Jacobian derivative about control-libraries

[buschbapti]

We had a discussion about that already hadn't we? I believe, but I am not sure, that usually you need dJ * v and there is a function to get that in pinocchio. The problem with the Jacobian derivative is that it would require the usage of tensors which are part of the Eigen-unsupported https://eigen.tuxfamily.org/dox/unsupported/eigen_tensors.html. We could include that as well but I am not familiar with it.

[buschbapti]

Actually I just checked the documentation of the function you mention and it is a 6 x n_joints matrix because it is the time derivative of the Jacobian and not Jacobian partial derivative (hessian). Are you sure this is what you need? In that case yes this could easily be added. I would recommend a nomenclature as compute_jacobian_time_derivative().

[patr3]

Actually I need to compute dJ*dq

[buschbapti]

Right indeed not dJ * v which is incorrect. I know there is such a function in pinocchio. Not sure what it is though.