[Cartographer]初始化

1.Cartographer文件树

cartographer算法包括cartographer和cartographer_ros两部分&＃xff0c;其中cartographer为算法的核心实现&＃xff0c;cartographer_ros为算法和ROS的对接的桥梁&＃xff0c;方便用户使用。

.
├── cartographer
│ ├── bazel
│ │ └── third_party
│ ├── cartographer
│ │ ├── cloud
│ │ ├── common
│ │ ├── ground_truth
│ │ ├── io
│ │ ├── mapping
│ │ ├── metrics
│ │ ├── sensor
│ │ └── transform
│ ├── cmake
│ │ └── modules
│ ├── configuration_files
│ ├── docs
│ │ ├── assets
│ │ └── source
│ └── scripts
└── cartographer_ros├── cartographer_ros│ ├── cartographer_ros│ ├── configuration_files│ ├── launch│ ├── scripts│ └── urdf├── cartographer_ros_msgs│ ├── msg│ └── srv├── cartographer_rviz│ ├── cartographer_rviz│ └── ogre_media├── docs│ └── source└── scripts


2.Cartographer初始化

1. 从src/cartographer_ros/cartographer_ros/launch/backpack_2d.launch进入&＃xff0c;可以看到cartographer_node节点的启动

<launch><param name&＃61;"robot_description"textfile&＃61;"$(find cartographer_ros)/urdf/backpack_2d.urdf" /><node name&＃61;"robot_state_publisher" pkg&＃61;"robot_state_publisher"type&＃61;"robot_state_publisher" /><node name&＃61;"cartographer_node" pkg&＃61;"cartographer_ros"type&＃61;"cartographer_node" args&＃61;"-configuration_directory $(find cartographer_ros)/configuration_files-configuration_basename backpack_2d.lua"output&＃61;"screen"><remap from&＃61;"echoes" to&＃61;"horizontal_laser_2d" />node><node name&＃61;"cartographer_occupancy_grid_node" pkg&＃61;"cartographer_ros"type&＃61;"cartographer_occupancy_grid_node" args&＃61;"-resolution 0.05" />
launch>


2. 可以发现cartographer_node是在src/cartographer_ros/cartographer_ros/cartographer_ros/node_main.cc中创建的&＃xff0c;这里主要关注函数cartographer_ros::Run()。

int main(int argc, char** argv) {// note: 初始化glog库google::InitGoogleLogging(argv[0]);// 使用gflags进行参数的初始化. 其中第三个参数为remove_flag// 如果为true, gflags会移除parse过的参数, 否则gflags就会保留这些参数, 但可能会对参数顺序进行调整.google::ParseCommandLineFlags(&argc, &argv, true);/*** &＃64;brief glog里提供的CHECK系列的宏, 检测某个表达式是否为真* 检测expression如果不为真, 则打印后面的description和栈上的信息* 然后退出程序, 出错后的处理过程和FATAL比较像.*/CHECK(!FLAGS_configuration_directory.empty())<< "-configuration_directory is missing.";CHECK(!FLAGS_configuration_basename.empty())<< "-configuration_basename is missing.";// ros节点的初始化::ros::init(argc, argv, "cartographer_node");// 一般不需要在自己的代码中显式调用// 但是若想在创建任何NodeHandle实例之前启动ROS相关的线程, 网络等, 可以显式调用该函数.::ros::start();// 使用ROS_INFO进行glog消息的输出cartographer_ros::ScopedRosLogSink ros_log_sink;// 开始运行cartographer_roscartographer_ros::Run();// 结束ROS相关的线程, 网络等::ros::shutdown();
}


3. 在cartographer_ros::Run()函数中创建了map_builder和node&＃xff0c;node控制着整个系统&＃xff0c;而map_builder如名字一样&＃xff0c;用于地图的构建&＃xff0c;另外map_builder这个变量移动到MapBuilderBridge中的std::unique_ptr map_builder_

void Run() {constexpr double kTfBufferCacheTimeInSeconds &＃61; 10.;tf2_ros::Buffer tf_buffer{::ros::Duration(kTfBufferCacheTimeInSeconds)};// 开启监听tf的独立线程tf2_ros::TransformListener tf(tf_buffer);NodeOptions node_options;TrajectoryOptions trajectory_options;// c&＃43;&＃43;11: std::tie()函数可以将变量连接到一个给定的tuple上,生成一个元素类型全是引用的tuple// 根据Lua配置文件中的内容, 为node_options, trajectory_options 赋值std::tie(node_options, trajectory_options) &＃61;LoadOptions(FLAGS_configuration_directory, FLAGS_configuration_basename);// MapBuilder类是完整的SLAM算法类// 包含前端(TrajectoryBuilders,scan to submap) 与 后端(用于查找回环的PoseGraph) auto map_builder &＃61;cartographer::mapping::CreateMapBuilder(node_options.map_builder_options);// Node类的初始化, 将ROS的topic传入SLAM, 也就是MapBuilderNode node(node_options, std::move(map_builder), &tf_buffer,FLAGS_collect_metrics);......
}


4. 从node的构造开始&＃xff0c;初始的流程如图所示