Timo Stienstra

[location](https://docs.sympy.org/latest/modules/matrices/matrices.html#sympy.matrices.matrices.MatrixBase.LUsolve) of `LUsolve` For `inv()` I would guess [this location](https://docs.sympy.org/latest/modules/matrices/matrices.html?highlight=inv#sympy.matrices.matrices.MatrixBase.inv). I could not find anything for `hadamard_product`.

The problem with automatic detection is that you only know the objective and some variables may be missing from there. If one unknown trajectory misses, then you by definition don't...

Another option would be to change [the signature of `Problem`](https://github.com/csu-hmc/opty/blob/6a23d17493341ad5698e61c6a2dd6e1353b47440/opty/direct_collocation.py#L94). This could also enhance user experience. This could look a bit like this: ```python # Assume obj and obj_grad are...

Not sure if this is behaviour we should want. The task of `ang_vel_in` is not to store the velocity, but to compute it. Personally I would keep this explicit, also...

It would be an option to set it, when `subs` is specified. Currently it does not save it, see [#23741](https://github.com/sympy/sympy/issues/23741). Personally, I do also have a bit of a preference...

Just tested the implementation with line profile for one of the Whipple bicycle models I've written. For comparison I left in the old implementation. It has a total of 8...

> This example uses the auxiliary speed functionality on the bicycle model: [pydy/pydy#462](https://github.com/pydy/pydy/pull/462) Could be a quick way to ensure the aux speeds are still handled. Thanks for sharing. Just...

@brocksam thank you for all the feedback. Just finished updating the docs and code accordingly. Will be updating some more to pass the failing tests.

@brocksam thank you for all the feedback.

A disadvantage about the current check, whether the kinematical differential equations (kdes) defined in joints are also used in `System` can be a little frustrating. Just had the problem myself...