Closed watakandai closed 4 years ago
Is this for franka or have to be done when doing simulation too?
for both.
I assume JointTrajectory
may not ensure that it has reached to the assigned goal.
Maybe it does. I'm not sure.
We do need to verify that the state, let's saytheta
has reached theta_{command}
.
In order to do that we need to subscribe to the current position theta
Ahh, now I get it, that can be easily be done
yep, thanks!
Is this still relevant @watakandai ?
No, but we need to test if the controller can reach the desired position in a certain time. For panda, I found out that the PID controller implemented by panda_simulation is a shit. We need to tune the PID parameters and test them in the future
Okay I guess that tuning is a project in itself. Shall we make a different issue about it in the kanban board?
Yes please
can we close this issue @watakandai ?
👍
cool
https://github.com/HIRO-group/ros_robotic_skin/blob/master/scripts/joint_movement.py
rad
. Instead of commanding1
, it might be better to commandpi/4
, so that the angle rotates 45 degrees.As @calebescobedo has commented, we do need to verify that the joint has reached the goal. I guess we need to subscribe to the current joint position and compute the error between the current joint position and the commanded joint position.