AI is a wonder created by humans. Today, NASA is taking generative AI to space. 

The potential of ML to truly be integrated into scientific decision-making may be severely inhibited by transitional failures: key moments of breakdown in the evolution of complex AI methods and pipelines, which is also a contributing factor to AI’s reproducibility problem. 

The broad reach of AI and ML algorithms is evident in space examinations for navigating, routing the safest route for the spacecraft and more. The amount of data space agencies receive with the help of AI is beyond imagination. 

Therefore, the integrity of AI in science applications needs a quality control system that can form a foundation for effective handover. This requirement is reflected in NASA Science Mission Directorate’s (SMD) commitment to Open Science, which directs NASA Science to emphasize developing and implementing specific Open Science capabilities: continuous evolution of data and computing systems, better engagement of the scientific community, and, turn strategic partnerships for innovation, and provide data access to the wider research community for robust validation of published research results. 

Furthermore, scientists at NASA are using an AI-powered technique called “generative design” to speed up the process of designing hardware for upcoming missions. One of the first AI-designed missions will use a massive balloon to lift a telescope to the stratosphere. 

Image Source: NASA

What is generative design? 

The world is mediated by electronic systems and devices. The role of design is shifting in response to these changes. In this role, established design practices are often found to be inadequate or insufficient. Anthony Dunne sees design as relocating the electronic product beyond a culture of relentless innovation for its own sake, based simply on what is technologically possible and semiologically consumable to a broader context of critical things about its role in everyday life.  

The generative methodology offers an unconventional way of conceptualizing and working in design. Research in generative systems is closely tied to the general concept of synthesis, most viscerally apparent in nature and natural systems. 

Generative design to mission hardware 

NASA has used generative AI to design mission hardware, and it resembles bones left by alien species. However, the model weighs less, tolerates high structural loads, and requires a fraction of the time parts designed by humans take to develop. 

According to Engineer Ryan McClelland, they look somewhat alien and weird, but it makes sense once you see them in function. To create these Parrs, a computer-assisted design (CAD) specialist starts with the mission’s requirements and draws in the surfaces where the part connects to the instrument or spacecraft and any bolts and fittings for electronics and other hardware.  

Once all off-limits areas are defined, the AI connects the dots, producing complex structure designs in as little as an hour or two. These evolved structures save up to two-thirds of the weight compared to traditional components and can be milled by commercial vendors.  

Parts are also analyzed using NASA-standard validation software and processes to identify potential points of failure. McClelland’s evolved components have been adopted by NASA missions in different stages of design and construction, including astrophysics balloon observatories, Earth-atmosphere scanners, planetary instruments, space weather monitors, space telescope and even the Mars Sample Return mission.  

NASA’s custom parts 

AI-assisted design is a growing industry, with everything from equipment parts to entire car and motorcycle chassis being developed by computers. According to McClelland, 3D printing with resins and metals will unlock the future of AI-assisted design, enabling larger components such as structural trusses, complex systems that move or unfold, or advanced precision optics.  

These techniques could enable NASA and commercial partners to build larger components in orbit that would not otherwise fit in a standard launch vehicle; they could even facilitate construction on the Moon or Mars using materials found in those locations. 

NASA

Design Research Society