American multinational technology company Nvidia has announced updates to Isaac Sim, its robotics simulation tool, which was designed to build and test virtual robots in realistic environments in a range of operating conditions.
The platform will now be accessible from the cloud, with the company saying this means global teams working on robotics projects can collaborate with increased accessibility, agility and scalability for testing and training virtual robots.
This function was added to support the improvement of training data, the poor quality of which can often delay deployment when building new robotics facilities or scaling existing autonomous systems.
Nvidia has also said that, with humans increasingly working alongside collaborative robots (cobots) or autonomous mobile robots (AMRs), it’s critical to the development process that people’s common behaviours are added to simulations.
As a response to this, the platform’s new people simulation capability was designed to let users add human characters to a warehouse or manufacturing facility and make them perform typical tasks, such as stacking packages or pushing carts.
What’s more, accurate simulation in training can improve the ability of human operators to work with and program robots, the company has said.
This, Nvidia claims, is to minimise the difference between results observed in a simulated environment, compared to real-world environments.
Using NVIDIA RTX technology, Isaac Sim can now reportedly render physically accurate data from sensors in real time, with the platform also offering various new simulation-ready 3D assets, which Nvidia reports as critical to the building of physically accurate virtual environments.