Usage

1. Necessary Files

You will need a map file in .pcd format containing the 3D point cloud data of the map.

You will also need a separate map information file in .yaml format. This contains the location of the point cloud map origin in geocentric coordinates.

The map information file (YAML file) should have the following format:

# example_map.yaml
---
map_config:
    latitude: 37.380811523812845
    longitude: -121.90840595108715
    elevation: 16.0
    roll: 0.0
    pitch: 0.0
    yaw: 0.0

where latitude, longitude and elevation are the WGS84 coordinates of the origin of the point cloud map, and roll, pitch and yaw describe the orientation of the map reference frame.

2. Create/edit a parameter file to reference the map YAML file

An example parameter file for the map_publisher node can be found in the following location (inside the ade environment):

/opt/AutowareAuto/share/ndt_nodes/share/param/map_publisher.param.yaml

Alternatively if building from source code the example parameter file will be located here:

~/AutowareAuto/install/ndt_nodes/share/ndt_nodes/param/map_publisher.param.yaml

Copy the file to another location and edit (and uncomment) the map_pcd_file and the map_yaml_file parameters of the file map_publisher.param.yaml to refer to the point cloud map and the YAML map information files:

map_pcd_file: "/path/to/map_data.pcd"

map_yaml_file: "/path/to/map_info.yaml"

Be carefull to enter appropriate values for map parameters such as minimum and maximum points.

3. Run the map_publisher node

Inside the ade, source the workspace. If you are using precompiled version source from /opt/AutowareAuto/setup.bash. if you are developing and have built from source code, source the from ~/AutowareAuto/install/setup.bash

Run the node with the paramter file as an argument:

ros2 run ndt_nodes ndt_map_publisher_exe --ros-args --params-file path/to/map_publisher.param.yaml

Alternatively use a launch file: ros2 launch ndt_nodes map_provider.launch.py

which launches the map provider node with the default parameter values in map_publisher.param.yaml, and also launches a voxel_grid_node to sub-sample the point cloud for visualization.

4. Output

By default (as defined in the parameter file), the map_provider node will publish the following information

the ndt map on the topic ndt_map
the point cloud map visualization on the topic viz_ndt_map
if a voxel_grid_node is used to sub-sample the point cloud as in the launch file, it will be published on the topic viz_map_subsampled

Configuration of map provider node

The configurable parameters passed by the parameter file map_publisher.param.yaml of the node are:

Parameter Name	Usage
map_pcd_file	file name of point cloud data file
map_yaml_file	file name of map information file
map_frame	frame for map point coordinates
map_config.capacity	max ndt map voxel capacity
map_config.min_point.x	min x bound for ndt map data
map_config.min_point.y	min y bound for ndt map data
map_config.min_point.z	min z bound for ndt map data
map_config.max_point.x	max x bound for ndt map data
map_config.max_point.y	max y bound for ndt map data
map_config.max_point.z	max z bound for ndt map data
map_config.voxel_size.x	voxel x size for ndt map data
map_config.voxel_size.y	voxel y size for ndt map data
map_config.voxel_size.z	voxel z size for ndt map data
viz_map	flag for whether to publish map for visualization

A voxel_grid_node is used to down sample the point cloud for visualization. An example parameter file for the voxel grid is given at

/opt/AutowareAuto/share/ndt_nodes/share/param/pcl_map_voxel_grid_downsample.param.yaml: