Isaac ROS 0.30.0 (DP3)

Author: jaiveersinghNV

profile/README.md

@@ -1,6 +1,6 @@
 # NVIDIA Isaac ROS
 
-Welcome to Isaac ROS, a collection of ROS2 packages for making autonomous robots.
+Welcome to Isaac ROS, a collection of hardware-accelerated, high performance, low latency ROS 2 packages for making autonomous robots which leverage the power of Jetson and other NVIDIA platforms.
 
 * To get started, visit the [summary of packages](https://github.com/NVIDIA-ISAAC-ROS/.github/blob/main/profile/package-summary.md) and [examples](https://github.com/NVIDIA-ISAAC-ROS/.github/blob/main/profile/package-summary.md) of how to use them.
 * Compare the [performance](https://github.com/NVIDIA-ISAAC-ROS/.github/blob/main/profile/performance-summary.md) of the packages on various platforms.

profile/blog-posts.md

@@ -6,7 +6,7 @@ Read the latest posts from NVIDIA for more about Isaac ROS.
 
 - [Build Connected Robots with NVIDIA Isaac Dispatch and Client](https://gateway.on24.com/wcc/experience/elitenvidiabrill/1407606/3998202/isaac-ros-webinar-series)
 - [Accelerate YOLOv5 and Custom AI Models in ROS with NVIDIA Isaac](https://gateway.on24.com/wcc/experience/elitenvidiabrill/1407606/3998202/isaac-ros-webinar-series)
-- [Using ML Models in ROS2 to Robustly Estimate Distance to Obstacles](https://gateway.on24.com/wcc/experience/elitenvidiabrill/1407606/3998202/isaac-ros-webinar-series)
+- [Using ML Models in ROS 2 to Robustly Estimate Distance to Obstacles](https://gateway.on24.com/wcc/experience/elitenvidiabrill/1407606/3998202/isaac-ros-webinar-series)
 - [Pinpoint, 250 fps, ROS 2 Localization with vSLAM on Jetson](https://gateway.on24.com/wcc/experience/elitenvidiabrill/1407606/3998202/isaac-ros-webinar-series)
 
 ## NVIDIA Developer Blogs
@@ -23,7 +23,7 @@ Read the latest posts from NVIDIA for more about Isaac ROS.
 ## ROS Discourse
 
 - [Isaac ROS Oct update, Mission Dispatch & Client, Freespace and Image compression](https://discourse.ros.org/t/isaac-ros-oct-update-mission-dispatch-client-freespace-and-image-compression/27857)
-- [Isaac Mission Dispatch & Client w/ VDA5050 open source release for ROS2](https://discourse.ros.org/t/isaac-mission-dispatch-client-w-vda5050-open-source-release-for-ros2/27856)
+- [Isaac Mission Dispatch & Client w/ VDA5050 open source release for ROS 2](https://discourse.ros.org/t/isaac-mission-dispatch-client-w-vda5050-open-source-release-for-ros2/27856)
 - [Isaac ROS 1-year anniversary, Sept incremental release for Jetpack 5.0.2](https://discourse.ros.org/t/isaac-ros-1-year-anniversary-sept-incremental-release-for-jetpack-5-0-2/27190)
 - [Isaac ROS July update, Humble with hardware accelerated node graphs](https://discourse.ros.org/t/isaac-ros-july-update-humble-with-hardware-accelerated-node-graphs/26308)
 - [New type adaptation open source examples for Humble](https://discourse.ros.org/t/new-type-adaptation-open-source-examples-for-humble/26241)

profile/faq.md

@@ -2,15 +2,15 @@
 
 ## What is Isaac ROS?
 
-Isaac ROS is a collection of ROS2 packages for making autonomous robots. The Isaac ROS suite has been developed and released by NVIDIA to leverage the power of hardware acceleration on NVIDIA Jetson and discrete GPUs for standard robotics applications.
+Isaac ROS is a collection of ROS 2 packages for making autonomous robots. The Isaac ROS suite has been developed and released by NVIDIA to leverage the power of hardware acceleration on NVIDIA Jetson and discrete GPUs for standard robotics applications.
 
 A summary of all available Isaac ROS packages is available [here](./package-summary.md).
 
 ## How can I use a package?
 
 Each Isaac ROS package has a detailed quick start guide in the main-level `README.md` file. Additionally, many Isaac ROS packages include more advanced tutorials, such as camera-to-detection compute graph examples and integration with NVIDIA [Isaac Sim](https://developer.nvidia.com/isaac-sim).
 
-## I have native ROS2 code, will it work with Isaac ROS packages?
+## I have native ROS 2 code, will it work with Isaac ROS packages?
 
 Yes. Isaac ROS packages use standard ROS interfaces on input and output topics. Where appropriate, Isaac ROS packages are also designed to be drop-in replacements for the commonly-used, CPU-based ROS implementations familiar to robotics developers.  
 

profile/hardware-setup.md

@@ -5,15 +5,17 @@
 ### Discrete GPU
 
 1. Prepare a NVIDIA-powered platform with the following min specs:
+    * Ubuntu 20.04+
+      * **Experimental**: WSL2 on Windows 11
     * 16 GB general RAM
     * Discrete NVIDIA GPU with the following specs:
-        * supports CUDA 11.7 with Volta or newer
+        * supports CUDA 11.8 with Volta or newer
         * minimum 8GB of VRAM and recommended 12GB+
 
 ### Jetson Platform
 
 1. Install Jetpack using the steps found [here](https://docs.nvidia.com/jetson/jetpack/install-jetpack/index.html).
-2. After bootup confirm that you have installed `Jetpack 5.0.2` by running the following command. The output should have the terms `R35 (release), REVISION: 1.0`.
+2. After bootup confirm that you have installed `Jetpack 5.1.1` by running the following command. The output should have the terms `R35 (release), REVISION: 3.1`.
 
     ```bash
     cat /etc/nv_tegra_release

profile/package-summary.md

@@ -1,45 +1,58 @@
-# Isaac ROS Realsense Setup
+# Isaac ROS RealSense Setup
 
-1. Clone the `librealsense` repo setup udev rules. Remove any connected relasense cameras when prompted:
+## Camera Compatibility
 
-   ```bash
+| RealSense Model | Supported? |
+| --------------- | ---------- |
+| D455            | ✓    |
+| D435            | ✓    |
+| D415            | ✗    |
+
+> **Note:** For best results, please ensure that the camera's firmware has been updated.
+
+## Setup Instructions
+
+> **Note:** This tutorial assumes that you have set up your development environment by following the instructions [here](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_common/blob/main/docs/dev-env-setup.md).
+
+1. Clone the `librealsense` repo setup udev rules. Remove any connected RealSense cameras when prompted:
+
+    ```bash
     cd /tmp && \
     git clone https://github.com/IntelRealSense/librealsense && \
     cd librealsense && \
     ./scripts/setup_udev_rules.sh
     ```
 
-2. Clone the isaac_ros_common and the `4.51.1` release of the realsense repository:
+2. Clone the isaac_ros_common and the `4.51.1` release of the `realsense-ros` repository:
 
-    ```bash
-    mkdir -p ~/workspaces/isaac_ros-dev/src && cd ~/workspaces/isaac_ros-dev/src
-    ```
+    > Note: `${ISAAC_ROS_WS}` is defined to point to either `/ssd/workspaces/isaac_ros-dev/` or `~/workspaces/isaac_ros-dev/`.
 
     ```bash
+    cd ${ISAAC_ROS_WS}/src
     git clone https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_common
-    ```
-
-    ```bash
     git clone https://github.com/IntelRealSense/realsense-ros.git -b 4.51.1
     ```
 
-3. Plug in your realsense camera before launching the docker container in the next step.
+3. Plug in your RealSense camera before launching the docker container in the next step.
+
 4. Configure the container created by `isaac_ros_common/scripts/run_dev.sh` to include `librealsense`. Create the `.isaac_ros_common-config` file in the `isaac_ros_common/scripts` directory:
 
     ```bash
-    cd ~/workspaces/isaac_ros-dev/src/isaac_ros_common/scripts && \
-    touch .isaac_ros_common-config && \
-    echo CONFIG_IMAGE_KEY=humble.nav2.realsense > .isaac_ros_common-config
+    cd ${ISAAC_ROS_WS}/src/isaac_ros_common/scripts && \
+      touch .isaac_ros_common-config && \
+      echo CONFIG_IMAGE_KEY=ros2_humble.realsense > .isaac_ros_common-config
     ```
 
-5. Launch the Docker container using the `run_dev.sh` script:
+5. Launch the Docker container
 
     ```bash
-    cd ~/workspaces/isaac_ros-dev/src/isaac_ros_common && \
-      ./scripts/run_dev.sh
+    cd ${ISAAC_ROS_WS}/src/isaac_ros_common && \
+      ./scripts/run_dev.sh ${ISAAC_ROS_WS}
     ```
 
-6. At this point, you can check that the realsense camera is connected by running `realsense-viewer`:
+    This will rebuild the container image using `Dockerfile.realsense` in one of its layered stage. It will take some time for rebuilding.
+
+6. Once container image is rebuilt and you are inside the container, you can run `realsense-viewer` to check that the RealSense camera is connected.
 
    ```bash
    realsense-viewer
@@ -48,7 +61,3 @@
    If you turn on the "Stereo Module" in the GUI, you should see something like the following:
 
 <div align="center"><img src="../resources/isaac_ngc/realsense/realsense_viewer.png" width=800px/></div>
-
-> Note: This tutorial has been tested with a RealSense D455 or D435 connected to a Jetson Orin or Xavier AGX.
-
-> Note: In our testing it was important that the camera had a recent version of the firmware installed.

profile/performance-summary.md

@@ -1,5 +1,42 @@
 # Release Notes
 
+## Isaac ROS 0.30.0 DP (developer preview) April 5, 2023
+
+Packages available as `DP 3 Release` tagged `v0.30.0-dp` on GitHub.
+
+### What's New
+
+Updates including:
+
+- support for JetPack 5.1.1
+- support for Jetson Orin NX, and Jetson Orin Nano 8GB
+- support for RTX 4K series (Lovelace GPU) with CUDA 11.8 on x86_64
+- new package `isaac_ros_map_localization` using LIDAR to automatically find the pose of the robot for Nav2
+- `isaac_ros_vslam` improvements for better covariance estimation and planar constraints for mobile robots
+- `isaac_ros_nvblox` improvements for DNN-based person detection and removal from 3D scene reconstruction
+- `isaac_ros_h264_encoder` improvement adding H.264 P-frame encode for higher performance and smaller rosbags
+- `isaac_ros_h264_decoder` improvement adding H.264 decompression on Jetson (aarch64)
+- `isaac_ros_dnn_stereo_disparity` improvement of ESS DNN trained with real data and SDG (synthetic data generation)
+- new package `ros2_benchmark` for performance measurement of graphs of nodes in open source
+- new package `isaac_ros_benchmark` building upon `ros2_benchmark` to provide configurations to benchmark Isaac ROS graphs of nodes
+- open source release for multiple NITROS-enabled hardware accelerated packages
+
+### Limitations
+
+This release has the following known limitations, with workarounds available in the troubleshooting section:
+
+- `isaac_ros_triton` can under certain conditions report errors of the form `Incomplete inference; response appeared out of order`.
+- `isaac_ros_h264_encoder` can become unresponsive when starting a stream after stopping because of a known synchronization bug.
+
+### Feature Requests & Issues Addressed
+
+This release contains new features and fixes to address user feedback, including:
+
+- [`isaac_ros_visual_slam #53`](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_visual_slam/issues/53): Add instructions to clone the dependency `isaac_ros_nitros`
+- [`isaac_ros_visual_slam #54`](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_visual_slam/issues/54): Update depth emitter parameters passed to RealSense node
+- [`isaac_ros_nvblox #40`](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_nvblox/issues/40): Update `realsense-ros` branch name due to upstream change (kudos to [@ripdk12](https://github.com/ripdk12))
+- [`isaac_ros_mission_client #2`](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_mission_client/issues/2): Change parameter name to avoid namespace duplication (kudos to [@bobbleballs](https://github.com/bobbleballs))
+
 ## Isaac ROS 0.20.0 DP (developer preview) October 19, 2022
 
