Diego Ferigo

A while ago I started playing a bit around with this type of inversion, but I couldn't get any better performance than calling `jax.numpy.linalg.inv`. I came up to the conclusion...

This is basically the updated version of the problem occurring in the previous OOP APIs in #84 cc @flferretti

I guess it needs https://github.com/ami-iit/rod/pull/27/commits/0ce60b22e18432f6a557e83f2a1100cb0bd4cdd4 included in https://github.com/ami-iit/rod/pull/27.

Also #83 actually needed https://github.com/ami-iit/rod/pull/27 for proper color support, we need to bump the dependencies before releasing jaxsim `v0.2.0`.

Thanks for the investigation @flferretti. Since `JaxSimModelData` contains many JAX arrays as attributes, I believe that the `__hash__` and `__eq__` methods need to be properly implemented. I'll have a look...

I had preliminary results of differentiating against the hardware parameters introduced in #101 (and refined in #120). I created a simple toy problem of a box falling for 1 second...

A good example that compares a derivative computed analytically w.r.t. the same quantity obtained by AD is: - #171

Another example of using AD is the validation of $\dot{M}(\mathbf{q}) = C(\mathbf{q}, \boldsymbol{\nu}) + C^\top(\mathbf{q}, \boldsymbol{\nu})$ performed in: - #172

> > These transform are complex. Can you try to set the w_H_B to identity and the joint position to zero, and try again? In that condition the W_H_F rotation...

> Now, since this test is passing for the majority of the frames and failing only for 6 of them, what I'm suspecting is that maybe is not always true...