In its default state, with the arm straight up (all joints 0), using rqt to change the value of joint 7 (the joint that attaches to the gripper) from +3 to -3 causes it to get stuck and not respond. This change to joint 7 prevents the joint from getting stuck.
This didn't seem necessary for other joints so I left them untouched, though comparing with the ur10.urdf.xacro, the ur10 has a range of max_effort values for the different joints, so there is precedence to change the others. I am going for minimal diff for the iiwa to start with.
Hopefully fixes https://bitbucket.org/osrf/ariac/issues/126
In its default state, with the arm straight up (all joints 0), using rqt to change the value of joint 7 (the joint that attaches to the gripper) from +3 to -3 causes it to get stuck and not respond. This change to joint 7 prevents the joint from getting stuck.
This didn't seem necessary for other joints so I left them untouched, though comparing with the ur10.urdf.xacro, the ur10 has a range of max_effort values for the different joints, so there is precedence to change the others. I am going for minimal diff for the iiwa to start with.
Also since the PID on the iiwa is a bit slow, the tolerances on the JointTrajectoryController have been relaxed in https://bitbucket.org/osrf/ariac/commits/3b9f46bfdcc7aac6858ffc2de2ac708a4d91e72e. An interesting thing to note is that even if the trajectory is reported as aborted, movement is not stopped: the rest of the trajectory is still executed.