Cartographer源码阅读(2)：Node和MapBuilder对象

　　上文提到特别注意map_builder_bridge_.AddTrajectory(x,x)，查看其中的代码。两点：

　　首先是map_builder_.AddTrajectoryBuilder(...)，调用了map_builder_对象的方法。其次是sensor_bridges_键值对的赋值。

int MapBuilderBridge::AddTrajectory(const std::unordered_set<std::string>& expected_sensor_ids,  const TrajectoryOptions& trajectory_options)
{
     const int trajectory_id = map_builder_.AddTrajectoryBuilder(expected_sensor_ids, trajectory_options.trajectory_builder_options,                   
                  ::std::bind(&MapBuilderBridge::OnLocalSlamResult, this,
                  ::std::placeholders::_1, ::std::placeholders::_2,
                  ::std::placeholders::_3, ::std::placeholders::_4,
                  ::std::placeholders::_5));
     LOG(INFO) << "Added trajectory with ID '" << trajectory_id << "'.";
 
     // Make sure there is no trajectory with 'trajectory_id' yet.
     CHECK_EQ(sensor_bridges_.count(trajectory_id), 0);
     sensor_bridges_[trajectory_id] = cartographer::common::make_unique<SensorBridge>(
          trajectory_options.num_subdivisions_per_laser_scan,
          trajectory_options.tracking_frame,
          node_options_.lookup_transform_timeout_sec, tf_buffer_,
          map_builder_.GetTrajectoryBuilder(trajectory_id));
     auto emplace_result =  trajectory_options_.emplace(trajectory_id, trajectory_options);
     CHECK(emplace_result.second == true);
     return trajectory_id;
}

　　其中map_builder_.AddTrajectoryBuilder(...)是Cartographer项目中的代码了。

int MapBuilder::AddTrajectoryBuilder( const std::unordered_set<std::string>& expected_sensor_ids, const proto::TrajectoryBuilderOptions& trajectory_options,
 LocalSlamResultCallback local_slam_result_callback) 
{
    const int trajectory_id = trajectory_builders_.size();//生成trajectory_id
    if (options_.use_trajectory_builder_3d()) 
    {
        CHECK(trajectory_options.has_trajectory_builder_3d_options());
        trajectory_builders_.push_back(common::make_unique<CollatedTrajectoryBuilder>(
            &sensor_collator_, trajectory_id, expected_sensor_ids,
            common::make_unique<mapping::GlobalTrajectoryBuilder<
                mapping_3d::LocalTrajectoryBuilder,
                mapping_3d::proto::LocalTrajectoryBuilderOptions,
                mapping_3d::PoseGraph>>(
                trajectory_options.trajectory_builder_3d_options(),
                trajectory_id, pose_graph_3d_.get(),
                local_slam_result_callback)));//注意此处的push_back()方法
    }
    else
    {
         CHECK(trajectory_options.has_trajectory_builder_2d_options());
         trajectory_builders_.push_back(common::make_unique<CollatedTrajectoryBuilder>(
            &sensor_collator_, trajectory_id, expected_sensor_ids,
            common::make_unique<mapping::GlobalTrajectoryBuilder<
                mapping_2d::LocalTrajectoryBuilder,
                mapping_2d::proto::LocalTrajectoryBuilderOptions,
                mapping_2d::PoseGraph>>(
                trajectory_options.trajectory_builder_2d_options(),
                trajectory_id, pose_graph_2d_.get(),
                local_slam_result_callback)));//注意此处的push_back()方法
    }
    if (trajectory_options.pure_localization()) 
    {
         constexpr int kSubmapsToKeep = 3;
         pose_graph_->AddTrimmer(common::make_unique<PureLocalizationTrimmer>(trajectory_id, kSubmapsToKeep));
    }
    if (trajectory_options.has_initial_trajectory_pose())
    {
        const auto& initial_trajectory_pose = trajectory_options.initial_trajectory_pose();
        pose_graph_->SetInitialTrajectoryPose(trajectory_id, initial_trajectory_pose.to_trajectory_id(),
        transform::ToRigid3(initial_trajectory_pose.relative_pose()), common::FromUniversal(initial_trajectory_pose.timestamp()));
    }
    return trajectory_id;
}

