Shulou(Shulou.com)05/31 Report--

This article mainly introduces how to run the fusion obstacle visualization tool in Apollo, which has a certain reference value, and interested friends can refer to it. I hope you will gain a lot after reading this article.

Apollo created the LiDAR obstacle Visualization tool, an offline visualization tool that displays LiDAR-based obstacle perception results, see how to run Perception Visulizer offline (https://github.com/ApolloAuto/apollo/blob/master/docs/howto/how_to_run_offline_perception_visualizer_cn.md).

However, the tool lacks the ability to visualize the obstacle perception results based on radar and the fusion results based on its two sensors.

Apollo developed a second visualization tool, the Fusion obstacle Visualization tool, to complement the LiDAR obstacle Visualization tool. The Fusion obstacle Visualization tool shows the obstacle perception results of these modules:

Algorithm module based on LiDAR

Algorithm module based on radar

Fusion algorithm module, used to debug and test a complete obstacle sensing algorithm

All visualization is based on LiDAR data visualization. The source data from LiDAR (a set of 3D points used to reshape any object in the scene) is superior to radar in depicting the visual features of the entire scene. For more information about the Fusion obstacle Visualization tool, please refer to Demo Video (http://apollo.auto/platform/perception.html).

In general, you can build and run the Fusion obstacle Visualization tool in Docker in three steps:

Prepare the source data.

Build a visual tool for fusion obstacles.

Run the tool.

These three steps are described in detail below.

Prepare source data

Before you run the Fusion obstacle Visualization tool, you need to make the following preparations:

Obstacle perception based on LiDAR

Point cloud data (PCD) fil

Car posture

Obstacle perception based on Radar

Protobuf format of obstacle data acquired by Radar

Car posture

Vehicle speed

To facilitate data extraction, Apollo provides a tool called export_sensor_data to export data from ROS packages.

Steps

1. Build the data export utility with the following command:

Cd / apollobazel build / / modules/perception/tool/export_sensor_data:export_sensor_data

2. Run the data export utility with the following command:

/ apollo/bazel-bin/modules/perception/tool/export_sensor_data/export_sensor_data

3. Run ROS bag:

The default directory for ROS bag is / apollo/data/bag. The following example shows ROS bag. Exe with the file name example.bag.

Use the following command:

Cd / apollo/data/bagrosbag play-clock example.bag-rate=0.1

To ensure that no frame data is lost during the ROS message callback, it is recommended that you reduce the playback rate, which is set to 0.1 in the example above.

When you play the package, all data files are dumped frame by frame to the export directory, using a timestamp as the file name.

The default LiDAR data export directory is / apollo/data/lidar.

The radar data export directory is / apollo/data/radar.

These directories are defined in the file / apollo/modules/perception/tool/export_sensor_data/conf/export_sensor_data.flag via flag lidar_path and radar_path.

In the directory specified by lidar_path, files with two different suffixes are generated: .pcd and .pose.

In the directory specified by radar_path, files with three different suffixes are generated: .radar, .pose, and and .velocity.

Building Visualization tools for Fusion obstacles

Apollo uses Bazel to build a fusion obstacle visualization tool.

Build the blend obstacle visualization tool with the following command:

Cd / apollobazel build-c opt / / modules/perception/tool/offline_visualizer_tool:offline_sequential_obstacle_perception_test

The-c opt option is used to optimize the performance of the program being built, which is essential for the real-time visualization of offline simulation and perception modules.

(optional) if you want to run the awareness module in GPU mode, use the following command:

Bazel build-c opt-- cxxopt=-DUSE_GPU / / modules/perception/tool/offline_visualizer_tool:offline_sequential_obstacle_perception_test running tool

Before running the Fusion obstacle Visualization tool, you can set the source data directory and algorithm module settings: / apollo/modules/perception/tool/offline_visualizer_tool/conf/offline_sequential_obstacle_perception_test.flag in the configuration file.

The default source data directories lidar_path and radar_path correspond to / apollo/data/lidar and / apollo/data/radar, respectively.

The visualization-enabling Boolean flag is true, and by default, the type of obstacle result to display is fused (a fusion obstacle result based on LiDAR and RADAR sensors). You can change fused to lidar or radar to show pure obstacle results from obstacle perception based on a single sensor.

Run the Fusion obstacle Visualization tool with the following command:

/ apollo/bazel-bin/modules/perception/tool/offline_visualizer_tool/offline_sequential_obstacle_perception_test

You can see the following results:

A pop-up window showing the perceived results of the point cloud frame by frame

Origin clouds are shown in gray

Detected bounding box (with red arrow indicating title):

Vehicle (green)

Pedestrians (pink)

Bicycle (blue)

Unrecognized element (purple)

Thank you for reading this article carefully. I hope the article "how to run the Fusion Barrier Visualization tool in Apollo" shared by the editor will be helpful to you. At the same time, I also hope that you will support and pay attention to the industry information channel. More related knowledge is waiting for you to learn!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.