According to ABI Research, the number of industrial and commercial robots installed will grow more than 6.4 times from 3.1 million in 2020 to 20 million in 2030. To develop, validate, and deploy these new AI robots, companies will need to use simulation techniques to put these robots into real-world scenarios.
NVIDIA announced a major update to Isaac Sim at CES. The robot simulation tool can build and test virtual robots in a variety of different realistic environments. Isaac Sim is built on the foundation of the meta-universe app creation and running platform, NVIDIA Omniverse, which is now open to cloud access.
Powerful AI capabilities for roboticists
As humans begin to work more and more side by side with cooperative robots (Cobots) or autonomous mobile robots (AMRs), it has become critical to include humans and their common behaviors in simulations.
Isaac Sim's new human simulation features add human characters to a warehouse or production facility and have them perform common tasks like stacking packages and pushing carts. The tool now supports many common behaviors, so simulating them is as simple as issuing commands.
In order to minimize the gap between what is observed in the simulated world and what is seen in the real world, it is necessary to have a sensor model that conforms to physics.
With NVIDIA's RTX technology, Isaac Sim can now render physics-compliant data from sensors in real time. When using RTX analog Lidar, ray tracing can provide more accurate sensor data in a variety of lighting conditions or in response to reflective materials.
Isaac Sim also offers a number of new simulable 3D assets, which are critical to building a physics simulation environment. Everything from warehouse parts to robots is readily available, so developers and users can start building quickly.
The Isaac Gym for reinforcement learning and the Isaac Cortex for collaborative robot programming have both made advances, and these new capabilities are important for robotics researchers. In addition, a new tool, Isaac ORBIT, provides a simulated operating environment and benchmark for robot learning and motion planning.
Aiming at the large robotics operating system (ROS) developer community, Isaac Sim has upgraded support for ROS 2 Humble and Windows. All Isaac ROS software is now available in simulations.
Accelerate adoption by expanding the capabilities and ecosystem of the Isaac platform
The vast and complex robot ecosystem covers industries ranging from logistics and manufacturing to retail, energy and sustainable agriculture.
More than a million developers already use NVIDIA's Isaac robot platform, and more than a thousand enterprises rely on one or more parts of the platform, including many that have deployed physical robots that use Isaac Sim to develop and test in virtual worlds.
Telexistence has deployed drink-restocking robots in 300 convenience stores across Japan; To improve safety, Deutsche Bahn is training AI models to handle unexpected extreme events that are crucial but rarely happen in the real world, such as luggage falling on train tracks; Sarcos Robotics is developing robots that pick and place solar panels in renewable energy installations.
Festo uses the Isaac Cortex to simplify the programming of cooperative robots and transfer simulation skills to physical robots; Fraunhofer is developing advanced AMRs using Isaac Sim's physics-compliant full fidelity visualization capabilities; Flexiv is using Isaac Replicator to generate composite data for training AI models.
While training robots is important, simulation also plays a key role in training human operators to work with and program robots. Ready Robotics is using Isaac Sim to teach programming for industrial robots; Universal Robotics is using Isaac Sim to build a workforce and train terminal operators in the cloud.
Cloud access puts the Isaac platform within reach
With Isaac Sim now open to cloud access, multidisciplinary teams working on robotics projects around the world can collaborate on testing and training virtual robots, improving accessibility, flexibility and scalability.
When using robotic systems to build new facilities or extend existing autonomous systems, a lack of adequate training data is often a "roadblock" to deployment. Isaac Sim uses the Isaac Replicator to enable developers to create large-scale baseline real-world data sets that mimic real-world physical environments.
After deployment, as automation requirements expand, dynamic route planning is required to operate an efficient fleet of hundreds of robots. NVIDIA cuOpt is a real-time fleet task allocation and route planning engine that improves operational efficiency through automation.
Start using Isaac Sim
Download Isaac Sim right away and learn more technical details about the features and capabilities of the new platform.
Watch NVIDIA's special presentation at CES and hear NVIDIA executives announce products, cooperation projects and services in autonomous machines, robotics, design, simulation, etc.
about NVIDIA
Since its founding in 1993, NVIDIA (NASDAQ: NVDA) has been a pioneer in accelerated computing. NVIDIA's 1999 invention of the GPU has driven the growth of the PC gaming market and redefined modern computer graphics, ushering in the era of modern AI and driving the creation of the meta-universe. NVIDIA is now a full-stack computing company whose data-center scale products are reshaping entire industries.
