interface realsense camera with rtabmap_ros package

[tanmayyb]

For wireless SLAM to take place, the RGB-D data has to go through the ROS2 interface to the appropriate RTAB-map topics to be processed. Currently I am facing bugs at each step of the process.
Once the issue is resolved, we should be able to obtain RTABmaps through the ROS2 DDS (hence wirelessly).

[tanmayyb]

Error:

Warning: Invalid frame ID "base_link" passed to canTransform argument target_frame - frame does not exist

link: ros-rtabmap-does-not-create-point-cloud
link: rtabmap_ros-with-realsense-odometry-could-not-get-transform-from-base_link-to-camera_color_optical_frame

[tanmayyb]

update:
the rgbd_odometry node expects the tf frames of id base_link. but I was sending realsense_node_link.
When setting the params camera_name: base, and base_frame_id :link for the camera node, the tf frames arrives as base_link to the rgb_odometry node. Thus solving this issue.

[tanmayyb]

code and commands for reference:
launch.py for realsense_camera_node:

import os
from ament_index_python import get_package_share_directory
from launch import LaunchDescription
from launch.actions import DeclareLaunchArgument
from launch.actions import IncludeLaunchDescription
from launch.actions import GroupAction
from launch.launch_description_sources import PythonLaunchDescriptionSource
from launch.substitutions import LaunchConfiguration
from launch.substitutions import TextSubstitution
from launch_ros.actions import Node
from launch_ros.actions import PushRosNamespace

def generate_launch_description():
    # for run file
    realsense_camera_node = Node(
            package='realsense2_camera',
            executable='realsense2_camera_node',
            name='realsense2_camera_node', #name of the node
            remappings=[
                ('/color/image_raw', '/rgb/image'),
                ('/color/camera_info', '/rgb/camera_info'),
                ('/depth/image_rect_raw', '/depth/image'),
            ],
            parameters=[
                #{"base_frame_id": "link"}, (default)
                {"camera_name": "base"},
            ]
        )
    return LaunchDescription([
        realsense_camera_node
    ])

run as launch.py and with this in local dir:

ros2 launch launch.py

For rtabmap's rgbd_odometry node:

ros2 run rtabmap_ros rgbd_odometry

[tanmayyb]

This issue existed due to incorrect usage of the namespace. The custom launchfile fixes this. <3