The partnership between Cadence and Nvidia in the field of robotics.

      The two companies revealed an enhanced partnership during a Cadence conference in Santa Clara on Wednesday. Their objective is to improve the accuracy of robot training data to expedite the deployment of physical AI systems in the real world.

      Cadence Design Systems and Nvidia have come together to address one of the most enduring challenges in robotics: the disparity between how robots learn in computer simulations and their actual performance in physical environments. The collaboration, presented by both companies’ CEOs at the Cadence conference in Santa Clara, California, merges Cadence’s high-fidelity physics simulation engines with Nvidia’s AI training platforms, which encompass its Isaac open-source simulation libraries and Cosmos open-world models.

      Cadence is primarily recognized as a leading supplier of software for designing advanced computing chips. However, it also develops physics engines that simulate interactions among real-world materials, the deformation of metals, fluid dynamics, and surface contact.

      These simulations are utilized in aerospace, automotive, and semiconductor design, but are now being repurposed to address a new challenge: producing the training data necessary for robot AI systems to learn object handling and physical navigation.

      Training robots in a simulated environment is faster and more cost-effective than training them in reality, but the usefulness of the training data is limited by the accuracy of the physics engine.

      "The more accurate the generated training data is, the better the model will be," stated Cadence CEO Anirudh Devgan at the Santa Clara conference.

      Nvidia CEO Jensen Huang clearly outlined the partnership’s extent: “We’re working with you across the board on robotic systems.”

      The integrated system will connect Cadence’s multiphysics simulation with Nvidia’s model training pipelines, deploying the results on Nvidia’s Jetson robotics and edge AI hardware.

      This will create a streamlined workflow that encompasses world-model training, physics simulation, and real-world deployment feedback, all coordinated by AI agents throughout the process.

      This announcement is part of Nvidia's broader strategy of forming deep simulation partnerships in the field of industrial engineering. The company has also announced collaborations with Siemens and Dassault Systèmes to develop industrial AI platforms and virtual twins.

      For Cadence, venturing into robotics signifies a substantial extension of its simulation software into the AI infrastructure layer at a time when there is a rapidly increasing need for precise robot training data.