ICE Kinematics IK Up Vector

[Sven Kempermann]

I'm playng around with the new ICE Kinematics, and it's a lot of fun. But I have a problem I'm not able to solve. I need an up-Vector for my IK Chain. Although I know the math for this task, I'm not able to get it working inside a ICE-Tree.. has anyone done this yet?

Regards,

Sven Kempermann

[Oscar Juarez]

nice, I was going to start tackling that task, how much time did it take you?

a tutorial would be great.

thanks for sharing

[Michal Doniec]

I've got basic analitycal 2D IK solver running in a couple of hours, but no upvector. Isn't that a just matter of controlling IK chain rotation axis by aim constraint?  (really impressed with ICE Kinematics).

[Eric Thivierge / XSI Database]

Anyone have any reference material on building an IK solver from
scratch? I'd like to read up on it a bit. Sorry to not be able to
contribute to the issue at hand.

Thanks,

--------------------------------------------
Eric Thivierge
Technical Director
http://www.ethivierge.com

[Sven Kempermann]

I think so, but I'm struggling to get this work together with the IK-Solver.

@Eric

I would also would appreciate such an information. Couldn't find anything on the web but mostly robotic stuff

2010/4/12 Michal Doniec <don...@gmail.com>

[Michal Doniec]

Advanced Animation and Rendering Techiques by Mark and Alan Watt

 

Here is an article which got me started

 

http://freespace.virgin.net/hugo.elias/models/m_ik2.htm

[Ahmidou Lyazidi]

Hi
Here is my IK compound with upvector support. It must be plugged in the FK to IK compound.
it is based on this code:
http://graphics.cs.cmu.edu/nsp/course/15-464/Spring07/assignments/jlander_gamedev_sept98.pdf
but was customized for my own purpose
Have fun!

Ahmidou Lyazidi
Director | TD | CG artist

2010/4/12 Michal Doniec <don...@gmail.com>

[Michal Doniec]

Just to quickly add to this article, there is an error in the equation 6 and in the code

It should be:

O1 =atan( -(L1*sin(O2) )*x+( L1+L2*cos(O2))*y / (L2 * sin(O2)) *y + (L1+L2*cos(O2))*x)

[Raffaele Fragapane]

For a two bones IK law of cosine is the starting point for everything and trivial to understand and implement. If you want more than two bones then it becomes non deterministic very quickly unless you introduce a fair few additional variables (2D, 2D with planar offsets, or 3D, stiffness factors to distribute angles within the solver and so on).
Law of cosine or similar doubled up for Z legs though will get you there for 99% of creatures. It's only for particular effects or things like tails, tentacles, effects rigging and so on that you will need something different really, and in those cases you usually aren't dealing with just solving between root and effector for a bunch of bones anymore anyway :)

[Robert Moodie]

An old skool classic:

http://www.isner.com/Transform/IsnerTransformManip_04.htm
http://www.isner.com/tutorials/quatSpells/quaternion_spells_14.htm

Helped me a lot...

_Rob

[Grahame Fuller]

Apologies if this is a double post.

The error is corrected in the follow-up article, which also goes on to discuss techniques for 3+ bone chains.

http://graphics.cs.cmu.edu/nsp/course/15-464/Spring07/assignments/jlander_gamedev_nov98.pdf

gray

From: softimag...@listproc.autodesk.com [mailto:softimag...@listproc.autodesk.com] On Behalf Of Michal Doniec
Sent: Monday, April 12, 2010 10:49 AM
To: soft...@listproc.autodesk.com
Subject: Re: ICE Kinematics IK Up Vector

[Grahame Fuller]

[Oscar Juarez]

thats a compound wich is inside the 2 bone IK compound

the fk to ik compound is the one that blends between FK and IK solver.

but Ahmidou's solver uses Y as bone direction so youll have to orient your controllers like that

for the FK

On Thu, Apr 15, 2010 at 9:43 AM, Eric Thivierge / XSI Database <ethiv...@gmail.com> wrote:

Hi Ahmidou,

Which compound do you speak up when you say that this compound has to
be plugged into the "FK to IK compound"?

