NVIDIA Isaac ROS

The latest Humble ROS 2 release improves performance on compute platforms that offer hardware accelerators. It enables accelerated computing features for type adaptation and type negotiation, eliminating software/CPU overhead and improving performance of hardware acceleration.

The NVIDIA implementation of type adaption and negotiation is called NITROS. These are ROS processing pipelines made up of Isaac ROS hardware-accelerated modules (a.k.a. GEMs). The source code of NITROS is now available for developers to modify, extend, and use in your applications.

NVIDIA Isaac Transport for ROS

H.264 video encode and decode hardware-accelerated packages for NITROS are used for compressed camera data recording and playback for development of AI models and perception functions. They compress two 1080p stereo cameras at 30fps (>120fps total) and reduce the data footprint by ~10X.

Knowledge of a robot’s position alone isn't enough to safely navigate complex environments. Robots must also be able to discover obstacles on their own. NvBlox (preview) uses RGB-D data to create a dense 3D representation of the robot's environment. It includes unforeseen obstacles that could cause a danger to the robot if not observed in real time. This data helps generate a temporal costmap for navigation stack.

Isaac ROS NvBlox

DNN Inference GEM is a set of ROS 2 packages that lets developers use any of NVIDIA’s numerous inference models available on NGC, or even provide their own DNN. Further tuning of pre-trained models or optimizations of developers' own models can be done with the NVIDIA TAO Toolkit.

After optimization, these packages are deployed by the NVIDIA TensorRT™ high-performance inference SDK or Triton™ , NVIDIA’s inference server. If the desired DNN model isn't supported by TensorRT, then Triton can be used to deploy the model.

Additional GEMs incorporating model support are available and include support for U-Net and DOPE. The U-Net package, based on TensorRT, can be used for generating semantic segmentation masks from images and the DOPE package can be used for 3D pose estimation for all detected objects.

DNN Inference GEM is the fastest way to incorporate performant AI inference in a ROS 2 application. The pretrained model—PeopleSemSegNet, pictured in the image (right)—runs at 325fps @544p on NVIDIA Jetson AGX Orin.

Isaac ROS DNN Inference
Isaac ROS Pose Estimation
Isaac ROS Image Segmentation

Stereo perception DNN-based GEMs are designed to help roboticists with common perception tasks.

Enhanced Semi-Supervised stereo disparity (ESS) is a DNN for stereo camera disparity prediction and Bi3D is a DNN for vision-based proximity detection.

Both are pretrained for robotics applications using synthetic data and are intended for commercial use.

Isaac ROS DNN Stereo Disparity
Isaac ROS Proximity Segmentation

As autonomous machines move around in their environments, they must keep track of where they are. VSLAM provides a method for visually estimating the position of a robot relative to its start position, known as VO (visual odometry).The Isaac ROS GEM for VSLAM provides this powerful functionality to ROS 2 developers.

This GEM offers the best accuracy for a real-time stereo-camera VSLAM solution. You can find publicly available results based on the widely used KITTI database here. For the KITTI benchmark, the algorithm achieves a drift of ~1% in localization and an orientation error of 0.003 degrees per meter of motion. In addition to being very accurate, this GPU-accelerated package runs extremely fast. The package uses cuVSLAM library to find and match more key points in real-time, with fine tuning to minimize overall reprojection error. This is attained by using a combination of visual data and IMU measurements

Isaac ROS Stereo Visual SLAM

Isaac Mission Dispatch allows a cloud/edge system to send and monitor tasks from a ROS 2 robot with Isaac Mission Client using industry standards for production deployments. Mission Dispatch is a cloud-native microservice that can be integrated as part of larger fleet management systems.

Mission Dispatch and Mission Client are both available in open source and can be used to test robots in simulation for automating test portions of continuous integration and continuous deployments (CI/CD), performing a series of predefined tasks evaluated against expected results. This benefit is in addition to the primary usage of assigning tasks to robots in operation.

Mission Dispatch can be integrated into fleet management systems (e.g., Anyfleet, Roborunner FleetGateway) with Mission Client on the ROS 2 robot. It will also interoperate with other ROS 2 Clients built on VDA5050.

Isaac ROS Mission Dispatch
Isaac ROS Mission Client

The image shows a lens-distorted camera image (left) and rectified image using LDC GEM (right).

In a typical robotics image-processing pipeline, raw data from the camera sensor must be processed before being passed off to a DNN or classic computer vision module for perception processing. This image processing consists of things like Lens Distortion Correction (LDC), image resizing, and image format conversion. If stereo cameras are involved, then estimating disparity is also required. The image processing GEMs have been designed to take advantage of the specialized computer vision hardware available in Jetson solutions, like the GPU, the VIC (Video and Image Compositor), and the PVA (Programmable Vision Accelerator).

For robots using cameras connected via a CSI interface, NVIDIA provides the Argus package for ROS hardware acceleration.

Isaac ROS Image Processing