Jump to content
  • Advertisement
Sign in to follow this  
lucky6969b

Two-handed Reaching with OpenTissue

This topic is 894 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I came across the following problem when I was using OpenTissue to solve the 2 handed reaching problem.

I found that there would be a "dragging" force on both arms which cancels out each other.

For example,

the left arm IK takes effect, it will drag the body towards the left side,

And the right arm IK also in effect, it will pull back the body towards the right side.

The end result is nothing.

 

So I decided to do it one after the other.

the first frame solves for the left arm, the second frame the right arm...etc

alternately.

 

But still didn't quite get any pleasing results.

Any advice for improving this way of tackling the problem?

Thanks

Jack

 

 

 

Share this post


Link to post
Share on other sites
Advertisement

You are my hero for helping me discover this library. I would give you +5 if I could, because this library looks like it can save me a lot of trouble.
 

Anyways. It's a two handed reaching problem. It often implies that both hands must be processes-ed at the same time and compared on a graph. And that there must be a base case in which the algorithm ends. Which means that they both depend on the torso.

The problem is that the torso is being evenly effected by both arms. When in reality if you ever tried reaching, there are actually points in which the torso will favor one arm, or will favor both at the same time and lock one of it's axis of motion.

So here's the deal.

If both arms should reach to the right. Then the torso must favor the left arm to position it evenly with the body. That means that the effecting force of the right arm has must scale down, and the effecting force of the left arm has over everything must scale up. Eventually both results will balance out and should theoretically produce a more visually pleasing effect. To reiterate, you are looking for a point of intersection on the force of the body where the forces cancel out, and you're going to stop there.

Try doing this in real life, and remember that the body is always in a state of tug of war. There's simply just stale mates that determines when the body is in the correct position that it needs.

These kind of problems suck when you don't compare it to the actual human body.

Edited by Tangletail

Share this post


Link to post
Share on other sites

What about controlling the dof of the torso instead, because I cannot find a user guide on

OpenTissue, there is some documentation on the library, but very limited.

I have to collect them piece by piece, cos there is no centralized manual or user guide...

What a pity.

Thanks

Jack

Share this post


Link to post
Share on other sites

The alternate approach works, and I had to make some constraints on the joint angles.

Seems like the weight is not implemented...

 

Final Result:

[attachment=32425:2016-06-29 21_54_17-TestSB.png]

 

 

Thanks

Jack

Edited by lucky6969b

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!