Talk:Rodrigues’ rotation formula

From Rosetta Code

Background

I needed code with no dependencies to model the legs of a walking robot. "Spot" style creatures like from Boston Dynamics are very popular these days.

I prototype in perl, but my robot is likely to use C (on Arduino) or maybe circuit-python - undecided as yet - but that's the reason for the "no dependencies" requirement of course. Either way, there was plenty of *math* everywhere showing how to do what I need, but no actual *code* I could find (besides one confused stackexchange student question in Javascript).

Hopefully this can help everyone who needs to rotate things around in 3D...


Task incomplete – lacks test data

We need a few test rotations – simple and edge case input data – for each contribution to show a result for. Hout (talk) 21:36, 29 September 2021 (UTC)

Suggestion - make up some test cases with good coverage (e.g. include unit rotation vectors in one plane, as well as non-unit non-planar tests)
Sounds sensible – you'll need to add whichever values you choose to the task description itself, so that contributors can compare output.
You'll also need to sign these contributions so that readers can have a hope of making head or tail of who is speaking :-) Just add four tilde characters at the end, for an automatically derived name and date-stamp. (colon multiples indent responses) Hout (talk) 13:53, 30 September 2021 (UTC)
All entries apart from FreeBASIC seem to be using the vectors { 5, -6, 4 } and { 8, 5, -30 }. No idea what the correct wording would be to add that to the task description. --Pete Lomax (talk) 09:40, 8 November 2021 (UTC)