Parcels for the next day. Deliveries of new vehicles. Fresh organic products. Each of these modern conveniences are accelerated by fleets of mobile robots.
NVIDIA today announces updates to Nova Orin—a reference platform for autonomous mobile robots (AMR)—that advance its roadmap. We publish details of three reference platform configurations. Two use a single Jetson AGX Orin — running the NVIDIA Isaac robotics stack and the Robot Operating System (ROS) with the GPU-accelerated framework — and one is based on two Orin modules.
The Nova Orin platform was designed to improve reliability and reduce development costs for building and deploying AMRs worldwide.
AMRs are like self-driving cars, but for unstructured environments. You don’t need fixed, pre-programmed routes and you can avoid obstacles. This makes them ideal in logistics for moving items in warehouses, distribution centers and factories, or for applications in hospitality, cleaning, roaming security and last mile delivery.
For years, AMR manufacturers have been developing robots by sourcing and integrating computing hardware, software, and sensors in-house. This time-consuming effort requires years of engineering resources, lengthens go-to-market pipelines, and distracts from the development of domain-specific applications.
Nova Orin offers a better way forward with tested, industrial-grade configurations of sensors, software, and GPU computing capabilities. Leveraging the NVIDIA AI platform, developers can focus on building their unique software stack of robotic applications.
The stakes are high for the intralogistics enabled by AMRs across all industries, a market that ABI Research estimates is projected to grow nearly six-fold to $46 billion by 2030, up from $8 billion in 2021.
Design a high-performance, flexible reference architecture
The Nova Orin reference architecture designs are provided for specific use cases. There is an Orin-based design without safety-certified sensors and one that includes them along with a safety programmable logic controller. The third architecture has a dual Orin-based design that relies on Vision AI to enable functional safety.
Sensor support is included for stereo cameras, lidars, ultrasonic sensors, and inertial measurement units. The selected sensors were chosen to balance performance, price and reliability for industrial applications. The range of sensors provides a multi-modal coverage diversity required for the development and deployment of secure and collaborative AMRs.
The stereo cameras and fisheye cameras are custom developed by NVIDIA in coordination with camera partners. All sensors are calibrated and time-synchronized and come with drivers for reliable data collection. These sensors allow AMRs to detect objects and obstacles in a wide range of situations while enabling simultaneous localization and mapping (SLAM).
NVIDIA offers two lidar options, one for applications that don’t require functional safety-certified sensors and the other for those that do. In addition to these 2D lidars, Nova Orin supports 3D lidar for mapping and collecting ground truth data.
Building a comprehensive AI platform for OEMs, ISVs
NVIDIA drives the Nova Orin platform alongside the hardware and integration tools with extensive software support.
The base operating system includes drivers and firmware for all hardware and customization tools, as well as design guides for integration with robots. Nova easily integrates with a ROS-based robotic application.
The sensors have validated models in Isaac Sim for application development and testing without the need for a real robot.
The cloud-native data collection tools eliminate the tedious task of setting up data pipelines for the massive amount of sensor data needed for model training, debugging, and analysis. State-of-the-art GEMs developed for Nova sensors are GPU-accelerated with the Jetson Orin platform and offer key building blocks such as visual SLAM, stereo depth estimation, obstacle detection, 3D reconstruction, semantic segmentation, and pose estimation.
Nova Orin is also addressing the need to rapidly create high-resolution, city-scale 3D indoor maps in the cloud. These generated maps enable robot navigation, fleet planning and simulation. In addition, the maps can be continuously updated with the data from the robots.
Industrial grade AMRs
As robotic systems evolve, the need to securely deploy and manage the critical onboard AI software is paramount for future AMRs.
Nova Orin supports secure over-the-air updates and device management and monitoring to enable easy deployment and reduce maintenance costs. The open, modular design allows developers to take some or all of the platform’s capabilities and extend them to rapidly develop robotics applications.
NVIDIA is working closely with regulators to develop image-based security technology to further reduce costs and improve reliability of AMRs. And we provide a navigation software development kit to help developers quickly build applications.
Improving the productivity of factories and warehouses will depend on AMRs working safely and efficiently side-by-side at scale. A high level of autonomy, powered by Nova Orin’s 3D perception, will help fuel this revolution.
learn more about Nova Orin and Sign in to be notified its availability.