　　注意，trajectory_builders_是根据trajectory_id添加的。以后调用的时候根据trajectory_id调用。

　　2D/3D区分：同时可以看到，这里对2D和3D情况作了区分，根据options_.use_trajectory_builder_3d()确定使用的类型。

　　在ROS的主循环运行过程中，会不断处理传感器传入的数据。

　　以IMU数据为例，auto sensor_bridge_ptr = map_builder_bridge_.sensor_bridge(trajectory_id)，根据trajectory_id获取sensor_bridge_ptr。注意这里因为是订阅的其它ROS主题（Topic），所以sensor_id参数是从其他主题传入的。（即当前程序内部有一套主题名称的字符串，订阅了外部主题也有一套名称字符串表示。这样两者通过同样的名称字符串建立了关系）

void Node::HandleImuMessage(const int trajectory_id, const std::string& sensor_id, const sensor_msgs::Imu::ConstPtr& msg)
{
  carto::common::MutexLocker lock(&mutex_);
  if (!sensor_samplers_.at(trajectory_id).imu_sampler.Pulse()) 
  {
        return;
  }
  auto sensor_bridge_ptr = map_builder_bridge_.sensor_bridge(trajectory_id);
  auto imu_data_ptr = sensor_bridge_ptr->ToImuData(msg);
  if (imu_data_ptr != nullptr) 
  {
        extrapolators_.at(trajectory_id).AddImuData(*imu_data_ptr);
  }
  sensor_bridge_ptr->HandleImuMessage(sensor_id, msg);
}

　　最后调用了sensor_bridge_ptr->HandleImuMessage(sensor_id, msg);的代码。这里又通过trajectory_builder_调用了AddSensorData方法，由于之前做为参数传入的是CollatedTrajectoryBuilder，所以实际调用的是CollatedTrajectoryBuilder的AddSensorData方法。

void SensorBridge::HandleImuMessage(const std::string& sensor_id, const sensor_msgs::Imu::ConstPtr& msg) 
{
     std::unique_ptr<::cartographer::sensor::ImuData> imu_data = ToImuData(msg);
     if (imu_data != nullptr) 
    {
            trajectory_builder_->AddSensorData( sensor_id, cartographer::sensor::ImuData{imu_data->time, imu_data->linear_acceleration, imu_data->angular_velocity});
    }
}

　　SensorBridge类实现代码，消息转换函数查看msg_conversion.cc文件：

SensorBridge::SensorBridge(
    const int num_subdivisions_per_laser_scan,
    const std::string& tracking_frame,
    const double lookup_transform_timeout_sec, tf2_ros::Buffer* const tf_buffer,
    carto::mapping::TrajectoryBuilderInterface* const trajectory_builder)
    : num_subdivisions_per_laser_scan_(num_subdivisions_per_laser_scan),
      tf_bridge_(tracking_frame, lookup_transform_timeout_sec, tf_buffer),
      trajectory_builder_(trajectory_builder) {}

std::unique_ptr<::cartographer::sensor::OdometryData>
SensorBridge::ToOdometryData(const nav_msgs::Odometry::ConstPtr& msg) {
  const carto::common::Time time = FromRos(msg->header.stamp);
  const auto sensor_to_tracking = tf_bridge_.LookupToTracking(
      time, CheckNoLeadingSlash(msg->child_frame_id));
  if (sensor_to_tracking == nullptr) {
    return nullptr;
  }
  return ::cartographer::common::make_unique<
      ::cartographer::sensor::OdometryData>(
      ::cartographer::sensor::OdometryData{
          time, ToRigid3d(msg->pose.pose) * sensor_to_tracking->inverse()});
}

