NASA JPL
Designing an interface for scientists to program autonomous space vehicles
Human Centered Design Team
Highlights
Skills
Working with NASA Scientists
Deconstructing mission planning
Collaboration with NASA interface designers
Building Human - AI interfaces
User testing
Data Visualization
AI Systems
Rover science instruments
Mars rover control interfaces
Figma wire-framing
Technology
The Challenge
I was brought on as a designer at NASA to envision a first of its kind interface that would allow research scientists to program autonomous space vehicles.
Due to distance and environmental conditions, rovers require sophisticated autonomous systems. Our efforts were aimed at bridging the gaps between autonomous driving teams and mars research scientists.
Learning to see like a rover
Embarking on autonomy UX design required a monumental amount of research. Our team had to learn how to see like a rover, act like an operations organizer, be curious like a scientist, and reason like an engineer.
Design methodology
Navigating complicated operational and technical AI workflows
In order to make sound design systems, we had to understand the past, present, and future of rover interfaces. This required numerous conversations with cutting edge scientists and developers across NASA.
Fluency in active rover missions, instrumentation, operations workflows, and user challenges were essential to designing our interface.
Our team worked with individuals across NASA how contributed their research, expertise, and perspective to our final product.
Balancing Act
Most challenging was determining how to arrive at specific autonomous AI directions from the input of scientists that new little about the inner workings of rover autonomy.
We had to learn how to balance complexity and the interpretability of our product in a real setting.
Scientists needed to want this tool and so did autonomy engineers.
Prototyping
We filled notebooks, whiteboards, and Figma pages with prototypes as we iterated to our final project. Each visualization, decision, and button needed to be intentional and stand up to rigorous user testing sessions
Iterations
The interface went through three big iterations before we reached a final product. Each one brought new insights and findings to the NASA community.
Communication
Communication was key in relaying the ideas of our prototypes. Their were many layers of complexity to this project, so the ability to both visually and verbally express interface complexity became a major takeaway from this project.
Design Outcomes
Bridging the gap between scientists and engineers
You can see glimpses into our final interface in the gifs below. With this interface, scientists could program intent for the first time. This was accomplished through specifying a number of fundamental interests and instrument procedures in our interface.
Lessons Learned
Create a first of its kind interfaces that would allow research scientists to indicate their research preferences for an autonomous rover.
Further Questions
If you’d like to learn more about this project, please feel free to reach out. Research into rover autonomy interfaces is ongoing and
