Kinematic solvers


I have some problems related with this course. I understand the course and I think that i have some good experience in working with robotic arms and its kinematics, but I don’t find the relation between creating our own iksolver and working with the service compute_ik and compute_fk. ¿How can I implement my own kinematics to a robot having the solver of my robot? I think that this course must be more complete about working with our solvers combined with moveit.

Hello Lucas,

Thanks for your feedback. This course still has some improvements to be done on it. As per your suggestion, we were thinking of adding this topic (using your own IK solver in MoveIt) in another course we are planning to publish which will be called Advanced MoveIt. However, it might also be a good idea to add this to the Basic Kinematics course. We will definitely consider your suggestion.


Ok, that’s good, but i still want to know how to implement inverse kinematics to my own robot. Can you give me some tips or some suggestions on how to start doing it? I have been searching it but i don’t find the info. If you can help me it would be great.

Hello @LucasReyBraga,

One of the most extended solutions (and better documented) is to use IKFast. It is pretty easy to set up and works for almost any kind of manipulator. You can have a look at the following documentation on how to use it:

Hope this helps,

I have some doubts about this type of IK implementations. I have an URDF which I want to close with kinematics because it has a close chain. But i don’t know how to establish the relationship between the joints to compute it like if it was close. That’s the reason why I want to implement my own kinematics and I have found lots of problems with that. I think that Ikfast isn’t going to help me in this but if it can I want to know if you could help me to understand hoy

I would be really thankfull with you