FoundationPose Tracking

Overview

This tutorial walks you through a graph to run FoundationPose tracking using FoundationPose FoundationPose tracking has similar functionality as FoundationPose pose estimation. However, it utilizes the pose from the previous frame and requires only the refine network without batching which can significantly increase the speed of the pipeline. Therefore, you can use the FoundationPose pose estimation to obtain the initial pose on the first frame and then switch to FoundationPose tracking for subsequent frames.

Note

To ensure the tracking pipeline is able to get the initial pose estimation, please play the ROS bag after the launch file is fully started.

Tutorial Walkthrough

1. Complete the Isaac ROS FoundationPose Quickstart Guide.

2. Open a new terminal and launch the Docker container using the run_dev.sh script:

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

RosbagRealSense CameraHawk Camera

3. Continuing inside the container, install the following dependencies:

   sudo apt-get install -y ros-humble-isaac-ros-examples
   

4. Run the following launch files to spin up a demo of this package:

   Launch isaac_ros_foundationpose:

   ros2 launch isaac_ros_examples isaac_ros_examples.launch.py launch_fragments:=foundationpose_tracking interface_specs_file:=${ISAAC_ROS_WS}/isaac_ros_assets/isaac_ros_foundationpose/quickstart_interface_specs.json mesh_file_path:=${ISAAC_ROS_WS}/isaac_ros_assets/isaac_ros_foundationpose/Mustard/textured_simple.obj texture_path:=${ISAAC_ROS_WS}/isaac_ros_assets/isaac_ros_foundationpose/Mustard/texture_map.png score_engine_file_path:=${ISAAC_ROS_WS}/isaac_ros_assets/models/foundationpose/score_trt_engine.plan refine_engine_file_path:=${ISAAC_ROS_WS}/isaac_ros_assets/models/foundationpose/refine_trt_engine.plan rt_detr_engine_file_path:=${ISAAC_ROS_WS}/isaac_ros_assets/models/synthetica_detr/sdetr_grasp.plan
   

   Then open another terminal, and enter the Docker container again:

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

   Then, play the ROS bag:

   ros2 bag play -l ${ISAAC_ROS_WS}/src/isaac_ros_pose_estimation/resources/rosbags/foundationpose_tracking.bag/
   

Visualize Results

1. Open a new terminal inside the Docker container:

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

   RosbagRealSense CameraHawk Camera

   2. launch RViz2 to visualize the output

      rviz2 -d  $(ros2 pkg prefix isaac_ros_foundationpose --share)/rviz/foundationpose_tracking.rviz
      

   3. You should see a RViz2 window open as shown below showing the 3D bounding box overlaid over the input image