scan (sensor_msgs/LaserScan)：Laser scans to create the map from
Required tf Transforms:
<the frame attached to incoming scans> → base_link：usually a fixed value, broadcast periodically by a robot_state_publisher, or a tf .
base_link → odom：usually provided by the odometry system (e.g., the driver for the mobile base)
Provided tf Transforms:
map → odom：the current estimate of the robot's pose within the map frame
使用记录下的tf以及laser scan data构建地图的步骤如下：
1. 键盘或手柄控制机器人在空间中运动时，使用rosbag记录激光及tf数据包，记录完成后按Ctrl+C键结束。
$ rosbag record -O my_scan_data /scan /tf
2. 设置参数，确保在任何节点使用前use_sim_time参数为true。我们重播一个记录历史文件时，里面记录的是历史时间，所以我们需要告诉ROS从现在起开始启用模拟时间。This basically tells nodes on startup to use simulated time (ticked here by rosbag) instead of wall-clock time (as in a live system). It avoids confusing time-dependent components like tf, which otherwise would wonder why messages are arriving with timestamps far in the past. 关于时钟问题可以参考.
Normally, the ROS client libraries will use your computer's system clock as a time source, also known as the "wall-clock" or "wall-time" (like the clock on the wall of your lab). When you are running a simulation or playing back logged data, however, it is often desirable to instead have the system use a simulated clock so that you can have accelerated, slowed, or stepped control over your system's perceived time. For example, if you are playing back sensor data into your system, you may wish to have your time correspond to the timestamps of the sensor data.
$ rosparam set use_sim_time true
下图是use_sim_time参数为false时的情况：
设置use_sim_time为true，rosbag回放开始后ROS Time与Bag Time一致：
3. 运行slam_gmapping节点，它将在scan主题上监听激光扫描数据并创建地图（可以在命令行中设置建图参数：比如地图分辨率、粒子数目、迭代次数、地图更新间隔等参数）
$ rosrun gmapping slam_gmapping scan:=scan _xmin:=-2.5 _xmax:=2.5 _ymin:=-2.5 _ymax:=2.5 ...
比较重要的几个参数有：
slam_gmapping节点用到的参数相当多，有很多参数需要在实际中测试多次来确定其值。如果参数太多在命令行中输入会不太方便，可以写成launch文件来运行：
<launch> <arg name="scan_topic" default="scan" /> <!-- laser的topic名称，与自己的激光的topic相对应 --> <arg name="base_frame" default="base_link"/> <!-- 机器人基坐标系 --> <arg name="odom_frame" default="odom"/> <!-- 里程计坐标系 --> <arg name="map_frame" default="map" /> <!-- 地图坐标系 --> <!-- 启动slam_gmapping节点 --> <node pkg="gmapping" type="slam_gmapping" name="slam_gmapping" output="screen" clear_params="true"> <!-- clear_params: Delete all parameters in the node's private namespace before launch --> <!-- Parameters used by gmapping wrapper --> <param name="base_frame" value="$(arg base_frame)"/> <!-- The frame attached to the mobile base --> <param name="odom_frame" value="$(arg odom_frame)"/> <!-- The frame attached to the odometry system --> <param name="map_frame" value="$(arg map_frame)" /> <!-- The frame attached to the map --> <param name="map_update_interval" value="0.5"/> <!-- 地图更新时间间隔 Lowering this number updates the occupancy grid more often, at the expense of greater computational load. --> <param name="throttle_scans" value="1"/> <!-- throw away every nth laser scan. set it to a higher number to skip more scans --> <!-- Parameters used by gmapping itself --> <!-- Laser Parameters --> <param name="maxUrange" value="1.8"/> <!-- maximum range of the laser scanner that is used for map building. set maxUrange < maximum range of the real sensor <= maxRange --> <param name="maxRange" value="2.0"/> <!-- maximum range of the laser scans. Rays beyond this range get discarded completely --> <param name="sigma" value="0.05"/> <!-- standard deviation for the scan matching process --> <param name="kernelSize" value="3"/> <!-- search window for the scan matching process --> <param name="lstep" value="0.05"/> <!-- The optimization step in translation 平移优化步长 --> <param name="astep" value="0.05"/> <!-- The optimization step in rotation 旋转优化步长--> <param name="iterations" value="5"/> <!-- number of refinement steps in the scan matching 扫描匹配迭代步数--> <param name="lsigma" value="0.075"/> <!-- standard deviation for the scan matching process --> <param name="ogain" value="3.0"/> <!-- gain for smoothing the likelihood --> <param name="lskip" value="5"/> <!-- 0表示所有的激光都处理，如果计算压力过大可以将该值调大。 take only every (n+1)th laser ray for computing a match (0 = take all rays) --> <param name="minimumScore" value="80"/> <!-- 判断scanmatch是否成功的阈值，过高的话会使scanmatch失败，从而影响地图更新速率 --> <!-- Motion Model Parameters (all standard deviations of a gaussian noise model). 运动模型的噪声参数 --> <param name="srr" value="0.01"/> <!-- linear noise component (x and y) --> <param name="stt" value="0.02"/> <!-- angular noise component (theta) --> <param name="srt" value="0.02"/> <!-- linear -> angular noise component --> <param name="str" value="0.01"/> <!-- angular -> linear noise component --> <!-- Initial map dimensions and resolution --> <param name="xmin" value="-2.5"/> <!-- minimum x position in the map [m] --> <param name="xmax" value="2.5"/> <!-- maximum x position in the map [m] --> <param name="ymin" value="-2.5"/> <!-- minimum y position in the map [m] --> <param name="ymax" value="2.5"/> <!-- maximum y position in the map [m] --> <param name="delta" value="0.05"/> <!-- size of one pixel [m], 地图分辨率 --> <!-- Likelihood sampling (used in scan matching) --> <param name="llsamplerange" value="0.01"/> <!-- linear range --> <param name="lasamplerange" value="0.005"/> <!-- linear step size --> <param name="llsamplestep" value="0.01"/> <!-- linear range --> <param name="lasamplestep" value="0.005"/> <!-- angular step size --> <!-- Others --> <param name="linearUpdate" value="0.05"/> <!-- 机器人移动linearUpdate距离，进行scanmatch --> <param name="angularUpdate" value="0.0436"/> <!-- 机器人选装angularUpdate角度，进行scanmatch --> <param name="resampleThreshold" value="0.5"/> <!-- 重采样门限Neff. threshold at which the particles get resampled. Higher means more frequent resampling --> <param name="particles" value="100"/> <!-- 滤波器中粒子数目 Each particle represents a possible trajectory that the robot has traveled --> <remap from="scan" to="$(arg scan_topic)"/> </node> </launch>
4. 在新终端中启动bag包回放，将数据提供给slam_gmapping节点
$ rosbag play my_scan_data.bag