| Description | Username | Hostname (Computer Name) | IP | Password | OS | ROS |
|---|---|---|---|---|---|---|
| Stretch Robot (Yellow) | hello-robot | stretch-re1-1027 | 192.168.1.64 | hello2020 | Ubuntu 18.04 | Melodic |
| Stretch Robot (Grey) | hello-robot | stretch-re1-1028 | 192.168.1.28 | hello2020 | Ubuntu 18.04 | Melodic |
The commonly used commands after sshing into the robots are given below.
| Tool | Utility |
|---|---|
| stretch_robot_home.py | Commonly run after booting up the robot in-order to calibrate the joints |
| stretch_robot_system_check.py | Scans for all hardware devices and ensure they are present on the bus and reporting valid values. Useful to verify that the robot is in good working order prior to commanding motion. It will report all success in green, failures in red. |
| stretch_robot_stow.py | Useful to return the robot arm and tool to a safe position within the base footprint. It can also be useful if a program fails to exit cleanly and the robot joints are not backdriveable. It will restore them to their 'Safety' state. |
| stretch_robot_battery_check.py | Quick way to check the battery voltage / current consumption |
| stretch_xbox_controller_teleop.py | Useful to quickly test if a robot can achieve a task by manually teleoperating the robot |
| stretch_robot_dynamixel_reboot.py | This will reset all Dynamixels in the robot, which may be needed if a servo overheats during high use and enters an error state. |
Note: The table above is copied from here
Gazebo Lab world is here. Clone it to your catkin_ws/src computer, execute catkin_make from your catkin_ws and add the following lines to ~/.bashrc and then source.
export GAZEBO_MODEL_PATH=~/catkin_ws/src/AssistiveRobot-SimulationFiles/lab_gazebo/models
export GAZEBO_RESOURCE_PATH=~/catkin_ws/src/AssistiveRobot-SimulationFiles/lab_gazebo/worlds
Then, follow the instructions to setup the stretch gazebo as described here. (You may also want to see)
To run the gazebo simulation, I slightly edited default gazebo.launch file to make the gazebo world as lab instead of empty world and also added position parameters to specify the initial position of the stretch robot in the lab. The new launch file is here. Copy it into stretch_ros/stretch_gazebo/launch/ next to gazebo.launch file. After that, you can run
roslaunch stretch_gazebo gazebo_stretch_lab.launch rviz:=true
Now you should be able to see Gazebo and Rviz with a Stretch Robot in the lab world.
If you will open Rviz with another launch use:
roslaunch stretch_gazebo gazebo_stretch_lab.launch rviz:=false
Note: you can use killall gzserver and killall gzclient commands to shutdown Gazebo.
With the real stretch helloe robots one would run roslaunch stretch_navigation mapping.launch as described in here to start 2D mapping. This launch file launches stretch_driver and lidar sensors from stretch_core package. It also launches keyboard teleoperation with topic name /stretch/cmd_vel, a configured rviz to visualize the map, and finally launches gmapping package for the mapping.
For the Gazebo simulation, however, we don't need to launch stretch_driver and lidar sensors from stretch_core package because,although differently, they are already provided with previously mentioned Gazebo package launching command roslaunch stretch_gazebo gazebo_stretch_lab.launch rviz:=false (ie. /scan topic by lidar sensor is already published by gazebo and /stretch_diff_drive_controller/cmd_vel topic is listened to move the base, instead of /stretch/cmd_vel, also /tf is also provided by gazebo which means there is no need for stretch_core package).
Based on explanation above, I edited the mapping.launch file as mapping_gazebo_2D.launch that can be found in here. Copy it into stretch_ros/stretch_navigation/launch/ next to mapping.launch file. After that, you can run:
roslaunch stretch_navigation mapping_gazebo_2D.launch
Now you can start 2D mapping in simulation with your keyboard input. (You may need to install ROS teleop_twist_keyboard package with command sudo apt-get install ros-melodic-teleop-twist-keyboard)
Note: How to save the constructed map is described in here. For example: I created an example 2D map of the simulated lab in gazebo with command rosrun map_server map_saver -f ./maps/lab2d in maps directory of this repo. Then I also had to edit the generetad lab2d.yaml file with replacing the line image: ./maps/lab2d.pgm to image: ./lab2d.pgm
(You may need to install ROS AMCL localization package with command sudo apt-get install ros-melodic-amcl)
Similar to the previous section, with the real stretch hello robots one would run roslaunch stretch_navigation navigation.launch map_yaml:=<path-to-maps>/maps/<map_name>.yaml as described in here to start 2D navigation. However, we will run:
roslaunch stretch_navigation navigation_gazebo_2D.launch map_yaml:=<path-to-maps>/maps/<map_name>.yaml
after copying navigation_gazebo_2D.launch into stretch_ros/stretch_navigation/launch/ next to navigation.launch file.
(For example, I run roslaunch stretch_navigation navigation_gazebo_2D.launch map_yaml:=/home/burak/common/RESEARCH/HelloStretchRobotFiles/HelloStretchSandbox/maps/lab2d.yaml)
Then, as explained in here,
-
In the top bar of the opened Rviz window, use 2D Pose Estimate to lay an arrow down roughly where the robot is located in the real space. AMCL, the localization package, will better localize our pose once we give the robot a 2D Nav Goal. In the top bar of Rviz, use 2D Nav Goal to lay down an arrow where you'd like the robot to go. In the terminal, you'll see move_base go through the planning phases and then navigate the robot to the goal. If planning fails, the robot will begin a recovery behavior: spinning around 360 degrees in place.
-
It is also possible to send 2D Pose Estimates and Nav Goals programatically. In your own launch file, you may include navigation.launch to bring up the navigation stack. Then, you can send move_base_msgs::MoveBaseGoal messages in order to navigate the robot programatically.