ROS turtlebot_follower 学习
首先在catkin_ws/src目录下载源码，地址：https://github.com/turtlebot/turtlebot_apps.git
了解代码见注释（其中有些地方我也不是很明白）
follower.cpp

#include <ros/ros.h>
#include <pluginlib/class_list_macros.h>
#include <nodelet/nodelet.h>
#include <geometry_msgs/Twist.h>
#include <sensor_msgs/Image.h>
#include <visualization_msgs/Marker.h>
#include <turtlebot_msgs/SetFollowState.h>

#include "dynamic_reconfigure/server.h"
#include "turtlebot_follower/FollowerConfig.h"

#include <depth_image_proc/depth_traits.h>


namespace turtlebot_follower
{

//* The turtlebot follower nodelet.
/**
 * The turtlebot follower nodelet. Subscribes to point clouds
 * from the 3dsensor, processes them, and publishes command vel
 * messages.
 */
class TurtlebotFollower : public nodelet::Nodelet
{
public:
  /*!
   * @brief The constructor for the follower.
   * Constructor for the follower.
   */
  TurtlebotFollower() : min_y_(0.1), max_y_(0.5),
                        min_x_(-0.2), max_x_(0.2),
                        max_z_(0.8), goal_z_(0.6),
                        z_scale_(1.0), x_scale_(5.0)
  {

  }

  ~TurtlebotFollower()
  {
    delete config_srv_;
  }

private:
  double min_y_; /**< The minimum y position of the points in the box. */
  double max_y_; /**< The maximum y position of the points in the box. */
  double min_x_; /**< The minimum x position of the points in the box. */
  double max_x_; /**< The maximum x position of the points in the box. */
  double max_z_; /**< The maximum z position of the points in the box. 框中 点的最大z位置，以上四个字段用来设置框的大小*/
  double goal_z_; /**< The distance away from the robot to hold the centroid 离机器人的距离，以保持质心*/
  double z_scale_; /**< The scaling factor for translational robot speed 移动机器人速度的缩放系数*/
  double x_scale_; /**< The scaling factor for rotational robot speed 旋转机器人速度的缩放系数*/
  bool   enabled_; /**< Enable/disable following; just prevents motor commands 启用/禁用追踪; 只是阻止电机命令，置为false后，机器人不会移动，/mobile_base/mobile_base_controller/cmd_vel topic 为空*/

  // Service for start/stop following
  ros::ServiceServer switch_srv_;

  // Dynamic reconfigure server 动态配置服务
  dynamic_reconfigure::Server<turtlebot_follower::FollowerConfig>* config_srv_;

  /*!
   * @brief OnInit method from node handle.
   * OnInit method from node handle. Sets up the parameters
   * and topics.
   * 初始化handle，参数，和话题
   */
  virtual void onInit()
  {
    ros::NodeHandle& nh = getNodeHandle();
    ros::NodeHandle& private_nh = getPrivateNodeHandle();
   //从参数服务器获取设置的参数（launch文件中设置数值）
    private_nh.getParam("min_y", min_y_);
    private_nh.getParam("max_y", max_y_);
    private_nh.getParam("min_x", min_x_);
    private_nh.getParam("max_x", max_x_);
    private_nh.getParam("max_z", max_z_);
    private_nh.getParam("goal_z", goal_z_);
    private_nh.getParam("z_scale", z_scale_);
    private_nh.getParam("x_scale", x_scale_);
    private_nh.getParam("enabled", enabled_);
    //设置机器人移动的话题（用于机器人移动）：/mobile_base/mobile_base_controller/cmd_vel（换成你的机器人的移动topic）
    cmdpub_ = private_nh.advertise<geometry_msgs::Twist> ("/mobile_base/mobile_base_controller/cmd_vel", 1);

    markerpub_ = private_nh.advertise<visualization_msgs::Marker>("marker",1);
    bboxpub_ = private_nh.advertise<visualization_msgs::Marker>("bbox",1);
    sub_= nh.subscribe<sensor_msgs::Image>("depth/image_rect", 1, &TurtlebotFollower::imagecb, this);

    switch_srv_ = private_nh.advertiseService("change_state", &TurtlebotFollower::changeModeSrvCb, this);

    config_srv_ = new dynamic_reconfigure::Server<turtlebot_follower::FollowerConfig>(private_nh);
    dynamic_reconfigure::Server<turtlebot_follower::FollowerConfig>::CallbackType f =
        boost::bind(&TurtlebotFollower::reconfigure, this, _1, _2);
    config_srv_->setCallback(f);
  }

