Closed rafaelBauer closed 1 month ago
Finally got it working. My problem was my control mode. I changed to pd_joint_pos
and it worked perfectly :)
EDIT: Finally got it working. I had two problems.
The target pose was likely leading to a collision, since I was trying to go to [0, 0, 0.5]
and the reference is the robot base link. So I changed the target pose to Pose([0.615, 0, 0.5][0, 1, 0, 0])
and the planning was working.
Then I had a problem, that the actions were not really controlling the robot in a desired way. My actions are coming only from the position of the planning. So I changed the ManiSkill control_mode to pd_joint_pos
and it worked perfectly :)
Thanks for the follow-up! "IK Failed! Cannot find valid solution" can also often be triggered by collision. I am not sure why the control mode in maniskill should have anything to do with IK in mplib.
Oh no, I also changed the pose. I updated my comment from before just to keep this complete. Thanks for the reply!
Hello, this is not a bug, but more a call for help :)
I am new into robot control and looked through the examples and I have been trying to create a motion plan, but I cannot really make it work.
My setup is a bit different. I have a panda robot and I am using ManiSkill. But somehow I am struggling to make a simple trajectory.
My move_group is the
panda_hand_tcp
and initially its pose isPose([0.00369545, 0.029311, 0.184837], [0.00760545, 0.999395, 0.0339194, 0.000981587])
(I got this information from SAPIEN Viewer). I am trying to make a planning toPose([0, 0, 0.05], [0, 1, 0, 0])
, but I am constantly getting this inverse kinematics error.. I am not sure if there is any configuration I am doing wrong.My ManiSkill environment control_mode is
pd_ee_delta_pose
.I appreciate if anyone has a tip on how I could proceed, or debug what is my problem :)