BMW’s Virtual Factory Uses AI to Hone the Assembly Line

0
26

German carmaker BMW plans to start out making drivetrains for electrical autos at an unlimited manufacturing facility in Regensburg, Bavaria, later in 2021. Nicely earlier than any new elements roll off the manufacturing line, the complete manufacturing course of will run in stunningly reasonable element inside a digital model of the manufacturing facility.

The simulation permits managers to plan the manufacturing course of in larger element than was beforehand attainable, says Markus Grüneisl, who leads manufacturing technique at BMW. “We now have an ideal digital twin of our real-time manufacturing,” he says.

The simulation is a part of BMW’s plan to make use of extra synthetic intelligence in manufacturing. Grüneisl says machine-learning algorithms can simulate robots performing advanced maneuvers to seek out essentially the most environment friendly course of. Over time, BMW desires to make use of the simulation to have robots learn to carry out more and more advanced jobs.

BMW used a software program platform known as Omniverse, developed by the chipmaker Nvidia, to recreate the Regensburg manufacturing line. Final 12 months, BMW stated it was utilizing an AI platform from Nvidia known as Isaac to coach robots for sure new duties.

“Sooner or later, I’m very optimistic that we will simply put a brand new robotic on this facility, and say, ‘OK, discuss with the opposite robots and discover the easiest way to supply this physique,’” Grüneisl says.

Producers have used laptop simulations to hone their meeting traces for a while. However Omniverse permits the complete manufacturing course of to be simulated with photo-realistic particulars, and with bodily properties like gravity and totally different supplies. It’s attainable to put out the manufacturing course of from begin to end, and to see how modifications to at least one half may need knock-on results on one other. It’s simpler to construct a extra advanced digital setting as a result of totally different 3D fashions may be imported into the system. Omniverse makes use of an open file normal appropriate with quite a few computer-aided design packages.

The software program can even simulate avatars of human employees grabbing elements and instruments, and assembling elements, to seek out one of the best process and reduce ergonomic issues. It may additionally make it attainable for fewer employees to finish a specific job, Grüneisl says.

“We do AI simulation of how individuals transfer across the manufacturing facility,” says Richard Kerris, normal supervisor for Omniverse at Nvidia. He calls the venture “one of the crucial advanced simulations that is ever been finished.”

There’s rising curiosity in utilizing AI to regulate robots and different industrial machines. Inspired by latest progress in AI, some startups are targeted on having robots be taught in simulation how you can carry out fiendishly tough duties like greedy irregular objects, expertise that would ultimately assist automate a lot ecommerce and logistics work. This usually makes use of an AI method known as reinforcement studying, which includes an algorithm experimenting and studying, from optimistic suggestions, how you can obtain a selected objective.

“That is undoubtedly the way in which to go,” says Ding Zhao, a professor at Carnegie Mellon College who focuses on AI and digital simulations. Zhao says simulations are essential to utilizing AI for industrial purposes, partly as a result of it’s unimaginable to run machines by hundreds of thousands of cycles to assemble coaching knowledge. As well as, he says, it’s essential for machine-learning fashions to be taught by experimenting with unsafe conditions, corresponding to two robots colliding, which can’t be finished with actual {hardware}. “Machine studying is data-hungry, and amassing it in the actual world is dear and dangerous,” he says.

Willy Shih, a professor at Harvard Enterprise College who focuses on manufacturing expertise, says the sophistication of simulation has been rising steadily, and he says simulation primarily saves money and time by preempting future manufacturing issues.

Shih says there may be loads of hype round AI for manufacturing however provides, “There are tons and plenty of purposes” for the expertise too.

Nvidia CEO Jensen Huang mentioned BMW’s use of Omniverse throughout his keynote on the firm’s annual GTC convention, held nearly on Monday. Nvidia initially made graphics chips for gaming however broadened its focus when these chips proved adept at coaching AI packages. The corporate has since leapt into a number of different industries the place AI is essential, together with automotive and medical imaging.

Leave a reply