Europe’s largest R&D organisation Fraunhofer Society is focusing on applied research of vital technologies ranging from AI to cybersecurity to medicine, NVIDIA announced.
In a statement, NVIDIA said the Fraunhofer Society’s Fraunhofer IML unit aims to push the limits of logistical and robotics. The German researchers, according to NVIDIA, are using NVIDIA Isaac Sim to develop advances in robot design through simulation.
Like companies such as BMW, Amazon, and Siemens, Fraunhofer IML relies on NVIDIA Omniverse, utilising it to advance applied research in logistics for fulfilment and manufacturing.
According to NVIDIA, Fraunhofer’s newest innovation, O3dyn, employs NVIDIA simulation and robotics technologies to create an indoor-outdoor autonomous mobile robot (AMR). Its purpose is to ease the transition from automated guided vehicles to fast-moving AMRs that aren’t yet on the market. This level of automation progress offers a significant increase in logistical acceleration.
“We are looking at how we can go as fast and as safely as possible in logistics scenarios,” Fraunhofer IML robotics and AI researcher Julian Eber said.
In 2013, Fraunhofer launched a groundbreaking robot known as STR, now widely used by BMW in its assembly plants and others. STR is a workhorse for industrial manufacturing employed to transport goods for production lines. According to NVIDIA, the STR and other improvements to the robotics platform, including the O3dyn, benefit from Fraunhofer IML’s AI work.
Fraunhofer IML’s goal is to develop AMRs representing a new state-of-the-art state. The O3dyn navigates using a variety of camera and sensor inputs from the NVIDIA Jetson edge AI and robotics platform.
It can travel up to 30 miles per hour and has wheels that can move in any direction to operate in confined spaces, advancing speed and agility.
“The omnidirectional dynamics is very unique and there’s nothing like this that we know of in the market,” Fraunhofer IML AI and autonomous systems head Sören Kerner said.
NVIDIA stated that Fraunhofer IML’s latest study aims to build and validate these AMRs in simulation by closing the sim-to-real gap using Isaac Sim. By exercising the robot in photorealistic, physically correct 3D settings, the researchers rely on Isaac Sim for virtual development of its highly dynamic autonomous mobile robot.
This allows Fraunhofer to load its robot’s more than 5,400 parts from computer-aided design software into the virtual world. It can then use Omniverse PhysX to rig them to physically precise specifications.
As a result, NVIDIA said, the virtual robot version can move as quickly in simulation as the genuine robot can in real life. Using the virtual environment enables Fraunhofer to accelerate development, boost accuracy for real-world deployment, and scale up more quickly.
According to NVIDIA, Fraunhofer is releasing the AMR simulation model into open source so developers can enhance it to achieve results more quickly.
“This is important for the future of logistics. We want to have as many people as possible work on the localisation, navigation, and AI of these kinds of dynamic robots in simulation,” Kerner stated.
