A Self-Driving Lab

An autonomous robot in BU’s College of Engineering is teaching itself to create the world’s most shock-absorbent shape. Meet MAMA BEAR. If it succeeds, the product could revolutionize helmets, packaging, car bumpers, and more.

MAMA BEAR is short for Mechanics of Additively Manufactured Architectures Bayesian Experimental Autonomous Researcher. For the past three years, MAMA BEAR has been repeatedly creating and crushing 3D-printed plastic structures and recording resulting details in a vast database.

First conceptualized in 2018 by Keith Brown, an associate professor of mechanical engineering, and his team in the KABlab, MAMA BEAR has been running continuously since 2021. Learning as it goes, the robot creates a structure, compresses it, and measures how much energy it absorbs and how its shape changes—these determine what’s known as its mechanical energy-absorption efficiency. Each iteration is a tweaked version of its predecessor, the design and dimensions adjusted by the robot’s computer algorithm based on all past experiments. To date, this “self-driving lab” has filled dozens of boxes with more than 25,000 shapes.

It’s all about finding the sweet spot. The ideal structure can’t be so strong that it damages whatever it’s supposed to protect, but it should be strong enough to absorb impact.

Before MAMA BEAR, the best structure anyone observed was about 71% efficient at absorbing energy, says Brown. But one January afternoon in 2023, Brown and his team saw their robot hit 75% efficiency, breaking the known record. The results were published in Nature Communications in May 2024 and have already attracted the attention of the US military as it seeks to design new helmet padding for soldiers.

“We’re excited that there’s so much mechanical data here and we’re using this to learn lessons about design more generally,” Brown says.