Scientists and engineers collaborating….

In developing science-driven space missions, scientists and engineers have the same goal—a successful mission—yet their relationship is almost adversarial by nature. Scientists want to maximize the science return (resulting in a complex spacecraft & mission) and engineers want the technical solution with the highest probability of success (typically a simple spacecraft & mission).

I’ve heard this specific relationship referred to as “positive tension.” It’s positive because the groups have opposing motivations and must compromise to find a technical solution that meets in the middle, achieving the scientific objectives within engineering and resource constraints. Doing so requires good communication, which can be challenging in any project, let alone one as interdisciplinary and complex as a space mission.

Despite being a science team member on NASA’s MESSENGER mission to Mercury, I had little exposure to space mission engineering. (I was only around for the last 6 months of operations.) The spacecraft was a magic black box that got me data.

This is not unusual: most planetary scientists’ backgrounds are in geology or astronomy/physics, neither of which involves a lick of engineering. We tend to be end users of mission data with little exposure to the processes and hardware that obtained it.

An artist’s impression of NASA’s MESSENGER spacecraft orbiting Mercury. MESSENGER was an incredibly successful Discovery-class mission (which means it was relatively low cost by NASA standards) that I worked on before leaving academia for industry. … — An artist’s impression of NASA’s *MESSENGER* spacecraft orbiting Mercury. *MESSENGER* was an incredibly successful Discovery-class mission (which means it was relatively low cost by NASA standards) that I worked on before leaving academia for industry. (NASA/JHU-APL/CIW)

Arriving at Planetary Resources after being on a NASA science team was a culture shock. I was the sole planetary scientist in a sea of engineers. Their jargon was a foreign language to me, and I was always asking for definitions and explanations. Once I accidentally triggered a long discussion about the differences between fasteners, screws, bolts, rivets, and nuts. The more you know….

But even after picking up the lingo, there was plenty of miscommunication because our motivations and contexts are different. For example, one engineer leading the design of our deployable asteroid penetrator wanted to know what asteroid surface compressive strength she should assume. I assumed she was asking for a precise number, laughed, and told her scientists have no idea; later discussions revealed that she was looking for a range of orders of magnitude (!), which is a totally tractable problem.

Conversely, I occasionally made the mistake of saying I wanted a particular instrument or operation, only to learn that my requests involved engineering assumptions. My science-driven requirements needed to be implementation-agnostic in order to maintain the positive tension between the science team who wants all the data and the engineering team who wants the simplest implementation.

This is a common mistake that I’ve seen planetary scientists make in discussing ideas for planetary science missions. They have a vision in their head of what the mission looks like and figure the engineers will make it work. However, we don’t necessarily have the background to understand what it takes to implement that vision.

Not that brainstorming doesn’t have value of course, but it’s important at the early stages of a designing a mission for scientists to keep an open mind about what the best engineering solution might be for the science objectives. If scientists are trying to do engineering, the “positive tension” that’s critical to mission success becomes unbalanced and will likely result in negative tension instead.

Not surprisingly, poor communication between scientists and engineers can be among the biggest challenges that mission teams face. An interesting paper by Susan Niebur (2010) surveyed 55 Discovery mission principal investigators (PI), project managers, and other leaders on these lower-cost NASA missions. On some missions, the scientists and engineers didn’t interact much; on others, they were more intertwined.

One PI who interacted heavily with engineers highlighted the value of explaining the mission’s scientific motivations to the engineers. Based on personal experience, I’ve found this to be profoundly critical, to the point that I’m baffled that some teams’ scientists and engineers didn’t interact much. Only when engineers understand why scientists want a particular measurement can we be confident that their design will adequately address it.

Lesson learned: those black boxes that get scientists data from space are not magic but rather the result of good communication, iterative design, and continual compromise.

Addendum - 2/24/19

I’m reading Seeing Like a Rover: How Robots, Teams, and Images Craft Knowledge of Mars by sociologist Janet Vertesi, who provides a unique perspective people and technology interact in a way that makes space missions work. Embedded on the Mars Exploration Rover science and operations teams, she noted the following observation, consistent with my previous writing on the sometimes tortured relationship between scientists and engineers:

The trade-off between scientific and driving observations throws into relief a relationship that Rover team members believe requires constant management: that between scientists and engineers. Stories abound on the team about fractured relations between the two camps on previous missions, and about the importance of paying attention to methods of communication between the two sides. Scientists, as characterized in these stories, always request too much without an understanding of what the spacecraft can and cannot do; engineers, for their part, are described as too protective of the vehicle, and as the saying goes, “would prefer to fly a spacecraft with no [scientific] instruments on it!”