Using Moveit to just calculate the Inverse Kinematics of an arm

[Sj]

Hey,

I don't know if this is the right place to be asking this, but I hope someone can guide me with this. I have the Forward Kinematics of a 6 DOF Robotic Arm. As of now I don't have a urdf for the arm but I wish to solve and obtain the Inverse Kinematics of the arm. Is this possible in Moveit? I wish to be able to have some kind of a "log" which can help me figure out what angles are required by the arm joints to move from one location to another.

I understand this is quite an ambiguous question, which might not pertain to Moveit, but can anyone please let me know how I should proceed with this query? Would appreciate it a lot. Thanks

[Ioan Sucan]

Hello Sj,

MoveIt does include a KDL-based inverse kinematics solver, but it is useful if you have a URDF for your robot. Since you do not have that, probably using KDL directly is more suitable:
http://www.orocos.org/kdl

Hope this helps!

Ioan

[Acorn Pooley]

Yes, this is the right place to ask.  Moveit can certainly do this, but it will require a URDF before it will work.  The URDF file is how moveit knows how your arm moves -- e.g. how long each link is and what kind of joints connect the links.

Once you have a URDF creating the SRDF is pretty straightforward -- 15 minutes or so with the tutorial.

http://moveit.ros.org/wiki/Capabilities#Configuration

With that you will be able to use KDL (built into moveit) to do inverse (and forward) kinematics.  KDL is a numeric solver and there may be faster ways to do inverse kinematics (depending on how your arm works), but KDL is generic and easy to get working.

We have some tutorials on how to do inverse/fwd kinematics once you have a URDF and SRDF

http://moveit.ros.org/wiki/Capabilities#Kinematics

-Acorn

[Sj]

Oh okay. Well I was trying to find a way to just find the inverse kinematics first instead of jumping on to the URDF so that I can use it as a simple simulation, but looks like working with the URDF is probably the best way to go for. Thank you!