NVIDIA ROS 2 Projects

NVIDIA provides packages for the development of AI applications for robotics.

ISAAC ROS Projects

  • Pre-built ROS 2 Humble support: Pre-built Debian packages for ROS 2 Humble on Ubuntu 20.04 for Jetson and other aarch64 platforms from the NVIDIA build farm.

  • CUDA with NITROS: This helps users develop their own CUDA enabled nodes that work with NITROS, the Isaac ROS implementation of type adaptation & type negotiation which enables accelerated computing in ROS 2.

  • Isaac ROS NITROS Bridge: NITROS bridge for optimized integration of Isaac ROS packages with existing ROS 1 applications. Use this to bridge your ROS applications to ROS 2 for accelerated computing with >2x speed up compared to using a traditional ROS bridge.

  • Nova Carter: A reference AMR for robotics development and research, powered by Isaac ROS and Nav2, and tuned with Open Navigation for teleoperation, mapping, and navigation.

  • Isaac ROS Nova: This repository provides a set of optimized packages to interface with the Isaac Nova Orin sensor suite.

  • Isaac ROS Pose Estimation: This repository contains ROS 2 packages to predict the pose of an object.

  • ROS2_Benchmark: ros2_benchmark provides the tools for measuring the throughput, latency, and compute utilization of these complex graphs without altering the code under test.

  • Isaac ROS Benchmark: This package builds upon the ros2_benchmark to provide configurations to benchmark Isaac ROS graphs.

  • Isaac ROS Map Localization: This module contains ROS 2 packages for lidar processing to estimate poses relative to a map. The Occupancy Grid Localizer processes a planar range scan to estimate pose in an occupancy grid map; this occurs in less than 1 second for most maps.

  • Isaac ROS Nitros: Isaac Transport for ROS package for hardware-acceleration friendly movement of messages.

  • Isaac ROS Compression: Hardware accelerated NITROS packages to compress camera data capture and playback for development of AI models and perception functions, compressing 4x 1080p cameras at 30fps (>120fps total) reducing data footprint by ~10x.

  • Isaac ROS DNN Stereo Depth: DNN Stereo Disparity includes packages for predicting disparity of stereo input.

  • Isaac ROS Depth Segmentation: Hardware-accelerated packages for depth segmentation.

  • Isaac ROS Nvblox : Hardware-accelerated 3D scene reconstruction and Nav2 local costmap provider using nvblox.

  • Isaac ROS Object Detection : Deep learning model support for object detection including DetectNet.

  • Isaac ROS DNN Inference : This repository provides two NVIDIA GPU-accelerated ROS 2 nodes that perform deep learning inference using custom models. One node uses the TensorRT SDK, while the other uses the Triton SDK.

  • Isaac ROS Visual SLAM : This repository provides a ROS 2 package that estimates stereo visual inertial odometry using the Isaac Elbrus GPU-accelerated library.

  • Isaac ROS Mission Client : This repository receives state and error updates from ROS and converts them to VDA5050 JSON messages for transmission by the ROS 2 -> MQTT node to Mission Dispatch.

  • Isaac ROS Argus Camera : This repository provides monocular and stereo nodes that enable ROS developers to use cameras connected to Jetson platforms over a CSI interface.

  • Isaac ROS Image Pipeline : This metapackage offers similar functionality as the standard, CPU-based image_pipeline metapackage, but does so by leveraging the Jetson platform’s specialized computer vision hardware.

  • Isaac ROS Common : Isaac ROS common utilities for use in conjunction with the Isaac ROS suite of packages.

  • Isaac ROS AprilTag : ROS 2 node uses the NVIDIA GPU-accelerated AprilTags library to detect AprilTags in images and publish their poses, ids, and additional metadata.

Additional Projects

Simulation Projects

  • Isaac Sim Nav2 : In this ROS 2 sample, we are demonstrating Omniverse Isaac Sim integrated with the ROS 2 Nav2 project.

  • Isaac Sim Multiple Robot ROS 2 Navigation : In this ROS 2 sample, we are demonstrating Omniverse Isaac Sim integrated with the ROS 2 Nav2 stack to perform simultaneous multiple robot navigation.