//设置默认值，详见catkin_ws/devel/include/turtlrbot_follower/FollowerConfig.h
  void reconfigure(turtlebot_follower::FollowerConfig &config, uint32_t level)
  {
    min_y_ = config.min_y;
    max_y_ = config.max_y;
    min_x_ = config.min_x;
    max_x_ = config.max_x;
    max_z_ = config.max_z;
    goal_z_ = config.goal_z;
    z_scale_ = config.z_scale;
    x_scale_ = config.x_scale;
  }

  /*!
   * @brief Callback for point clouds.
   * Callback for depth images. It finds the centroid
   * of the points in a box in the center of the image. 
   * 它找到图像中心框中的点的质心
   * Publishes cmd_vel messages with the goal from the image.
   * 发布图像中目标的cmd_vel 消息
   * @param cloud The point cloud message.
   * 参数：点云的消息
   */
  void imagecb(const sensor_msgs::ImageConstPtr& depth_msg)
  {

    // Precompute the sin function for each row and column wangchao预计算每行每列的正弦函数
    uint32_t image_width = depth_msg->width;
    ROS_INFO_THROTTLE(1, "image_width=%d", image_width);
    float x_radians_per_pixel = 60.0/57.0/image_width;//每个像素的弧度
    float sin_pixel_x[image_width];
    for (int x = 0; x < image_width; ++x) {
      //求出正弦值
      sin_pixel_x[x] = sin((x - image_width/ 2.0)  * x_radians_per_pixel);
    }

    uint32_t image_height = depth_msg->height;
    float y_radians_per_pixel = 45.0/57.0/image_width;
    float sin_pixel_y[image_height];
    for (int y = 0; y < image_height; ++y) {
      // Sign opposite x for y up values 
      sin_pixel_y[y] = sin((image_height/ 2.0 - y)  * y_radians_per_pixel);
    }

    //X,Y,Z of the centroid 质心的xyz
    float x = 0.0;
    float y = 0.0;
    float z = 1e6;
    //Number of points observed 观察的点数
    unsigned int n = 0;

    //Iterate through all the points in the region and find the average of the position 迭代通过该区域的所有点，找到位置的平均值
    const float* depth_row = reinterpret_cast<const float*>(&depth_msg->data[0]);
    int row_step = depth_msg->step / sizeof(float);
    for (int v = 0; v < (int)depth_msg->height; ++v, depth_row += row_step)
    {
     for (int u = 0; u < (int)depth_msg->width; ++u)
     {
       float depth = depth_image_proc::DepthTraits<float>::toMeters(depth_row[u]);
       if (!depth_image_proc::DepthTraits<float>::valid(depth) || depth > max_z_) continue;//不是有效的深度值或者深度超出max_z_
       float y_val = sin_pixel_y[v] * depth;
       float x_val = sin_pixel_x[u] * depth;
       if ( y_val > min_y_ && y_val < max_y_ &&
            x_val > min_x_ && x_val < max_x_)
       {
         x += x_val;
         y += y_val;
         z = std::min(z, depth); //approximate depth as forward.
         n++;
       }
     }
    }

    //If there are points, find the centroid and calculate the command goal.
    //If there are no points, simply publish a stop goal.
    //如果有点，找到质心并计算命令目标。如果没有点，只需发布停止消息。
    ROS_INFO_THROTTLE(1, " n ==%d",n);
    if (n>4000)
    {
      x /= n;
      y /= n;
      if(z > max_z_){
        ROS_INFO_THROTTLE(1, "Centroid too far away %f, stopping the robot\n", z);
        if (enabled_)
        {
          cmdpub_.publish(geometry_msgs::TwistPtr(new geometry_msgs::Twist()));
        }
        return;
      }

      ROS_INFO_THROTTLE(1, "Centroid at %f %f %f with %d points", x, y, z, n);
      publishMarker(x, y, z);

      if (enabled_)
      {
        geometry_msgs::TwistPtr cmd(new geometry_msgs::Twist());
        cmd->linear.x = (z - goal_z_) * z_scale_;
        cmd->angular.z = -x * x_scale_;
        cmdpub_.publish(cmd);
      }
    }
    else
    {

