agentleFS
Seeed-ProjectsSeeed-Projects/Seeed-Jetson-DevelopTool/skills/codex/deploy-nvblox/AGENTS.md

Seeed-Jetson-DevelopTool / deploy-nvblox

AGENTS.md57 starsChanged 4 months agoMITView on GitHub

Deploy Isaac ROS NVBlox with Orbbec RGB-D camera on Jetson AGX Orin for real-time GPU-accelerated 3D mapping (TSDF/ESDF). Sets up Docker CE, Isaac ROS 3.2, Orbbec SDK ROS2, and builds the NVBlox workspace for obstacle detection and 3D mesh generation. Requires JetPack 6.2 and ROS2 Humble.

  • Installs packages

The file

We split this AGENTS.md into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run one phase at a time. After each phase:

    line 14
  2. Very specific:

    If output contains [STOP] → stop, consult the failure decision tree.

    line 16
  3. Very specific:

    If output ends with [OK] → tell the user "Phase N complete" and proceed.

    line 17
  4. Distinctive:

    If you get a CDI error, fix with:

    line 92
  5. Distinctive:

    Run these commands on the host (not inside Docker):

    line 104

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Deploy NVBlox with Orbbec Camera on Jetson AGX Orin

NVBlox builds dense 3D maps in real-time from RGB-D camera input for robotic navigation. This skill deploys it on AGX Orin with Isaac ROS and Orbbec camera support.


Execution model

Run one phase at a time. After each phase:

  • Relay all output to the user.
  • If output contains [STOP] → stop, consult the failure decision tree.
  • If output ends with [OK] → tell the user "Phase N complete" and proceed.

Prerequisites

RequirementDetail
HardwareJetson AGX Orin (e.g. reComputer J50)
JetPack6.2
CameraOrbbec RGB-D camera (Gemini 2, 330 series, etc.)
ROS2Humble installed on host
NGC accountRequired for Isaac ROS Docker image pull

Phase 1 — Install basic dependencies (~3 min)

sudo apt update
sudo apt-get install -y python3-pip nvidia-jetpack git-lfs
sudo pip3 install jetson-stats

[OK] when all packages install. [STOP] if apt fails.


Phase 2 — Install Docker CE (~5 min)

sudo apt-get install -y apt-transport-https ca-certificates curl software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository "deb [arch=arm64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt-get update
sudo apt-get install -y docker-ce
sudo systemctl enable docker && sudo systemctl start docker
sudo usermod -aG docker $USER

Reboot after adding user to docker group:

sudo reboot

Verify after reboot:

docker info

[OK] when docker info succeeds without sudo. [STOP] if Docker fails to start.


Phase 3 — Install Isaac ROS 3.2 (~10–20 min)

mkdir -p ~/workspaces/isaac_ros-dev/src
echo "export ISAAC_ROS_WS=\${HOME}/workspaces/isaac_ros-dev/" >> ~/.bashrc
source ~/.bashrc
cd ${ISAAC_ROS_WS}/src
git clone -b release-3.2 https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_common.git

Pull and enter the Isaac ROS Docker container:

cd ${ISAAC_ROS_WS}/src/isaac_ros_common
./scripts/run_dev.sh

This will pull the NGC Docker image and start the container. You will need your NGC API Key.

If you get a CDI error, fix with:

sudo nvidia-ctk cdi generate --mode=csv --output=/etc/cdi/nvidia.yaml

[OK] when you are inside the Isaac ROS Docker container. [STOP] if NGC auth or Docker fails.


