NVIDIA-ISAAC-ROS / NVIDIA-ISAAC-ROS/isaac_ros_visual_slam

isaac_ros_visual_slam 4.5 landmarks_cloud and observations_cloud not showing at all in rviz2

Open
#227 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
1.5k
Forks
206
PR merge metrics
No merged PRs in 30d

Description

Hi all! I have this configuration for launching isaac_ros_vslam 4.5 on ROS2 Jazzy on Ubuntu 24.04:

launch file:

isaac_ros_visual_slam_node = ComposableNodeContainer( 
        name="vslam_container",
        namespace="",
        package="rclcpp_components",
        executable="component_container_mt",
        composable_node_descriptions=[
            ComposableNode(
                package="isaac_ros_visual_slam",
                plugin="nvidia::isaac_ros::visual_slam::VisualSlamNode",
                name="isaac_ros_visual_slam_node",
                extra_arguments=[{"use_intra_process_comms": False}], # some bug?
                parameters=[
                    vslam_config_file,
                    {
                        "camera_optical_frames": [
                            [front_camera_name, "_color_optical_frame"],
                        ],
                        "imu_frame": [front_camera_name, "_link"],        
                         
                        "use_sim_time": use_sim_time,  # "use_sim_time": use_sim_time, 
                    
                    }
                ],
                remappings=[
                    ("visual_slam/image_0", [front_camera_name, '/color/image_raw']),
                    
                    ("visual_slam/camera_info_0", [front_camera_name, '/color/camera_info']),

                    ("visual_slam/depth_0", [front_camera_name, '/depth/image_raw']),
                 
                ],
            ),
        ],
        parameters=[{
            'use_sim_time': use_sim_time,
        }],
        output="screen",
    )

The use_sim_time parameter is set correctly above this NodeContainer.

vslam_config.yaml:

isaac_ros_visual_slam_node:
  ros__parameters:
    tracking_mode: 2 # Multi-camera (0), Visual Inertial (1), or RGBD Tracking (2)
    num_cameras: 1  # stereo camera
    min_num_images: 1 # 1 for RGBD Tracking
    depth_camera_id: 0 # comment for tracking_mode: 0
    depth_scale_factor: 1000.0

    enable_image_denoising: false
    enable_ground_constraint_in_odometry: true
    enable_ground_constraint_in_slam: true
    enable_localization_n_mapping: false # If enabled, SLAM mode is on. If disabled, only odometry is on.

    image_jitter_threshold_ms: 400.0 

    map_frame: "map" 
    odom_frame: "odom"
    base_frame: "base_link" 

    camera_optical_frames:
      - "camera_front_color_optical_frame"
    imu_frame: "camera_front_link"

    rectified_images: false

    depth_enable_stereo_tracking: false

    publish_odom_to_base_tf: false # Enable tf broadcaster for odom_frame -> base_frame transform.
    publish_map_to_odom_tf: false # Enable tf broadcaster for map_frame -> odom_frame transform.

    verbosity: 999
    path_max_size: 2000
    enable_debug_mode: false

    enable_slam_visualization: true # Main flag to enable or disable visualization.
    enable_landmarks_view: true # If enabled, landmark pointcloud will be available for visualization.
    enable_observations_view: true # If enabled, 2D feature pointcloud will be available for visualization.

The path appears to be showing in RVIZ2, and the odometry as well. The tracking works pretty well on RGBD mode, but I am not getting any output out of observations_cloud or landmarks_cloud when selecting them as PointCloud2 in RVIZ2.
When I echo those topics in the terminal, nothing is showing. If I leave it on for long enough, landmarks_cloud will eventually display one brief output of landmark points, which also show in the terminal, but they stop working after one iteration.

Could anyone help me? Thank you!

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the shown composable-node launch configuration and vslam_config.yaml, then reproduce the issue while echoing observations_cloud and landmarks_cloud and viewing them in RViz2. Check whether either topic publishes continuously with the visualization flags enabled. Done means both expected PointCloud2 topics produce ongoing landmark and observation data.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
robotics
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.