Cadence, Nvidia Team Up for Robot Training
The partnership combines simulation software with AI models to improve robot training and cut deployment time.
Topics
News
Chip design software maker Cadence Design Systems and Nvidia are expanding their partnership to develop artificial intelligence tools for robotics.
The companies said at a Cadence event in Santa Clara, California, that they plan to combine Cadence’s physics simulation software with Nvidia’s AI models to improve virtual training environments where robots learn tasks before being deployed in the real world.
Nvidia Chief Executive Jensen Huang said the companies were working together “across the board on robotic systems.”
“The more accurate the generated training data is, the better the model will be,” said Cadence Chief Executive Anirudh Devgan, underscoring the role of simulation in speeding robot training.
The companies said the collaboration is aimed at cutting the time needed for robots to carry out useful tasks.
Training robots in simulation can be faster than in physical environments, but it depends on high-quality data generated by software such as Cadence’s physics engines.
Separately, Cadence said it is expanding its use of AI in chip design.
The company introduced a new AI agent for back-end chip design, the stage when circuit designs are translated into physical layouts on silicon, adding to an earlier tool focused on front-end design.
The new system will be available through Google Cloud.
This follows an earlier tool focused on the initial stages of chip design, where circuits are created in code-like formats.
“We help build AI systems, and then those AI systems can help improve the design process,” Devgan said.
