By starting with the mission’s requirements, a computer-assisted design specialist draws the surfaces where the part connects to the instrument or spacecraft; then, the AI software connects the dots to produce complex structure designs in as little as an hour or two.
“The algorithms do need a human eye,” McClelland said. “Human intuition knows what looks right, but left to itself, the algorithm can sometimes make structures too thin,” he added.
The evolved structures can tolerate higher structural loads, weigh less, and can be produced in as little as a week. They also need less help from people, which gives designers more time to work on other parts of the mission.
“We found it lowers risk. After these stress analyses, we find [that] the parts generated by the algorithm don’t have the stress concentrations you have with human designs. The stress factors are almost ten times lower than parts produced by an expert human,” McClelland explained.
According to Ryan McClelland, who pioneered the design of these parts, these evolved structures can save up to two-thirds of the weight compared to standard components and reduce the risk of failure.