Re: [eigen] Re: Rigid transformations in eigen: use of dual quaternions

[ Thread Index | Date Index | More Archives ]

> OK so we have several options here.
> The easiest is to add a method interpolate(q1,q2,s1,s2) to Quaternion,
> returning a Quaternion.
> It could even take care of normalizing the result (i guess you want that too?)
> Would that be enough for you?

I need quaternion multiplication with a scalar as well. And with this
function, it'd be a kludge.

> Otherwise, if we need expression templates in Quaternion, well I think
> I remember the big issue now: Quaternion already inherits
> RotationBase, so it would mean multiple inheritance if it also
> inherited MatrixBase. I didn't think more, so I don't know where to go
> from here.

How about making a brand new rigid transform class, you get all the
expression templates, size is fixed at compile time, so auto
vectorization. It could take as inputs the existing rotationXY,
translate, angleaxis classes as arguments and convert them to it;s
internal representation.

Rohit Garg

Senior Undergraduate
Department of Physics
Indian Institute of Technology

Mail converted by MHonArc 2.6.19+