Ros 学习（5）

https://github.com/LCAS/CMP9767M/wiki/Workshop-6---Localisation

Task1

按照上一个的内容读取bag，并在rviz中显示，add topic odometry/base_raw and odometry/filtered， 注意去掉rviz中的Covariance的选项，显示更加清晰

Adjust the display parameters for the odometry messages including arrow size and colour, tick off the covariance box and keep 1000 past readings to see the full trajectory.

If you are not confident with setting up rviz visualisations, you can alternatively load a configuration file which sets up the trajectory visualisation for you: uol_cmp9767m_tutorial/config/ekf.rviz #（停止bag运行，加载ekf.rviz），之后重新运行bag，此时rviz中的parameter已经是ekf.rviz预设置好的了。

Inspect the config file config/odom_imu_ekf.yaml and check the filter parameters. Read details on setting up the ekf node （理解不同参数的设置对定位的影响）and try different configurations for the IMU by switching on/off relevant parameters. How do they influence the results? 

Task2

roslaunch bacchus_gazebo vineyard_demo.launch fake_localisation:=false

roslaunch uol_cmp9767m_tutorial amcl.launch

#机器人不显示，且显示warn--No laser scan received

[ WARN] [1637683136.309457011, 95.710000000]: No laser scan received (and thus no pose updates have been published) for 95.710000 seconds.  Verify that data is being published on the /thorvald_001/scan topic.
 

#修复：

roscd uol_cmp9767m_tutorial/launch/ #(/opt/ros/melodic/share/uol_cmp9767m_tutorial/launch)

sudo gedit amcl.launch #修改

#<remap from="scan" to="$(arg robot_name)/scan" />     #为下面一行

#<remap from="scan" to="$(arg robot_name)/front_scan" />

In rviz add the following visualisation: particecloud。

键盘控制机器人：

Now, teleoperate the robot using keyboard (e.g. rosrun key_teleop key_teleop.py key_vel:=/thorvald_001/teleop_joy/cmd_vel)

Try out other critical parameters of the amcl node (number of particles (min, max), resample interval) and observe the difference in the results.#（打开roscd uol_cmp9767m_tutorial/launch/中的amcl.launch进行设置）