1
$\begingroup$

I am applying Recursive Newton Euler Inverse Dynamics to a simple two link model. If I apply the result torques and forces using PhysX or Bullet Physics I get unexpected results. The model spins and deforms instead of rising in the Y direction without rotating and deforming as the input acceleration indicates:

a busy cat

The Newton Euler Inverse Dynamics (tested with RBDL and Matlab spatial_v2). Returns:

Tau = 0,6,0, -6,0,0, -3,0,0 = (force A, Torque A, Torque B)

When I apply these forces and torques in PhysX and Bullet the bodies deform and rotate. The values that work in Bullet and PhysX are: Fa=0,6,0 Ta=3,0,0 Tb=3,0,0.

When I feed Tau into Forward Dynamics using Composite Body or Articulated Body then the original acceleration is returned.

I am wondering how one applies Tau to a model. Is there a transformation I am missing somewhere? All the joints in the model are aligned with the worldspace axis.

Any help would be appreciated. Thanks

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.