NVIDIA Introduces Isaac Foundation Models to Revolutionize Robotics Applications
The application of robotics is rapidly expanding in diverse environments such as smart manufacturing facilities, commercial kitchens, hospitals, warehouse logistics, and agricultural fields. The industry is shifting towards intelligent automation, which requires enhanced robot capabilities to perform functions including perception, mapping, navigation, load handling, object grasping, and intricate assembly tasks.
NVIDIA Isaac Perceptor
Autonomous mobile robots (AMRs) and automated guided vehicles (AGVs) are critical to assembly line efficiency, material handling, and healthcare logistics. As these robots navigate complex and unstructured environments, the ability to perceive and react to their surroundings becomes essential.
Isaac Perceptor, built on the NVIDIA Isaac Robot Operating System (ROS), empowers original equipment manufacturers (OEMs), freight service providers, software vendors, and the AMR ecosystem to accelerate development for robotics. Teams can equip mobile robots with perception capabilities for successful navigation and obstacle avoidance in unstructured environments.
Key features of Isaac Perceptor include multi-camera AI-based depth perception, visual inertial odometry, and real-time voxel grid mapping. Early collaborators include industry leaders like ArcBest, BYD Electronics, Gideon, KION, Kudan, idealworks, RGo, and Teradyne Robotics.
NVIDIA Isaac Manipulator
Isaac Manipulator is a collection of NVIDIA-accelerated libraries, AI models, and reference workflows designed to bring AI acceleration to robotic arms. It enables these arms to seamlessly perceive, understand, and interact with their environments.
The foundation models and accelerated libraries can be integrated as independent modules or as an entire workflow in solutions development. Early collaborators include companies such as Intrinsic, Siemens, Solomon, Techman Robot, Teradyne Robotics, Vention, and Yaskawa.
Key features include the cuMotion GPU-accelerated motion planner for faster path planning, FoundationPose for single-shot 6D pose estimation, and SyntheticaDETR for real-time object detection.
NVIDIA Isaac Sim 4.0
Isaac Sim allows developers to generate synthetic data and diverse, virtual complex test environments with industry-leading sensor and robot-type testing. This enables highly realistic simulations for testing thousands of robots simultaneously in real time.
Isaac Lab, built on the Isaac Sim platform, supports reinforcement learning, imitation learning, and transfer learning. It can train a wide range of robot embodiments, allowing developers to explore various designs and functionalities.
The new release brings ease of use with VSCode integration, multi-GPU support for reinforcement learning, performance improvements with RTX sensor tiled rendering, and more.
For more information, visit the NVIDIA Technical Blog.