void SensorBridge::HandleOdometryMessage(
    const std::string& sensor_id, const nav_msgs::Odometry::ConstPtr& msg) {
  std::unique_ptr<::cartographer::sensor::OdometryData> odometry_data =
      ToOdometryData(msg);
  if (odometry_data != nullptr) {
    trajectory_builder_->AddSensorData(
        sensor_id, cartographer::sensor::OdometryData{odometry_data->time,
                                                      odometry_data->pose});
  }
}

std::unique_ptr<::cartographer::sensor::ImuData> SensorBridge::ToImuData(
    const sensor_msgs::Imu::ConstPtr& msg) {
  CHECK_NE(msg->linear_acceleration_covariance[0], -1)
      << "Your IMU data claims to not contain linear acceleration measurements "
         "by setting linear_acceleration_covariance[0] to -1. Cartographer "
         "requires this data to work. See "
         "http://docs.ros.org/api/sensor_msgs/html/msg/Imu.html.";
  CHECK_NE(msg->angular_velocity_covariance[0], -1)
      << "Your IMU data claims to not contain angular velocity measurements "
         "by setting angular_velocity_covariance[0] to -1. Cartographer "
         "requires this data to work. See "
         "http://docs.ros.org/api/sensor_msgs/html/msg/Imu.html.";

  const carto::common::Time time = FromRos(msg->header.stamp);
  const auto sensor_to_tracking = tf_bridge_.LookupToTracking(
      time, CheckNoLeadingSlash(msg->header.frame_id));
  if (sensor_to_tracking == nullptr) {
    return nullptr;
  }
  CHECK(sensor_to_tracking->translation().norm() < 1e-5)
      << "The IMU frame must be colocated with the tracking frame. "
         "Transforming linear acceleration into the tracking frame will "
         "otherwise be imprecise.";
  return ::cartographer::common::make_unique<::cartographer::sensor::ImuData>(
      ::cartographer::sensor::ImuData{
          time,
          sensor_to_tracking->rotation() * ToEigen(msg->linear_acceleration),
          sensor_to_tracking->rotation() * ToEigen(msg->angular_velocity)});
}

void SensorBridge::HandleImuMessage(const std::string& sensor_id,
                                    const sensor_msgs::Imu::ConstPtr& msg) {
  std::unique_ptr<::cartographer::sensor::ImuData> imu_data = ToImuData(msg);
  if (imu_data != nullptr) {
    trajectory_builder_->AddSensorData(
        sensor_id, cartographer::sensor::ImuData{imu_data->time,
                                                 imu_data->linear_acceleration,
                                                 imu_data->angular_velocity});
  }
}

void SensorBridge::HandleLaserScanMessage(
    const std::string& sensor_id, const sensor_msgs::LaserScan::ConstPtr& msg) {
  ::cartographer::sensor::PointCloudWithIntensities point_cloud;
  ::cartographer::common::Time time;
  std::tie(point_cloud, time) = ToPointCloudWithIntensities(*msg);
  HandleLaserScan(sensor_id, time, msg->header.frame_id, point_cloud);
}

void SensorBridge::HandleMultiEchoLaserScanMessage(
    const std::string& sensor_id,
    const sensor_msgs::MultiEchoLaserScan::ConstPtr& msg) {
  ::cartographer::sensor::PointCloudWithIntensities point_cloud;
  ::cartographer::common::Time time;
  std::tie(point_cloud, time) = ToPointCloudWithIntensities(*msg);
  HandleLaserScan(sensor_id, time, msg->header.frame_id, point_cloud);
}

void SensorBridge::HandlePointCloud2Message(
    const std::string& sensor_id,
    const sensor_msgs::PointCloud2::ConstPtr& msg) {
  pcl::PointCloud<pcl::PointXYZ> pcl_point_cloud;
  pcl::fromROSMsg(*msg, pcl_point_cloud);
  carto::sensor::TimedPointCloud point_cloud;
  for (const auto& point : pcl_point_cloud) {
    point_cloud.emplace_back(point.x, point.y, point.z, 0.f);
  }
  HandleRangefinder(sensor_id, FromRos(msg->header.stamp), msg->header.frame_id,
                    point_cloud);
}

