A few months ago, NASA unveiled its next generation space suit that will be worn by astronauts when they return to the moon in 2024 as part of the agency’s plan to establish a permanent human presence on the lunar surface. The extravehicular mobility unit – or xEMU – is NASA’s first major upgrade to its space suit in nearly 40 years and is designed to make life easier for astronauts who will spend a lot of time raising moon dust. This will allow them to bend and stretch in ways they couldn’t before, to easily put on and take off the suit, to replace components for a better fit and to spend months without making repairs.
But the biggest improvements weren̵
Jesse Craft is a senior design engineer at Jacobs, a major Dallas-based engineering firm that was chosen by NASA to renovate the xEMU life support system. For Craft and the hundreds of other engineers working on the project, this requires a careful balance between competing priorities. The life support system must be safe, of course, but it must also be light enough to adapt to the weight limits for the lunar lander and strong enough to withstand the intense g forces and vibrations it will experience while launching a rocket. “It’s a real engineering challenge,” says Craft.
Squeezing more stuff into less space with less mass is the type of complex optimization problem that aerospace engineers continually face. But NASA wants boots on the moon by 2024, and compliance with that aggressive timeline meant that Craft and his colleagues could not spend weeks discussing the ideal shape of each widget. Instead, they are piloting new AI-based design software that can quickly develop new component designs.
“We believe AI is a technology that can do something faster and better than a trained human being can do,” says Jesse Coors-Blankenship, vice president of technology at PTC, the American company that made the software. “Some of the software technologies are already known to engineers, such as structural simulation and optimization. But with AI, we can do it faster. ”This approach to engineering is known as generative design. The basic idea is to provide the software with a series of requirements for the maximum size of a component, the weight it has to bear or the temperatures to which it will be exposed and let the algorithms understand the rest.
PTC’s software combines several approaches to AI, such as generative contradictory networks and genetic algorithms. A contradictory generative network is a game-like approach in which two machine learning algorithms compete in a competition to design the most optimized component. It’s the same technique used to generate photos of people who don’t exist. Genetic algorithms, on the contrary, are analogous to natural selection. They generate more designs, combine them, then take the best designs of the new generation and repeat themselves. In the past, NASA has used genetic algorithms to design optimal and bizarre antennas.
“The iterative process of the machine is 100 or 1,000 times greater than what we could do on our own and offers an optimized solution ideally within our limits,” says Craft. It is particularly useful as the final design of the spacesuit’s life support system is still evolving. Even a small change in requirements in the future could result in weeks of wasted work by the engineers.