Proving the Business Case for the Internet of Things

Nvidia improves BMW logistics with AI robots

Steve Rogerson
May 20, 2020

Car maker BMW has selected Nvidia’s Isaac robotics platform for its automotive factories. The logistics robots were built using AI computing and visualisation technologies.
The collaboration centres on implementing an end-to-end system based on Californian electronics company Nvidia’s technologies from training and testing to deployment, with robots developed using one software architecture, running on Nvidia’s open Isaac robotics platform.
BMW’s objective is to enhance logistics factory flow to produce custom-configured cars more rapidly and efficiently. Once developed, the system will be deployed to BMW factories worldwide.
“BMW Group’s use of Nvidia’s Isaac robotics platform to reimagine their factory is revolutionary,” said Jensen Huang, founder and CEO of Nvidia. “BMW Group is leading the way to the era of robotic factories, harnessing breakthroughs in AI and robotics technologies to create the next level of highly customisable, just-in-time, just-in-sequence manufacturing.”
The collaboration uses Nvidia’s DGX AI systems and Isaac simulation technology to train and test the robots and Quadro ray-tracing GPUs to render synthetic machine parts to enhance the training. Also included are multiple AI-enabled robots built on the Isaac software development kit (SDK), powered by Nvidia Jetson and EGX edge computers.
“BMW is committed to the power of choice for our customers – customisation of diverse features across diverse vehicles for diverse customers,” said Jürgen Maidl, senior vice president at BMW. “Manufacturing high-quality, highly customised cars, on multiple models, with higher volume, on one factory line requires advanced computing from end to end. Our collaboration with Nvidia allows us to develop the future of factory logistics today and to ultimately delight BMW Group customers worldwide.”
BMW’s supply chain takes millions of parts flowing into a factory from more than 4500 supplier sites, involving 230,000 unique part numbers, and in growing volumes as BMW’s car sales have doubled over the past ten years to 2.5 million vehicles. Moreover, BMW vehicles are offered to customers with an average of 100 different options, resulting in 99 per cent of customer orders being uniquely different for each other. This creates an immense challenge for factory logistics.
To optimise the complexity of this material flow, autonomous AI-powered logistics robots now assist the production process to assemble customised vehicles on the same production line.
“Ultimately, the sheer volume of possible configurations became a challenge to BMW Group production in three fundamental areas – computing, logistics planning and data analysis,” Maidl said.
BMW’s response is to use the Isaac robotics platform to develop five AI-enabled robots to improve their logistics workflow, powered by a variety of edge computers. These include navigation robots to transport material autonomously as well as manipulation robots to select and organise parts.
Developed on the Nvidia Isaac SDK, the robots use a number of powerful deep neural networks, addressing perception, segmentation, pose estimation and human pose estimation to perceive their environment, detect objects, navigate autonomously and move objects. These robots are trained on real and synthetic data using Nvidia GPUs to render ray-traced machine parts in various lighting and occlusion conditions to augment real data.
The real and synthetic data are then used to train deep neural networks on Nvidia DGX systems. The robots are then continuously tested in Isaac simulators for navigation and manipulation, operating on Nvidia’s Omniverse platform, where multiple BMW personnel in different geographies can all work in one simulated environment.