 Packages available as `DP 2 Release` tagged `v0.20.0-dp` on GitHub.
@@ -79,9 +116,9 @@ Packages available as `DP Release` tagged `v0.10.0-dp` on GitHub.
 
 ### What's New
 
-ROS2 Humble release of Isaac ROS with performance boost and new stereo perception functions including:
+ROS 2 Humble release of Isaac ROS with performance boost and new stereo perception functions including:
 
-- packages for ROS2 Humble (Foxy deprecated) on JetPack 5.0.1 DP
+- packages for ROS 2 Humble (Foxy deprecated) on JetPack 5.0.1 DP
 - support for Jetson Orin
 - support for Isaac Sim 2022.1 *(Isaac SIM 2021.2 is deprecated)*
 - `isaac_ros_nitros` implementing type adaptation & type negotiation features new to Humble boosting performance
@@ -106,7 +143,7 @@ Packages available as `EA3 Release` tagged `v0.9.3-ea3` on GitHub.
 
 GTC 2022 release of Isaac ROS with new packages for vision based autonomous navigation including:
 
-- packages for ROS2 Foxy on JetPack 4.6.1
+- packages for ROS 2 Foxy on JetPack 4.6.1
 - `isaac_ros_visual_slam` upgraded from SVIO as visual odometry source for Nav2 with save and load of feature maps
 - `isaac_ros_nvblox` for vision based temporary 3D obstacle map construction in real-time
 - `isaac_ros_object_detection` to detect and classify obstacles using DNN detection networks
@@ -119,7 +156,7 @@ Packages available as `EA2 Release` tagged `v0.9.1-ea2` on GitHub.
 
 Incremental update with fixes and new packages from previous ROSCon 2021 release, including:
 
-- packages for ROS2 Foxy on JetPack 4.6
+- packages for ROS 2 Foxy on JetPack 4.6
 - support for Isaac Sim 2021.3
 - `isaac_ros_pose_estimation` update adding CenterPose DNN for detection of pose
 - Isaac SIM <-> Isaac ROS tutorials with Jetson hardware in the loop
@@ -131,7 +168,7 @@ Incremental update with fixes and new packages from previous ROSCon 2021 release
 
 ROSCon 2021 release of Isaac ROS providing hardware acceleration for autonomous robots with new features including:
 
-- packages for ROS2 Foxy on JetPack 4.6
+- packages for ROS 2 Foxy on JetPack 4.6
 - `isaac_ros_visual_odometry` providing stereo visual inertial odometry
 - `isaac_ros_argus_camera` for hardware accelerated ISP of CSI and GMSL cameras
 - `isaac_ros_image_segmentation` for DNN based image segmentation including people detection pre-trained model
@@ -144,7 +181,7 @@ ROSCon 2021 release of Isaac ROS providing hardware acceleration for autonomous
 
 First release of Isaac ROS providing hardware acceleration for autonomous robots with new features including:
 
-- packages for ROS2 Foxy on JetPack 4.5
+- packages for ROS 2 Foxy on JetPack 4.5
 - `isaac_apriltag` providing hardware acceleration replacement for `apriltag`
 - `isaac_image_proc` providing hardware acceleration replacement for `image_proc`
 - `isaac_stereo_image_proc` providing hardware acceleration replacement for `image_proc`