Highlights:

  • The latest Isaac Sim incorporates human characters into operational environments for performing regular activities on the floor.
  • Researchers can now access an extensive collection of simulation-ready 3D models.

The demand for intelligent and autonomous robots has been surging due to tedious supply chain operations. Also, the allied tools require advanced simulation and Artificial Intelligence (AI).

Nvidia Corp. announced at CES 2023 that its robotic simulation tool, Isaac Sim, has been updated with advanced features. It enables engineers and designers to build and test virtual robots in operational environments, such as assembly lines, warehouses, and urban streets to train their AI and execute it better.

The simulation technology helps companies to access Nvidia Omniverse, a state-of-the-art platform that collaborates 3D worlds and metaverse applications in the cloud. This facilitates easy collaboration between teams anywhere in the world with minimal downtime.

Deepu Talla, vice president of embedded and edge computing at Nvidia, stated, “Developing robots is a multidisciplinary endeavor. Mechanical engineers, electrical engineers, computer scientists, and AI engineers come together to build a robot. With Isaac Sim in the cloud, these teams can be located around the globe and still collaborate efficiently.”

The new update of Isaac Sim simulates human characters in manufacturing or warehouses to perform regular activities, such as pushing carts and stacking boxes. The programmers can monitor robots following their commands. Meanwhile, the robot AI can avoid people’s actions and their erroneous movements.

Hence, in case of situational hazards, people could be evacuated before clubbing them with robots. This indicates that robots can be reprogrammed to deal with human presence on the floor, restructure the warehouse or factory for better accommodation of workers and robots, and comply with safety features to prevent any mishap.

Nvidia RTX technology, an updated platform for ray tracing and AI is used by Isaac Sim to build large-scale models, thus, enabling designers to render physical accurate data of simulated environments. Besides, the technology includes RTX-simulated LiDAR, a laser emission sensor that assists robots to guide themselves. In the virtual world, the robots behave pretty much like their real environment action in response to any reflective materials or lighting conditions on the floor.

Researchers can access a huge library of latest simulation-ready 3D models to create physically precise simulated environments. It includes popular robots, warehouse parts, and already built large-scale warehouses. Moreover, the libraries assist researchers to test robots in prototype environments and immediately start building.

Nvidia has advanced in several simulation tools, such as Isaac Gym, which provides AI reinforcement learning and Isaac Cortex, a collaborative robot programming tool. Furthermore, the company introduces Isaac ORBIT offering simulation to the manipulator robots to determine motions for moving, picking up, and putting down the objects.

Talla mentioned that the company has marched on “having a thousand to a million times more virtual robots for every physically robot deployed” adding to the platform’s growth. More than thousand companies and over a million developers have been already using the platform for different operations.

For instance, Telexistence Inc., a Japanese developer of remote-controlled and AI robotics, deployed over 300 beverage restocking robots across 300 convenience stores in the country. Deutsche Bahn AG, a German railway company, has been training the robotic AI models with the help of Isaac Sim to better handle the luggage falling off racks or carriages.