Nvidia’s Isaac Manipulator, introduced earlier this year, is a significant step forward in industrial automation. It consists of foundation models and GPU-accelerated libraries that enable scalable workflows for dynamic manipulation tasks. These foundation models, based on transformer deep learning architecture, allow robots to perceive and make decisions autonomously, similar to human-like understanding.
By partnering with Nvidia, Intrinsic aims to demonstrate the potential for a universally applicable robotic-grasping skill that can work seamlessly across different grippers, environments, and objects. Wendy Tan White, CEO of Intrinsic, emphasizes the transformative impact of foundation models in simplifying processing challenges, reducing development costs, and increasing flexibility for end-users.
The collaboration between Nvidia and Intrinsic also focuses on the development of next-generation robots, with a central goal of improving robot grip. Historically, programming and scaling robot grasping has been challenging. However, by leveraging Nvidia Isaac Sim on the Omniverse platform, Intrinsic has used simulation to generate synthetic data for vacuum grasping, enabling significant advancements in industrial automation.
Intrinsic Flowstate, a developer environment for AI-based robotics solutions, plays a crucial role in visualizing processes, perception, and motion planning. By incorporating Isaac Manipulator into the workflow, users can generate grasp poses and CUDA-accelerated robot motions. These can be evaluated in simulation with Isaac Sim before being deployed in the real world using the Intrinsic platform. This simulation step helps save costs and ensures more efficient deployment.
Nvidia and Intrinsic aim to bring state-of-the-art dexterity and modular AI capabilities to robotic arms by offering a comprehensive collection of foundation models and GPU-accelerated libraries. This will accelerate the development of new robotics and automation tasks.
At the Automate trade show, Nvidia senior research scientist Adithya Murali and Intrinsic chief science officer Torsten Kroeger will showcase their collaborative efforts in automating smart pick-and-place tasks using Intrinsic Flowstate and NVIDIA Isaac Manipulator.
Overall, the partnership between Nvidia and Intrinsic highlights the advancements being made in AI integration in the field of robotics. By leveraging foundation models and simulation tools, these companies are paving the way for more efficient and scalable industrial automation solutions. The possibilities for robotic grasping and industrial scalability are expanding, offering businesses new opportunities for growth and innovation.
