Cadence and Nvidia: Revolutionizing AI and Robotics

Cadence and Nvidia: A Strategic Alliance to Revolutionize AI and Robotics

At its annual CadenceLIVE event, Cadence Design Systems unveiled two major collaborations in the field of artificial intelligence. The company announced the expansion of its partnership with Nvidia and the integration of new solutions with Google Cloud. These initiatives aim to combine AI with physics-based simulations and accelerated computing, specifically for robotic systems and large-scale system design.

Both companies emphasized that this collaborative approach is intended to model and deploy solutions in key sectors such as semiconductors, robotics, and large-scale AI infrastructures. Nvidia, in particular, describes these robotic systems as embodying a form of "physical AI," a new frontier in the application of artificial intelligence.

Integration of Simulation and Design Tools

Cadence has integrated its multiphysics simulation and system design tools with Nvidia's CUDA-X libraries, as well as AI models and the Omniverse simulation environment. These tools allow for the modeling of thermal, electrical, and mechanical interactions, providing engineers with the ability to assess system behavior under real-world conditions. This approach is not limited to chip design but also extends to critical infrastructure components such as networking, cooling, and power systems.

The combined platform offers engineers the capability to simulate system behavior before physical deployment. The companies highlighted that the overall performance of a system depends on the harmonious interaction between computing, networking, cooling, and power systems.

Robotics Development and Simulation Training

The collaboration between Cadence and Nvidia also extends to robotics development. Cadence's physical engines, which model real-world material interactions, are paired with Nvidia's AI models. These models are used to train AI-driven robotic systems in simulated environments, thereby reducing the need for real-world data collection.

Nvidia's CEO, Jensen Huang, stated at the event: "We are working with you on all fronts regarding robotic systems." The companies clarified that the necessary datasets must be generated using physics-based models rather than collected from physical systems. These simulated datasets are then used to train models, whose accuracy depends on the fidelity of the underlying physical models. Cadence's CEO, Anirudh Devgan, remarked: "The more accurate the generated training data, the better the model will be."

Industry Adoption of Robotics

Nvidia revealed that industrial robotics companies, such as ABB Robotics, FANUC, YASKAWA, and KUKA, are utilizing its Isaac simulation frameworks and Omniverse-based digital twin tools. These tools enable testing of robotic systems before physical deployment by modeling complex robotic operations and entire production lines in physically accurate digital environments.

Cloud-Based Chip Design Automation

In parallel, Cadence introduced a new AI agent designed to automate advanced chip design tasks. This agent focuses on physical layout processes, translating circuit designs into silicon implementations. This innovation builds on a previous agent introduced for upstream chip design, where circuits are defined in code-like descriptions.

The new agent will be available via Google Cloud, allowing design teams to execute these workloads without relying on on-premises computing infrastructure. Cadence stated that the integration combines its electronic design automation tools with Google’s Gemini models, facilitating automated design and verification workflows.

Productivity Gains and Simulation Validation

Cadence reported significant productivity gains, up to tenfold during initial deployments in design and verification tasks. Although the company did not disclose specific details about implementations at client sites, it emphasized the use of digital twin models to test design trade-offs and optimize software configurations.

The companies also noted that the cost and complexity of large-scale data center infrastructures limit the use of trial-and-error deployment methods, making virtual simulations all the more crucial.

Announcement of Quantum Models

In a separate announcement, Nvidia introduced a new family of open-source quantum AI models, named NVIDIA Ising. These models, named after the Ising model, a mathematical framework used to represent interactions in physical systems, are designed to support the calibration of quantum processors and quantum error correction.

Nvidia claimed that these models offer performance up to 2.5 times faster and three times greater accuracy in decoding processes used for error correction. These models are intended for use in hybrid quantum-classical systems, where AI plays a crucial role in making quantum computing more practical and reliable. Jensen Huang stated: "With Ising, AI becomes the control plane — the operating system of quantum machines — transforming fragile qubits into scalable and reliable quantum-GPU systems."