Important results from the project
The project integrated Staer´s platform with Nvidia´s Omniverse and Isaac Sim tools. With this integration, it is now possible to generate synthetic data and simulated robot movements in warehouse environments for testing and training physical AI.
The integration also created Staer´s first public dataset, Staer Warehouses. It is an open synthetic dataset for physical AI research. The most densely annotated dataset that exists today for warehouse environments.
Expected long term effects
The project is expected to strengthen research and development in robotics and physical AI in industrial environments. Staer will continue to publish data and collaborate with leading research institutions in the field.
The relationships created during the US visit will help strengthen the Swedish ecosystem around robotics and physical AI towards international actors.
Approach and implementation
The project started with contacts to Nvidia. Then a couple of weeks of work to integrate the platforms. After that a couple of weeks of work to generate Staer Warehouses. The dataset was published and announced at the start of GTC in March followed by a week of meetings at the conference and in Bay Area.
Overall, the schedule was kept and the result was better than planned.
External links
The project description has been provided by the project members themselves and the text has not been looked at by our editors.