Phase 4 — Install Orbbec SDK ROS2 on host (~10 min)

Run these commands on the host (not inside Docker):

sudo apt install -y libgflags-dev nlohmann-json3-dev \
    ros-humble-image-transport ros-humble-image-transport-plugins ros-humble-compressed-image-transport \
    ros-humble-image-publisher ros-humble-camera-info-manager \
    ros-humble-diagnostic-updater ros-humble-diagnostic-msgs ros-humble-statistics-msgs ros-humble-xacro \
    ros-humble-backward-ros libdw-dev libssl-dev mesa-utils libgl1 libgoogle-glog-dev

mkdir -p ~/ros2_ws/src
cd ~/ros2_ws/src
git clone https://github.com/orbbec/OrbbecSDK_ROS2.git
cd OrbbecSDK_ROS2 && git checkout v2-main
cd ~/ros2_ws
colcon build --event-handlers console_direct+ --cmake-args -DCMAKE_BUILD_TYPE=Release

Install udev rules:

cd ~/ros2_ws/src/OrbbecSDK_ROS2/orbbec_camera/scripts
sudo bash install_udev_rules.sh
sudo udevadm control --reload-rules && sudo udevadm trigger

[OK] when colcon build completes and udev rules are installed. [STOP] if build fails.


Phase 5 — Build NVBlox in Isaac ROS container (~15 min)

Clone the Orbbec-adapted NVBlox source:

git clone https://github.com/jjjadand/isaac-NVblox-Orbbec.git
cp -r isaac-NVblox-Orbbec/src/isaac_ros_nvblox/ ${ISAAC_ROS_WS}/src/
cp -r isaac-NVblox-Orbbec/src/isaac_ros_nitros/ ${ISAAC_ROS_WS}/src/
cp -r isaac-NVblox-Orbbec/build/ ${ISAAC_ROS_WS}/

Enter the Docker container and build:

cd ${ISAAC_ROS_WS}/src/isaac_ros_common && ./scripts/run_dev.sh

# Inside container:
sudo apt update
sudo apt-get install -y ros-humble-magic-enum ros-humble-foxglove-msgs

echo 'CUDA_HOME=/usr/local/cuda
export PATH=$CUDA_HOME/bin:$PATH
export LD_LIBRARY_PATH=$CUDA_HOME/lib64:$LD_LIBRARY_PATH
export CUDACXX=$CUDA_HOME/bin/nvcc' >> ~/.bashrc

sudo ln -sf /opt/ros/humble/include/magic_enum.hpp /usr/include/magic_enum.hpp
sudo mkdir -p /opt/ros/humble/include/foxglove_msgs
sudo ln -sfn /opt/ros/humble/include/foxglove_msgs/foxglove_msgs/msg /opt/ros/humble/include/foxglove_msgs/msg

colcon build --symlink-install --cmake-args -DBUILD_TESTING=OFF
source install/setup.bash

[OK] when colcon build completes with all packages. [STOP] if compilation errors.


Phase 6 — Launch NVBlox (~2 min)

On the host, start the Orbbec camera node:

cd ~/ros2_ws
source install/setup.bash
ros2 launch orbbec_camera gemini2.launch.py
# Replace gemini2 with your camera model: gemini210, gemini2L, gemini_330_series, etc.

In the Isaac ROS Docker container, verify topics and launch NVBlox:

ros2 topic list
# Should show /camera/color/image_raw, /camera/depth/image_raw, etc.

cd ~/workspaces/isaac_ros-dev
source install/setup.bash
ros2 launch nvblox_examples_bringup orbbec_example.launch.py

[OK] when RViz shows 3D occupancy grids and meshes.


Failure decision tree

SymptomAction
NGC auth failsRegenerate API key at https://ngc.nvidia.com. Run ngc config set again.
CDI device error on run_dev.shRun: sudo nvidia-ctk cdi generate --mode=csv --output=/etc/cdi/nvidia.yaml.
Orbbec camera not detectedInstall udev rules (Phase 4). Replug camera. Check lsusb.
No camera topics in DockerOrbbec node must run on host, not in Docker. Verify ROS_DOMAIN_ID matches.
colcon build fails on NVBloxCheck magic_enum and foxglove_msgs are installed. Verify CUDA env vars.
RViz shows no dataConfirm camera topics are publishing: ros2 topic hz /camera/depth/image_raw.

Reference files

  • references/source.body.md — Full Seeed Wiki tutorial with Docker setup, build details, RViz configuration, and demo video (reference only)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.