const TfBridge& SensorBridge::tf_bridge() const { return tf_bridge_; }

void SensorBridge::HandleLaserScan(
    const std::string& sensor_id, const carto::common::Time time,
    const std::string& frame_id,
    const carto::sensor::PointCloudWithIntensities& points) {
  CHECK_LE(points.points.back()[3], 0);
  // TODO(gaschler): Use per-point time instead of subdivisions.
  for (int i = 0; i != num_subdivisions_per_laser_scan_; ++i) {
    const size_t start_index =
        points.points.size() * i / num_subdivisions_per_laser_scan_;
    const size_t end_index =
        points.points.size() * (i + 1) / num_subdivisions_per_laser_scan_;
    carto::sensor::TimedPointCloud subdivision(
        points.points.begin() + start_index, points.points.begin() + end_index);
    if (start_index == end_index) {
      continue;
    }
    const double time_to_subdivision_end = subdivision.back()[3];
    // `subdivision_time` is the end of the measurement so sensor::Collator will
    // send all other sensor data first.
    const carto::common::Time subdivision_time =
        time + carto::common::FromSeconds(time_to_subdivision_end);
    for (auto& point : subdivision) {
      point[3] -= time_to_subdivision_end;
    }
    CHECK_EQ(subdivision.back()[3], 0);
    HandleRangefinder(sensor_id, subdivision_time, frame_id, subdivision);
  }
}

void SensorBridge::HandleRangefinder(
    const std::string& sensor_id, const carto::common::Time time,
    const std::string& frame_id, const carto::sensor::TimedPointCloud& ranges) {
  const auto sensor_to_tracking =
      tf_bridge_.LookupToTracking(time, CheckNoLeadingSlash(frame_id));
  if (sensor_to_tracking != nullptr) {
    trajectory_builder_->AddSensorData(
        sensor_id, cartographer::sensor::TimedPointCloudData{
                       time, sensor_to_tracking->translation().cast<float>(),
                       carto::sensor::TransformTimedPointCloud(
                           ranges, sensor_to_tracking->cast<float>())});
  }
}

ToPointCloudWithIntensities(const sensor_msgs::PointCloud2& message) {
  PointCloudWithIntensities point_cloud;
  // We check for intensity field here to avoid run-time warnings if we pass in
  // a PointCloud2 without intensity.
  if (PointCloud2HasField(message, "intensity")) {
    pcl::PointCloud<pcl::PointXYZI> pcl_point_cloud;
    pcl::fromROSMsg(message, pcl_point_cloud);
    for (const auto& point : pcl_point_cloud) {
      point_cloud.points.emplace_back(point.x, point.y, point.z, 0.f);
      point_cloud.intensities.push_back(point.intensity);
    }
  } else {
    pcl::PointCloud<pcl::PointXYZ> pcl_point_cloud;
    pcl::fromROSMsg(message, pcl_point_cloud);

    // If we don't have an intensity field, just copy XYZ and fill in
    // 1.0.
    for (const auto& point : pcl_point_cloud) {
      point_cloud.points.emplace_back(point.x, point.y, point.z, 0.f);
      point_cloud.intensities.push_back(1.0);
    }
  }
  return std::make_tuple(point_cloud, FromRos(message.header.stamp));
}

