Map visualization

Adding topics in RViz

Start RViz in a new terminal:

rviz

On the Displays pane on the left side, click on Add. (If you don't have the Displays pane, you can get it via Panels -> Displays)

Click on the tab By topic and select a topic to visualize and click OK on the bottom right:

Note: The topics only appear in RViz after the map has been visualized from the maplab console. It means you might need to start maplab, load a map and call v.

Understanding visualization

Start the maplab console:

rosrun maplab_console maplab_console

Load a map, and visualize it:

load --map_folder MH_01_easy  # or any other VI map you have.
v

This will publish the following topics to RViz:

• vi_map_baseframe: Baseframes of the individual missions of the maps.
• vi_map_edges/viwls: Edges of the VI map. Visualization of edges:
• vi_map_vertices: Vertex transformations. Visualzation of edges and vertices:
• vi_map_landmarks: Point cloud of all landmarks. Visualization of edges, vertices and landmarks:

Loop closing a map (with the lc command) adds the visualization topic loop_closures, which will show merged landmark pairs:

Running relax adds the topic vi_map_edges/loop_closure to the list and displays loop closure edges between vertices that can be loop-closed:

Additional flags for visualization

• vis_scale: Sets the size of the edges.

• vis_color_by_mission: colors the landmarks and edges of each mission differently. This is enabled by default.

• vis_color_landmarks_by_height: Colors the landmarks by height. Use with these additional flags:

  • vis_color_by_height_period_m
  • vis_color_by_height_offset_m

  

• vis_color_salt: Changes the rotation of the colors, e.g.

Salt 1 (default value):

v --vis_color_by_mission --vis_color_salt 1

Salt 70:

v --vis_color_by_mission --vis_color_salt 70

Visualizing the inertial states

You can use

plot_vi_states_of_mission

to plot the VI states of a mission.

This will open a plot with the position, velocity, and accelerometer/gyroscope biases. This may be helpful for tuning the IMU parameters or better understanding what happens during the batch optimization.