perception-pointcloud-preprocess

Introduction

The point cloud preprocessing module preprocesses the point cloud data output by the driver. Deleting nan value points, points that are too far away, points scanned onto the self vehicle, and point that are too high.

Directory Structure

├── pointcloud_preprocess  // point cloud preprocess component
    ├── conf               // configuration folder
    ├── dag                // module startup file
    ├── data               // module configuration parameters
    ├── launch             // launch file
    ├── interface          // preprocess interface
    ├── proto              // preprocess module configuration proto file
    ├── preprocessor       // preprocess method
    ├── pointcloud_preprocess_component.cc // component entrance
    ├── pointcloud_preprocess_component.h
    ├── cyberfile.xml      // package management file
    ├── README.md          // readme file
    └── BUILD              // compile file

Modules

PointCloudPreprocessComponent

apollo::perception::lidar::PointCloudPreprocessComponent

Input

Point cloud data from driver: If there is one lidar, output point cloud after motion compensation. If there are multiple lidars, concatenate the point clouds into one frame after motion compensation. The default trigger channel is /apollo/sensor/velodyne64/compensator/PointCloud2. The detailed input channel information is in modules/perception/pointcloud_preprocess/dag/pointcloud_preprocess.dag file.

Output

Note: The output channel is structure type data. The message is defined in the modules/perception/common/onboard/inner_component_messages/lidar_inner_component_messages.h file. The output channel message data can be subscribed by components in the same process. The detailed output channel information is in modules/perception/pointcloud_preprocess/conf/pointcloud_preprocess_config.pb.txt file.

How to Launch

1. Add vehicle parameter configuration file to modules/perception/data/params, corresponding frame_id and sensor_name, launch transform

cyber_launch start modules/transform/launch/static_transform.launch

2. Modify modules/perception/launch/perception_lidar.launch

• select the dag file to start, use modules/perception/pointcloud_preprocess/dag/pointcloud_preprocess.dag here
• modify msg_adapter. It is used to wrap messages sent by other steps as /apollo/perception/obstacles, this can be used for individual debugging. Modify relevant channel configurations in modules/perception/data/flag/perception_common.flag

3. Modify parameters of modules/perception/pointcloud_preprocess/conf/pointcloud_preprocess_config.pb.txt

• sensor_name: sensor name
• lidar_query_tf_offset: tf time offset
• output_channel_name: output channel name
• plugin_param: plugin parameters
  • name: method name
  • config_path: configuration file path
  • config_file: configuration file name

4. Launch point cloud preprocess component

cyber_launch start modules/perception/pointcloud_preprocess/launch/pointcloud_preprocess.launch