      ROS_INFO_THROTTLE(1, "Not enough points(%d) detected, stopping the robot", n);
      publishMarker(x, y, z);

      if (enabled_)
      {

        cmdpub_.publish(geometry_msgs::TwistPtr(new geometry_msgs::Twist()));
      }
    }

    publishBbox();
  }

  bool changeModeSrvCb(turtlebot_msgs::SetFollowState::Request& request,
                       turtlebot_msgs::SetFollowState::Response& response)
  {
    if ((enabled_ == true) && (request.state == request.STOPPED))
    {
      ROS_INFO("Change mode service request: following stopped");
      cmdpub_.publish(geometry_msgs::TwistPtr(new geometry_msgs::Twist()));
      enabled_ = false;
    }
    else if ((enabled_ == false) && (request.state == request.FOLLOW))
    {
      ROS_INFO("Change mode service request: following (re)started");
      enabled_ = true;
    }

    response.result = response.OK;
    return true;
  }
 //画一个圆点，这个圆点代表质心
  void publishMarker(double x,double y,double z)
  {
    visualization_msgs::Marker marker;
    marker.header.frame_id = "/camera_rgb_optical_frame";
    marker.header.stamp = ros::Time();
    marker.ns = "my_namespace";
    marker.id = 0;
    marker.type = visualization_msgs::Marker::SPHERE;
    marker.action = visualization_msgs::Marker::ADD;
    marker.pose.position.x = x;
    marker.pose.position.y = y;
    marker.pose.position.z = z;
    marker.pose.orientation.x = 0.0;
    marker.pose.orientation.y = 0.0;
    marker.pose.orientation.z = 0.0;
    marker.pose.orientation.w = 1.0;
    marker.scale.x = 0.1;
    marker.scale.y = 0.1;
    marker.scale.z = 0.1;
    marker.color.a = 1.0;
    marker.color.r = 1.0;
    marker.color.g = 0.0;
    marker.color.b = 0.0;
    //only if using a MESH_RESOURCE marker type:
    markerpub_.publish( marker );
  }
 //画一个立方体，这个立方体代表感兴趣区域（RIO）
  void publishBbox()
  {
    double x = (min_x_ + max_x_)/2;
    double y = (min_y_ + max_y_)/2;
    double z = (0 + max_z_)/2;

    double scale_x = (max_x_ - x)*2;
    double scale_y = (max_y_ - y)*2;
    double scale_z = (max_z_ - z)*2;

    visualization_msgs::Marker marker;
    marker.header.frame_id = "/camera_rgb_optical_frame";
    marker.header.stamp = ros::Time();
    marker.ns = "my_namespace";
    marker.id = 1;
    marker.type = visualization_msgs::Marker::CUBE;
    marker.action = visualization_msgs::Marker::ADD;
    //设置标记物姿态
    marker.pose.position.x = x;
    marker.pose.position.y = -y;
    marker.pose.position.z = z;
    marker.pose.orientation.x = 0.0;
    marker.pose.orientation.y = 0.0;
    marker.pose.orientation.z = 0.0;
    marker.pose.orientation.w = 1.0;
    //设置标记物的尺寸大小
    marker.scale.x = scale_x;
    marker.scale.y = scale_y;
    marker.scale.z = scale_z;

    marker.color.a = 0.5;
    marker.color.r = 0.0;
    marker.color.g = 1.0;
    marker.color.b = 0.0;
    //only if using a MESH_RESOURCE marker type:
    bboxpub_.publish( marker );
  }

  ros::Subscriber sub_;
  ros::Publisher cmdpub_;
  ros::Publisher markerpub_;
  ros::Publisher bboxpub_;
};

PLUGINLIB_DECLARE_CLASS(turtlebot_follower, TurtlebotFollower, turtlebot_follower::TurtlebotFollower, nodelet::Nodelet);

}

接下来看launch文件follower.launch
建议在修改前，将原先的代码注释掉，不要删掉。

<!--
  The turtlebot people (or whatever) follower nodelet.
 -->
<launch>
  <arg name="simulation" default="false"/>
  <group unless="$(arg simulation)"> <!-- Real robot -->
    <include file="$(find turtlebot_follower)/launch/includes/velocity_smoother.launch.xml">
      <arg name="nodelet_manager"  value="/mobile_base_nodelet_manager"/>
      <arg name="navigation_topic" value="/cmd_vel_mux/input/navi"/>
    </include>
    <!--modify by 2016.11.07 启动我的机器人和摄像头，这里更换成你的机器人的启动文件和摄像头启动文件-->
    <include file="$(find handsfree_hw)/launch/handsfree_hw.launch">
    </include>
    <include file="$(find handsfree_bringup)/launch/xtion_fake_laser_openni2.launch">
    </include>


   <!-- 将原先的注释掉<include file="$(find turtlebot_bringup)/launch/3dsensor.launch">
      <arg name="rgb_processing"                  value="http://www.mamicode.com/true"/> 
      <arg name="depth_processing"                value="http://www.mamicode.com/true"/>
      <arg name="depth_registered_processing"     value="http://www.mamicode.com/false"/>
      <arg name="depth_registration"              value="http://www.mamicode.com/false"/>
      <arg name="disparity_processing"            value="http://www.mamicode.com/false"/>
      <arg name="disparity_registered_processing" value="http://www.mamicode.com/false"/>
      <arg name="scan_processing"                 value="http://www.mamicode.com/false"/>
    </include>-->
  <!--modify end -->
  </group>
  <group if="$(arg simulation)">
    <!-- Load nodelet manager for compatibility -->
    <node pkg="nodelet" type="nodelet" ns="camera" name="camera_nodelet_manager" args="manager"/>

    <include file="$(find turtlebot_follower)/launch/includes/velocity_smoother.launch.xml">
      <arg name="nodelet_manager"  value="camera/camera_nodelet_manager"/>
      <arg name="navigation_topic" value="cmd_vel_mux/input/navi"/>
    </include>
  </group>

  <param name="camera/rgb/image_color/compressed/jpeg_quality" value="22"/>

  <!-- Make a slower camera feed available; only required if we use android client -->
  <node pkg="topic_tools" type="throttle" name="camera_throttle"
        args="messages camera/rgb/image_color/compressed 5"/>

  <include file="$(find turtlebot_follower)/launch/includes/safety_controller.launch.xml"/>

  <!--  Real robot: Load turtlebot follower into the 3d sensors nodelet manager to avoid pointcloud serializing -->
  <!--  Simulation: Load turtlebot follower into nodelet manager for compatibility -->
  <node pkg="nodelet" type="nodelet" name="turtlebot_follower"
        args="load turtlebot_follower/TurtlebotFollower camera/camera_nodelet_manager">
     <!--更换成你的机器人的移动topic，我的是/mobile_base/mobile_base_controller/cmd_vel-->
    <remap from="turtlebot_follower/cmd_vel" to="/mobile_base/mobile_base_controller/cmd_vel"/>
    <remap from="depth/points" to="camera/depth/points"/>
    <param name="enabled" value="true" />
    <!--<param name="x_scale" value="http://www.mamicode.com/7.0" />-->
    <!--<param name="z_scale" value="http://www.mamicode.com/2.0" />
    <param name="min_x" value="http://www.mamicode.com/-0.35" />
    <param name="max_x" value="http://www.mamicode.com/0.35" />
    <param name="min_y" value="http://www.mamicode.com/0.1" />
    <param name="max_y" value="http://www.mamicode.com/0.6" />
    <param name="max_z" value="http://www.mamicode.com/1.2" />
    <param name="goal_z" value="http://www.mamicode.com/0.6" />-->


    <!-- test  可以在此处调节参数-->
    <param name="x_scale" value="1.5"/>
    <param name="z_scale" value="1.0" />
    <param name="min_x" value="-0.35" />
    <param name="max_x" value="0.35" />
    <param name="min_y" value="0.1" />
    <param name="max_y" value="0.5" />
    <param name="max_z" value="1.5" />
    <param name="goal_z" value="0.6" />
  </node>
  <!-- Launch the script which will toggle turtlebot following on and off based on a joystick button. default: on -->
  <node name="switch" pkg="turtlebot_follower" type="switch.py"/>
 <!--modify  2016.11.07 在turtlebot_follower下新建follow.rviz文件，加载rviz，此时rviz内容为空-->
  <node name="rviz" pkg="rviz" type="rviz" args="-d $(find turtlebot_follower)/follow.rviz"/>
<!--modify end -->
</launch>

编译，运行follow.launch 会将机器人和摄像头，rviz都启动起来，只需要运行这一个launch就可以了。
rviz中添加两个marker，pointcloud，camera。如图：

topic与frame名称与代码中要保持一致。
添加完之后，rviz显示如图：

红点代表质心，绿框代表感兴趣区域
当红点在我们身上时，机器人会跟随我们运动，注意：走动时，我们的速度要慢一点，机器人的移动速度也要调慢一点。
当感兴趣区域没有红点时，机器人停止跟随，直到出现红点。