The BMW virtual factory uses AI to perfect the assembly line

German car manufacturer BMW intends to start producing transmissions for electric vehicles at a large factory in Regensburg, Bavaria, later in 2021. Long before any new part comes off the production line, the entire manufacturing process will run into amazingly realistic details in -a virtual version of the factory.

The simulation allows managers to plan the production process in more detail than was previously possible, says Markus Grüeneisl, who leads BMW’s production strategy. “We now have a perfect digital twin of our production in real time,” he says.

The simulation is part of BMW’s plan to use more artificial intelligence in production. Grüeneisl says that machine learning algorithms can simulate robots performing complex maneuvers to find the most efficient process. Over time, BMW wants to use simulation for robots to learn how to perform increasingly complex tasks.

BMW used a software platform called Omniverse, developed by chip maker Nvidia, to recreate the Regensburg production line. Last year, BMW said it was using an Nvidia AI platform called Isaac to train robots for certain new tasks.

“In the future, I’m very sure we can put a new robot in this facility and say, ‘OK, talk to the other robots and find the best way to produce this body,'” says Grüeneisl.

Manufacturers have been using computer simulations to perfect their assembly lines for some time. But Omniverse allows the simulation of the entire production process with photo-realistic details and physical properties such as gravity and various materials. It is possible to determine the production process from start to finish and see how changes to one part could have an impact on the other. It is easier to build a more complex virtual environment because different 3D models can be imported into the system. Omniverse uses a standard open file that is compatible with many computer-aided design packages.

The software will also simulate the avatars of human workers grabbing parts and tools and assembling components to find the best procedure and minimize ergonomic issues. It may also be possible for fewer workers to complete a certain job, says Grüeneisl.

“We do AI simulations of how people move around the factory,” says Richard Kerris, CEO of Omniverse at Nvidia. He calls the project “one of the most complex simulations ever made.”

There is a growing interest in using AI to control robots and other industrial machines. Encouraged by recent advances in AI, some startups focus on robots learning in simulation how to perform extremely difficult tasks, such as grabbing irregular objects, a technology that could eventually help automate many business activities. electronics and logistics. It often uses an AI approach called reinforcement learning, which involves an algorithm that experiments and learns, from positive feedback, how to achieve a specific goal.

“This is definitely the way to go,” says Ding Zhao, a professor at Carnegie Mellon University who focuses on AI and digital simulations. Zhao says simulations are crucial to using AI for industrial applications, in part because it’s impossible to run cars through millions of cycles to collect training data. In addition, he says, it is important for machine learning models to learn by experiencing insecure situations, such as two colliding robots, which cannot be done with real hardware. “Machine learning is starving for data, and collecting it in the real world is expensive and risky,” he says.

Willy Shih, a professor at Harvard Business School who specializes in manufacturing technology, says the sophistication of simulation has grown steadily and says that simulation saves time and money primarily by preventing future manufacturing problems.

Shih says there is a lot of hype around AI for manufacturing, but adds, “There are a lot of applications” for technology as well.

Nvidia CEO Jensen Huang discussed BMW’s use of the Omniverse at its annual GTC conference, which took place on Monday. Nvidia initially made graphics chips for games, but expanded its focus when these chips proved to be capable of training AI programs. Since then, the company has jumped into several other industries where AI is important, including automotive and medical imaging.

Share this:

Related