Orb_slam_2 with the Intel RealSense T265

[rakshith95]

Hello, I try to use the orb_slam_2_ros_stereo node with the Intel T265, but I find that it does not really work unless I also give it a depth/image_raw input, (which the T265 does not have). I'm not sure why this is the case since the stereo node seems to only need the left, and right image topics.

My launch file:

`

launch

arg name="map_path" default="$(env HOME)/.pal/ari_maps/config"/> arg name="camera_model" default="realsense_t265" /> arg name="config_path" default="/home/pal/.pal/custom_configs"/>

node name="loc_orb_back" pkg="orb_slam2_ros" type="orb_slam2_ros_stereo" output="screen">

   <remap from="image_left/image_color_rect" to="/torso_back_camera/fisheye1/image_rect" />
   <remap from="image_right/image_color_rect" to="/torso_back_camera/fisheye2/image_rect" />
   <!-- <remap from="depth/image_raw" to="/torso_front_camera/aligned_depth_to_color/image_raw" /> -->

   <param name="publish_pointcloud" type="bool" value="true" />
   <param name="publish_pose" type="bool" value="true" />
   <param name="localize_only" type="bool" value="true" />
   <param name="reset_map" type="bool" value="false" />

   <!-- static parameters -->
   <param name="load_map" type="bool" value="true" />
   <param name="map_file" type="string" value="$(arg map_path)/pal.bin" />
   <param name="settings_file" type="string" value="$(arg config_path)/$(arg camera_model).yaml" />
   <param name="voc_file" type="string" value="$(find orb_slam2_ros)/orb_slam2/Vocabulary/ORBvoc.txt" />

   <param name="pointcloud_frame_id" type="string" value="back_orb_map" />
   <param name="camera_frame_id" type="string" value="tracked_camera_link_back" />
   <param name="min_num_kf_in_map" type="int" value="5" />

   <param name="correct_global_frame" type="bool" value="true" />

   <param name="minimum_travel_distance" type="double" value="0.2" />
   <param name="minimum_travel_heading" type="double" value="0.017" />
   <param name="transform_tolerance" type="double" value="0.25" />

`

And the settings yaml file: `%YAML:1.0

#--------------------------------------------------------------------------------------------

Camera Parameters. Adjust them!

#--------------------------------------------------------------------------------------------

Camera calibration and distortion parameters (OpenCV)

Camera.fx: 285.6816101074219 Camera.fy: 285.5213928222656 Camera.cx: 415.45660400390625 Camera.cy: 400.83740234375

Camera.k1: -0.003609714098274708 Camera.k2: 0.033376969397068024 Camera.p1: -0.030752409249544144 Camera.p2: 0.0039985291659832 Camera.k3: 0.0

Camera.width: 848 Camera.height: 800

Camera frames per second

Camera.fps: 30.0

stereo baseline times fx

Camera.bf: 18.24

Close/Far threshold. Baseline times.

ThDepth: 35

Deptmap values factor

DepthMapFactor: 1000.0

#--------------------------------------------------------------------------------------------

Stereo Rectification. Only if you need to pre-rectify the images.

Camera.fx, .fy, etc must be the same as in LEFT.P

#-------------------------------------------------------------------------------------------- LEFT.height: 848 LEFT.width: 800 LEFT.D: !!opencv-matrix rows: 1 cols: 5 dt: d data: [-0.003609714098274708, 0.033376969397068024, -0.030752409249544144, 0.0039985291659832, 0.0] LEFT.K: !!opencv-matrix rows: 3 cols: 3 dt: d data: [285.6816101074219, 0.0, 415.45660400390625, 0.0, 285.5213928222656, 400.83740234375, 0.0, 0.0, 1.0] LEFT.R: !!opencv-matrix rows: 3 cols: 3 dt: d data: [1, 0, 0, 0, 1, 0, 0, 0, 1] LEFT.P: !!opencv-matrix rows: 3 cols: 4 dt: d data: [285.6816101074219, 0.0, 415.45660400390625, 0.0, 0.0, 285.5213928222656, 400.83740234375, 0.0, 0.0, 0.0, 1.0, 0.0]

RIGHT.height: 800 RIGHT.width: 848 RIGHT.D: !!opencv-matrix rows: 1 cols: 5 dt: d data: [-0.0068389722146093845, 0.04215899854898453, -0.03935682028532028, 0.006755772978067398, 0.0]

RIGHT.K: !!opencv-matrix rows: 3 cols: 3 dt: d data: [285.343505859375, 0.0, 416.8861083984375, 0.0, 285.3443908691406, 404.5669860839844, 0.0, 0.0, 1.0]

RIGHT.R: !!opencv-matrix rows: 3 cols: 3 dt: d data: [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0]

RIGHT.P: !!opencv-matrix rows: 3 cols: 4 dt: d data: [285.343505859375, 0.0, 416.8861083984375, -18.261984375, 0, 285.3443908691406, 404.5669860839844, 0.0, 0.0, 0.0, 1.0, 0.0]

#--------------------------------------------------------------------------------------------

ORB Parameters

#--------------------------------------------------------------------------------------------

ORB Extractor: Number of features per image

ORBextractor.nFeatures: 1200

ORBextractor.scaleFactor: 1.2

ORBextractor.nLevels: 8

ORB Extractor: Fast threshold

Image is divided in a grid. At each cell FAST are extracted imposing a minimum response.

Firstly we impose iniThFAST. If no corners are detected we impose a lower value minThFAST

ORBextractor.iniThFAST: 20 ORBextractor.minThFAST: 7

`