Jump to content

Procedural, intelligent, dynamic character animation!


Recommended Posts

Has anyone ever tried anything like this:

 

 

Looks like it would be great fun to try and Houdini could be perfect for it! Then if it was integrated with basic crowd sim (Houdini 14 crowd tools??) rules it could give some very interesting results! I'm working with Miarmy on a job at the moment and the main that it lacks is this kind of intelligence and procedural animation.. Would be great to have the best of both!

  • Like 1
Link to comment
Share on other sites

Not to be a wet blanket, but...

Looks like something that took 6 PHds 5 years, 4 million dollars, 30 additional people, working 20 hour days, with the sly backing of at least 1 centuries old world class University on call.

edit: Feeling grumpy, don't mind me. Reach for the stars.

Edited by kleer001
Link to comment
Share on other sites

I would love to learn and try out more of these endorphin type of behaviors. Not quite sure where to start with this though.

I am currently working on a project that requires quite a bit of procedural animation including swarming etc. and tried to integrate some of the ideas of this for the locomotion aspect of it:
http://www.gamasutra.com/view/news/216973/Video_An_indie_approach_to_procedural_animation.php

 

I think Houdini is about to reach a critical mass stage whereby enough pieces of the puzzle are falling into place:

- Houdini 14 crowds and agents

- bullet for fast dynamics

- vops for fuzzy logic & ik-handles

- constraint networks for interaction, limits, springyness

- dop pops with sensor information

 

-- I don't know enough yet what h14 is offering in regards to the crowds, but I would love to try getting even some basic behaviours into houdini. I do think a lot is mixed from dynamics. And then the other half would come motion capture inpired rules like 'protect head' or 'wave arms to rebalance center of gravity',...  and mixing them together with fuzzy logic or using state machines like in games.

 

-- I also think that to start it off, you could start with a core skeleton representation using points, then build a physics ragdoll using bullet and constraints, and the ai and behavior blending... the blending itself could be vops and sopsolver. But the 'self-learning' aspect of it... I don't know much about it yet.

Link to comment
Share on other sites

Well I think its possible! Endorphin is incredibly impressive but I think something similar but simpler is doable. Yeah I saw that eetu, I think thats where I would start and I think that would be really interesting to develop balance and realistic self restitution for characters. Theres loads of research on this (e.g. https://www.youtube.com/watch?v=ZCenGtuYy-I, https://www.youtube.com/watch?v=3Lbmmoofu0E).

 

Yea Peter I think the tools currently in Houdini could get you most of the way if not all of the way there. Anyway I think i'll have a look into it when I get some time, but if anyone has any examples of something similar in Houdini or good reference i'd love to see it!

Link to comment
Share on other sites

Thanks Martin, thats a great talk and really interesting to see their progression.

 

I think if I ever found time to look at this I would start with the balance stuff and trying to get some intelligence into rag doll characters, since this is what is lacking in software like Miarmy. There are some features (e.g. the servo feature in Miarmy - 

) but this is just a generic animation applied on top of the dynamics simulation and doesn't really add any intelligence. Actions, like Peter mentioned, such as protecting your head, arms out to break the fall, and balancing I think wouldn't be too hard to implement and can be a big step towards adding realism for rag doll characters. Edited by michaelb-01
Link to comment
Share on other sites

  • 2 years later...

Hey all thought I'd resurrect this older thread...  So are there any ways to obtain a sort of semi ragdoll effect on a crowd character?  I currently have them running around but it would be great to get a little ragdoll blending in there as they run into each other. 

Currently I can add a state in that let's them switch from animated to ragdoll when they collide. But I'm just curious if it can blend a percentage of ragdoll and anim together?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...