　　查看CollatedTrajectoryBuilder的AddSensorData方法，在CollatedTrajectoryBuilder的头文件中，包括4个覆写的AddSensorData(x,x)方法，方法中通过sensor::MakeDispatchable转换为Dispatchable<DataType>类型。

 void AddSensorData(const std::string& sensor_id, const sensor::TimedPointCloudData& timed_point_cloud_data) override
 {
       AddSensorData(sensor::MakeDispatchable(sensor_id, timed_point_cloud_data));
 }

 void AddSensorData(const std::string& sensor_id,  const sensor::ImuData& imu_data) override
 {
       AddSensorData(sensor::MakeDispatchable(sensor_id, imu_data));
 }

  void AddSensorData(const std::string& sensor_id,  const sensor::OdometryData& odometry_data) override 
{
      AddSensorData(sensor::MakeDispatchable(sensor_id, odometry_data));
}

  void AddSensorData(const std::string& sensor_id, const sensor::FixedFramePoseData& fixed_frame_pose_data) override 
{
     AddSensorData(sensor::MakeDispatchable(sensor_id, fixed_frame_pose_data));
}

　　最终定位到了sensor_collator_对象的方法。

void CollatedTrajectoryBuilder::AddSensorData( std::unique_ptr<sensor::Data> data) 
{
      sensor_collator_->AddSensorData(trajectory_id_, std::move(data));
}

　　

---

　　查看几个类CollatedTrajectoryBuilder，mapping::GlobalTrajectoryBuilder

CollatedTrajectoryBuilder::CollatedTrajectoryBuilder(sensor::Collator* const sensor_collator, const int trajectory_id,  
           const std::unordered_set<std::string>& expected_sensor_ids, std::unique_ptr<TrajectoryBuilderInterface> wrapped_trajectory_builder)
    : sensor_collator_(sensor_collator),
      trajectory_id_(trajectory_id),
      wrapped_trajectory_builder_(std::move(wrapped_trajectory_builder)),
      last_logging_time_(std::chrono::steady_clock::now())
{
      sensor_collator_->AddTrajectory(trajectory_id, expected_sensor_ids, [this](const std::string& sensor_id, std::unique_ptr<sensor::Data> data) 
     {HandleCollatedSensorData(sensor_id, std::move(data));});
}

　　mapping::GlobalTrajectoryBuilder构造函数

GlobalTrajectoryBuilder(const LocalTrajectoryBuilderOptions& options, const int trajectory_id, 
PoseGraph* const pose_graph, const LocalSlamResultCallback& local_slam_result_callback)
      : trajectory_id_(trajectory_id), pose_graph_(pose_graph),
        local_trajectory_builder_(options), local_slam_result_callback_(local_slam_result_callback) 
{}

　　注意这里的继承关系：

　　　　class CollatedTrajectoryBuilder : public TrajectoryBuilderInterface

　　　　class GlobalTrajectoryBuilder : public mapping::TrajectoryBuilderInterface

　　在mapping_2d和mapping_3d两个命名空间下分别存在2个local_trajectory_builder_类，实现了局部的扫描匹配和子图构建。代码在cartographercartographerinternal文件夹下。

　　另外一个重要的Node类变量是extrapolators_，该对象在Node类的处理Odometry和IMU数据时都有用到，作用是位姿推算。在文一种Node::AddTrajectory方法中调用了AddExtrapolator(trajectory_id, options);

std::map<int, ::cartographer::mapping::PoseExtrapolator> extrapolators_;

void Node::AddExtrapolator(const int trajectory_id, const TrajectoryOptions& options)
{
  constexpr double kExtrapolationEstimationTimeSec = 0.001;  // 1 ms
  CHECK(extrapolators_.count(trajectory_id) == 0);
  const double gravity_time_constant =
      node_options_.map_builder_options.use_trajectory_builder_3d()
          ? options.trajectory_builder_options.trajectory_builder_3d_options()
                .imu_gravity_time_constant()
          : options.trajectory_builder_options.trajectory_builder_2d_options()
                .imu_gravity_time_constant();
  extrapolators_.emplace(
      std::piecewise_construct, std::forward_as_tuple(trajectory_id),
      std::forward_as_tuple(
          ::cartographer::common::FromSeconds(kExtrapolationEstimationTimeSec),
          gravity_time_constant));
}

 map的emplace方法，高效插入。http://en.cppreference.com/w/cpp/container/map/emplace