Joint Trajectory Action Server Error

[Fynn]

Hi all,

I was setting up a project with the sawyer robot with moveit.

In order to plan  and execute paths, I need the joint_trajectory_action_server running.

At first, everything worked like a charm.

But when a turtlebot was in the same ROS network. I first encountered the following error:

rosrun intera_interface joint_trajectory_action_server.py 
[INFO] [WallTime: 1528730135.036736] Initializing joint trajectory action server...
Traceback (most recent call last):
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/scripts/joint_trajectory_action_server.py", line 101, in <module>
    main()
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/scripts/joint_trajectory_action_server.py", line 97, in main
    start_server(args.limb, args.rate, args.mode, valid_limbs)
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/scripts/joint_trajectory_action_server.py", line 58, in start_server
    jtas.append(JointTrajectoryActionServer(limb, dyn_cfg_srv, rate, mode))
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/src/intera_joint_trajectory_action/joint_trajectory_action.py", line 64, in __init__
    self._cuff = intera_interface.Cuff(limb=limb)
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/src/intera_interface/cuff.py", line 48, in __init__
    timeout_msg=("Failed find cuff on limb '{0}'.".format(limb))
  File "/home/sawyer/ros_ws/src/intera_sdk/intera_interface/src/intera_dataflow/wait_for.py", line 42, in wait_for
    raise OSError(errno.ETIMEDOUT, timeout_msg)
OSError: [Errno 110] Failed find cuff on limb 'right'.

I am not sure if the turtlebot was the problem. But now, the action server always throws this error and is not even exiting properly. Its in an infinite loop and there is no way I can stop it except killing the process.

As far as I could tell, the action_server is waiting for a reply from the cuff state of sawyer, but this publisher is not sending any data.

 

Has anybody encountered this problem?

 

Thanks and best regards

Fynn

[Ian McMahon]

Hi Fynn,

How are you networked to your Sawyer robot? Both the physical ethernet connection, as well as the ROS environment variables from your ./intera.sh script will be useful:

$ env | grep ROS

That will hopefully determine whether you're networked to the ROS Master in your Sawyer or to the turtlebot.

Thanks!
~ Ian

 

[Fynn]

Hi Ian,

 

thanks for your quick reply. This is the ouput of env | grep ROS:

ROS_ROOT=/opt/ros/indigo/share/ros
ROS_PACKAGE_PATH=/home/sawyer/ros_ws/src:/opt/ros/indigo/share:/opt/ros/indigo/stacks
ROS_MASTER_URI=http://021606CP00055.local:11311
ROSLISP_PACKAGE_DIRECTORIES=/home/sawyer/ros_ws/devel/share/common-lisp
__ROS_PROMPT=1
ROS_DISTRO=indigo
ROS_IP=134.147.86.227
ROS_ETC_DIR=/opt/ros/indigo/etc/ros

The Sawyer and my workstation are in the same network via Ethernet cable. The turtlebot is in the network via WIFI, but I am pretty sure it is not a ROS master. If that is the only possible solution, then I will double check, but the trajectory action server still doesn't work even if the turtlebot is turned off...

 

Best,

Fynn

[Ian McMahon]

Hi Fynn,

One thing that looks unusual is your ROS_IP address: ROS_IP=134.147.86.227

Is it possible this is an external IP address, rather than a local IP? Could you change your intera.sh file to use your_hostname variable with your workstation's hostname.local as the value and then communicate with the robot? Please post your ROS environment variables after the change, I would expect to see ROS_HOSTNAME to be populated.

Thank you,
~ Ian

[Fynn]

Hi Ian,

this ROS_IP is not the issue. This is the IP of the workstation and it worked without any problems before the turtlebot connection.

just in case I changed the scripts to use the your_hostname variable, but this did not solve the issue. The joint_trajectory_action_server still throws this error message and hangs in the process.

The turtlebot also never was a ROS_Master. It is connected to the sawyer as a slave, so there are no conflicting Masters in one network.

 

Could this error originate from a hardware malfunction?

[Ian McMahon]

Hi Fynn,

Quote

Could this error originate from a hardware malfunction?

It is possible there is an underlying hardware issue. To debug this, please retrieve your logfiles by rebooting into the FSM and exporting them to a USB device, then sharing the resulting files with our Support team Support@rethinkrobotics.com :

http://sdk.rethinkrobotics.com/intera/Field_Service_Menu_(FSM)#Diagnostic_Functions

Thank you,

~ Ian