How To Connect The Kinematic Model An Visual Node The Right Way?

I’m in the process of writing an COLLADA 1.5 conform importer.

When it comes to kinematics I’m not 100% sure how to implement them the right way. I could not find the answer in the specs though.

I have a couple of sample COLLADA Files from different exporters, that seem to expect different behavior in terms of how to bind visual nodes to joints.

One requires the axis to return the relative matrix thus combining multiple axes in the visual scenes explicitly

If the IK parts are one-quarter as bad as Physics you’re in for a ride. The specs are not always coherent. You kind of have to intuit what the authors are thinking. Existing COLLADA software tends to be full of mistakes and corner cutting. So your judgement is as good as or better than any examples. Use the examples (other’s work) to see if they spark a moment of insight, but your own interpretation of the specification is best.

I wish I could be of more help. The only mention of IK I’ve ever seen is with OpenRave, since one of its principal authors was around when the 1.5 specification was being drafted. But the OpenRave physics interpretation is wrong, I can tell you. I am rewriting COLLADA-DOM, which is kind of interesting, because the COLLADA authors were using it in house to test out ideas, so you can glean from it a good idea of what they did more in practice than theory. It doesn’t have any IK or anything from 1.5.0. I think 1.5.0 was rushed to try to correct issues in the earlier specification and add things, but it didn’t get used before it seems like everyone lost interest in COLLADA.