Nvidia’s VP of Engineering Claire Delaunay announced that the Nvidia Issac SDK is available for a download – the announcement was made at TechCrunch’s TC Sessions: Robotics + AI event at UC Berkeley. The product announcement per se was made last month and is a part of Nvidia’s efforts to make robotics development available for a wider range of organizations.
Nvidia released a statement that said –
Using computational graphs and an entity component system, the Isaac Robot Engine allows developers to break down complex robotic tasks into a network of smaller, simpler steps. Developing a complex system is made easy using Gems, which are modular capabilities for sensing, planning, and actuation that can be easily plugged into a robotics application.
The new SDK comes with robust features such as –
- Robotics AI and ML
- Obstacle detection
- Speech recognition
- Stereo depth estimation
These capabilities are the key to build even the most basic of robotic systems.
The SDK will also work brilliantly with Nvidia’s in-house robotic components viz. the Jetson Nano and the Jetson AGX Xavier. Claire Delaunay took center stage and actually demonstrated the products herself at the event. Key demonstrations included Nvidia’s dual in-house referenced platforms, Carter and Kaya at the event.
The announcement automatically puts the AI-powered robotics industry in a win-win scenario. On the one hand, companies that are only just starting off get their hands on a readymade technology, powered by Nvidia’s experience to build a robot from scratch. Usually, this process would have required an organization to build a team of engineers, which in turn would have taken several months, maybe years to put a fully build robot into the market.
On the other hand, Nvidia’s robotics ‘raw materials’ are bound to be a go-to platform for a large number of robotics firms.