--------------------------------------------
Eric Thivierge
Technical Director
http://www.ethivierge.com

[Oscar Juarez]

in the installation path, theres a workgroup called ICE Kinematics, connect to it, and you'll find all of those compounds.

On Thu, Apr 15, 2010 at 10:55 AM, Duncan Greenwood <dun...@blackginger.tv> wrote:

I think I missed something. What is the "2 bone IK compound" ? I've been trying to catch up on the last 3 months of list emails and I can't seem to find and anything like that. Also I can't find it in xsi 2011 or on google. Can someone please tell me where I can find it?

Thanks

Duncan Greenwood

[Eric Thivierge / XSI Database]

Thanks Oscar,

Didn't know where to find it or dig it out from. :)

--------------------------------------------
Eric Thivierge
Technical Director
http://www.ethivierge.com

On Thu, Apr 15, 2010 at 11:00 AM, Oscar Juarez

[Duncan Greenwood]

hey thanks Oscar. I wonder why they didn't just integrate that into
normal "out of the box" xsi setup though ?

>> <mailto:don...@gmail.com>>

>> >>
>> >> Advanced Animation and Rendering Techiques by Mark and
>> Alan Watt
>> >>
>> >> Here is an article which got me started
>> >>
>> >> http://freespace.virgin.net/hugo.elias/models/m_ik2.htm
>> >>
>> >>
>> >>
>> >> On Mon, Apr 12, 2010 at 1:54 PM, Sven Kempermann
>> >> <sven.kem...@googlemail.com

>> <mailto:sven.kem...@googlemail.com>> wrote:
>> >>>
>> >>> I think so, but I'm struggling to get this work together
>> with the
>> >>> IK-Solver.
>> >>> @Eric
>> >>> I would also would appreciate such an information.
>> Couldn't find anything
>> >>> on the web but mostly robotic stuff
>> >>>
>> >>> 2010/4/12 Michal Doniec <don...@gmail.com

>> <mailto:don...@gmail.com>>

>> >>>>
>> >>>> I've got basic analitycal 2D IK solver running in a
>> couple of hours, but
>> >>>> no upvector. Isn't that a just matter of controlling IK
>> chain rotation axis
>> >>>> by aim constraint? (really impressed with ICE Kinematics).
>> >>>>
>> >>>> On Sat, Apr 10, 2010 at 12:44 AM, Oscar Juarez
>> >>>> <tridi.a...@gmail.com

>> <mailto:tridi.a...@gmail.com>> wrote:
>> >>>>>
>> >>>>> nice, I was going to start tackling that task, how much
>> time did it
>> >>>>> take you?
>> >>>>> a tutorial would be great.
>> >>>>> thanks for sharing
>> >>>>> On Fri, Apr 9, 2010 at 5:52 PM, Sven Kempermann
>> >>>>> <sven.kem...@googlemail.com

[Eric Thivierge / XSI Database]

I don't believe those compounds are officially supported. I think many
were built by Phil Taylor for demo purposes only.

--------------------------------------------
Eric Thivierge
Technical Director
http://www.ethivierge.com

>>> � � � �>>>>
>>> � � � �>>>
>>> � � � �>>
>>> � � � �>
>>> � � � �>
>>>
>>>
>>
>

[Eric Thivierge / XSI Database]

Hi Ahmidou,

Which compound do you speak up when you say that this compound has to
be plugged into the "FK to IK compound"?

--------------------------------------------
Eric Thivierge
Technical Director
http://www.ethivierge.com

[Duncan Greenwood]

I think I missed something. What is the "2 bone IK compound" ? I've been trying to catch up on the last 3 months of list emails and I can't seem to find and anything like that. Also I can't find it in xsi 2011 or on google. Can someone please tell me where I can find it?

Thanks

Duncan Greenwood


Oscar Juarez wrote: