Connection issue with real UR5

[kirstyellis]

Hi,

We are trying to run robo-gym with a real UR5. We successfully ran the simulation beforehand, but after trying to run the gym with a real robot, we are hitting an error. This is the code that we are running:

`import gym import robo_gym from robo_gym.wrappers.exception_handling import ExceptionHandling

env = gym.make('EndEffectorPositioningURRob-v0', rs_address='127.0.0.1:50051') env = ExceptionHandling(env)

num_episodes = 10

for episode in range(num_episodes): print("start episode") done = False env.reset() while not done: state, reward, done, info = env.step(env.action_space.sample()) ` This is the error that we are getting:

env.reset() File "/home/kirstyellis/.local/lib/python3.8/site-packages/gym/wrappers/order_enforcing.py", line 18, in reset return self.env.reset(**kwargs) File "/home/kirstyellis/robogym_ws/src/robo-gym/robo_gym/envs/ur/ur_base_env.py", line 104, in reset if not self.client.set_state_msg(state_msg): File "/home/kirstyellis/.local/lib/python3.8/site-packages/robo_gym_server_modules/robot_server/client.py", line 17, in set_state_msg msg = self.robot_server_stub.SetState(state_msg, timeout = 60) File "/home/kirstyellis/.local/lib/python3.8/site-packages/grpc/_channel.py", line 946, in call return _end_unary_response_blocking(state, call, False, None) File "/home/kirstyellis/.local/lib/python3.8/site-packages/grpc/_channel.py", line 849, in _end_unary_response_blocking raise _InactiveRpcError(state) grpc._channel._InactiveRpcError: <_InactiveRpcError of RPC that terminated with: status = StatusCode.DEADLINE_EXCEEDED details = "Deadline Exceeded" debug_error_string = "UNKNOWN:Deadline Exceeded {created_time:"2022-10-26T11:29:40.308375201-04:00", grpc_status:4}"

We entered the rs_address as 127.0.0.1 and the port given by the robot server. We found that if we entered a completely random port number, we had the same output. Is there a way to determine if a connection is made with the robot server? The ur5 arm is launching correctly, we can see that in rviz the joint states are being read so I don't think it is an issue with the robot side of things.

If you could provide any suggestions as to how we can get around this issue, that would be great. Thanks!

[Adarsh3559]

Hii @kirstyellis , I am getting the exact same error, everything works fine in simulation but when trying with real UR10e I am getting the same error, Please also let me know if you come up with the solution. I would highly appreciate that.

[kirstyellis]

hi @Adarsh3559 I saw your issue, the error is slightly different to the one I am seeing.

I can reproduce your error if I change the rs_address IP to an incorrect IP.

I have my robot server running on my local machine and so am using ip:port as 127.0.0.1:50051. I changed the IP to an arbitrary value and then got the exact error that you have reported.

[kirstyellis]

@Adarsh3559 I found the problem on our setup. The controller that the joint_trajectory_command_handler.py was publishing to was different to the controller that we had loaded. By default the controller that is loaded by the ur bringup launch is scaled_pos_joint_traj_controller. The publisher in robo-gym servers was publishing to pos_traj_controller.

See this line: https://github.com/jr-robotics/robo-gym-robot-servers/blob/master/ur_robot_server/scripts/joint_trajectory_command_handler.py#L16

[Adarsh3559]

@kirstyellis , Can you help me out also, as i am new to ROS, i am also running the robot server on my local machine, but i have connected my robot via Ethernet cable and the "robot IP is static : 192.168.1.70 with subnet mask: 255.255.255.0" while i have changed my "Ubuntu IP to 192.168.1.4 with subnet mask:255.255.255.0" otherwise when i ping the robot its says host unreachable if i don't change IP on Ubuntu.

And then i start the robot server and bring up the UR10e robot i can also he UR10e arm is launching correctly, I can also see that in rviz the joint states are also being read as you mentioned above.

**I am running the below code.

%%

import gym import robo_gym

initialize environment

env = gym.make('BasicAvoidanceURRob-v0', ur_model='ur10e', rs_address='192.168.1.4:50051')

%%

num_episodes = 10

for episode in range(num_episodes): done = False env.reset() while not done:

random step in the environment

    state, reward, done, info = env.step(env.action_space.sample())

%%**

Can you tell me what changes i need to make in order to make it work with my UR10e. I would highly appreciate your help, Also, if you want we can talk via email My email is : adarsh3559@gmail.com. Please help me out as i am stuck on this problem for too long

[kirstyellis]

I would try rs_address='127.0.0.1:50051'

[Adarsh3559]

@kirstyellis I tried but same error .do i need to make changes in https://github.com/jr-robotics/robo-gym-robot-servers/blob/master/ur_robot_server/scripts/joint_trajectory_command_handler.py#L16 ?

[kirstyellis]

Is the error the same as you had before or the same as the error I had? The error you reported in the issue that you opened is different to the one I reported.

You will need to change that line in the handler if the name of the controller you are using has a different name. You can check the name of the controllers that you have by making a rosservice call to list_controllers. Then check if the name of your controller matches the one given in the link.

[KarimHP]

hat the joint_trajectory_command_handler.py was publishing to was different to the controller that we had loaded. By default the controller t @kirstyellis

So im facing the same issue. Could you elaborate what you changed? Did you change the ur5_bringup.launch file from the UR_Robot_driver? or did you change the line the script you are referring to? I know its been a while since u posted but i just saw it and i have the exact same problem so i would be really really grateful for any help!

Update: I changed the line in the trajectory handler to:

        if self.real_robot:
            self.jt_pub = rospy.Publisher('/scaled_pos_traj_controller/command', JointTrajectory, queue_size=10)
           # self.jt_pub = rospy.Publisher('/pos_traj_controller/command', JointTrajectory, queue_size=10)

but still am getting the error:

  logger.warn(
Error occurred while calling the reset function. Restarting Robot server ...
Traceback (most recent call last):
  File "/home/moga/Desktop/robo-gym/robo_gym/wrappers/exception_handling.py", line 19, in reset
    return self.env.reset(**kwargs)
  File "/home/moga/.local/lib/python3.8/site-packages/gym/wrappers/order_enforcing.py", line 42, in reset
    return self.env.reset(**kwargs)
  File "/home/moga/.local/lib/python3.8/site-packages/gym/wrappers/env_checker.py", line 45, in reset
    return env_reset_passive_checker(self.env, **kwargs)
  File "/home/moga/.local/lib/python3.8/site-packages/gym/utils/passive_env_checker.py", line 192, in env_reset_passive_checker
    result = env.reset(**kwargs)
  File "/home/moga/Desktop/robo-gym/robo_gym/envs/ur/ur_ee_positioning.py", line 236, in reset
    if not self.client.set_state_msg(state_msg):
  File "/home/moga/.local/lib/python3.8/site-packages/robo_gym_server_modules/robot_server/client.py", line 17, in set_state_msg
    msg = self.robot_server_stub.SetState(state_msg, timeout = 60)
  File "/home/moga/.local/lib/python3.8/site-packages/grpc/_channel.py", line 1030, in __call__
    return _end_unary_response_blocking(state, call, False, None)
  File "/home/moga/.local/lib/python3.8/site-packages/grpc/_channel.py", line 910, in _end_unary_response_blocking
    raise _InactiveRpcError(state)  # pytype: disable=not-instantiable
grpc._channel._InactiveRpcError: <_InactiveRpcError of RPC that terminated with:
        status = StatusCode.DEADLINE_EXCEEDED
        details = "Deadline Exceeded"
        debug_error_string = "UNKNOWN:Deadline Exceeded {created_time:"2023-03-31T17:39:31.690481676+02:00", grpc_status:4}"
>

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/moga/Desktop/start.py", line 13, in <module>
    env.reset()
  File "/home/moga/Desktop/robo-gym/robo_gym/wrappers/exception_handling.py", line 22, in reset
    self.env.restart_sim()
  File "/home/moga/.local/lib/python3.8/site-packages/gym/core.py", line 241, in __getattr__
    return getattr(self.env, name)
  File "/home/moga/.local/lib/python3.8/site-packages/gym/core.py", line 241, in __getattr__
    return getattr(self.env, name)
AttributeError: 'EndEffectorPositioningURRob' object has no attribute 'restart_sim'

[kirstyellis]

@KarimHP, which controller is being loaded? Essentially, if it is the same problem that I had, you should do a rosservice call to check which controller is loaded, that then needs to match the topic that is being published to in joint_trajectory_command_handler. You can then check the info of that topic to see that the nodes publishing and subscribing to it look correct.