entire 1999 Mars budget to no avail. This has led to ill-informed speculation and calls for the end of the “better, faster, cheaper” approach to space exploration. As a former insider, I’d like to provide insight into what happened, and urge our leaders to maintain the “better, faster, cheaper” approach now under fire.

Despite publicity to the contrary, the 1999 Mars missions violated the primary tenets of the better, faster, cheaper missions and paid the price. The essence of this approach is: Handpick a small, talented team that works on the project from cradle to grave and locate that team at one site for face-to-face communication. Encourage an innovative culture and regard everyone as a systems engineer.

The Jet Propulsion Laboratory, which built Pathfinder, did not construct these two spacecraft. Both spacecraft were outsourced to a large aerospace contractor in Denver. With one team in Denver and another in Pasadena, there was little chance for the face-to-face contact required to successfully engineer a complex interplanetary spacecraft.

The problem was further compounded by the fact that JPL, although stellar at building spacecraft, is poor at pushing paper (managing contracts). NASA’s Climate Orbiter failure report stated that in both projects, there “is evidence of inadequate communications between the project elements, including the development and operations teams, the operations navigation and operations teams, the project management and technical teams, and the project and technical line management.” Translation: Nobody was communicating.

The people who flew the spacecraft did not build it. A new group was brought in to operate the spacecraft after construction was completed. The new team was unfamiliar with the spacecraft’s design, and in the case of the Climate Orbiter, this led directly to catastrophic failure. The orbiter calculated, and transmitted to Earth, the velocity change data that was being computed erroneously on the ground. Had the operations team understood the spacecraft, it might have uncovered the units problem that ultimately led to the spacecraft’s loss. On Pathfinder, the team who built the spacecraft tested and flew it. They had an intimate knowledge of their child and understood all of its possible misbehaviors.

Not only did the 1999 operations team have to learn about the spacecraft, it had to learn about three spacecraft. The operations organization was simultaneously flying the Mars Global Surveyor, Mars Climate Orbiter (destroyed), and the Mars Polar Lander (destroyed). Controlling one spacecraft is a dangerous and unforgiving task, controlling three is an impending disaster. Besides diluting the focus on any one mission, overworking the operations team prevents them from planning the analysis, tests and procedures required for a safe journey to Mars.

Finally, there are questions about the management of the undertaking. Anyone who works on a complex project knows that success requires more than just technical prowess. An innovative environment is vital and at the heart of the “better, faster, cheaper” approach. A remotely located defense contractor is a tough place to foster such an environment.

The Climate Orbiter failure report cited a lack of systematic analyses of what could go wrong for both missions. This is surprising since moderate paranoia is the spacecraft designer’s friend. It raises questions about how much proprietorship the engineers felt toward their spacecraft.

The 1999 Mars missions were simply examples of poor execution. Pathfinder proved that “better, faster, cheaper” is not a pipe dream. Space exploration is far too dangerous for the Pathfinder project to simply have been lucky. When flying the Pathfinder spacecraft, the team would joke about the Galactic Ghoul--a fierce creature who lurks within the orbit of Mars devouring robotic spacecraft. But with these two missions, the real culprit is an old mind-set, which will have to be transcended if we expect our descendants to ever walk on Mars.