Jason K. Moore
Jason K. Moore
I've been doing some work on this. Haven't yet quite got it where I want it.
@pydy/pydy-developers I've almost got this example working. There are two issues (possible related) that I could use some more eyes helping me spot the bug: 1. I get different results...
@chrisdembia Would you have a few minutes to look this over? You should be able to run it with PyDy master.
No it should be minimal. We could just copy over the `dtk.bicycle` module (or relevant functions) to here too. I just have all the basumandall stuff coded up there as...
That's awesome! So there is now just something minor (and probably even harder to find...). Do you want to make a PR against this branch?
Yeh, I can just pull your branch and merge.
Note that I was able to get very high precision matching with my old Autolev code: http://moorepants.github.io/dissertation/eom.html#non-linear-validation, so I'm inclined to believe that this model still has a small issue,...
These two don't seem to give the same results for some reason. The lower one caused the results in the manual kanes equations functions to give slightly different final xd...
I've noticed that if I use `0.0`, `0`, or `sympy.S(0)` for the zero quantities when setting velocities to zero and the inertia scalars to zero that I get different results...
Oscar shared this on the SymPy list that could help with the cse comparisons: ```python rom sympy.utilities.iterables import postorder_traversal def cse_compare(expr, syms, vals): result = lambdify(syms, expr)(*vals) syms_all, vals_all =...