Inverse Kinematics
Author
Hi guys - I have recently been looking into inverse Kinematics with a view to creating an animated crciket character to help with the coaching i do at a local club.
I have quite a good understanding of maths but am a little confused at the moment.
Anyway to start off with i was wondering whether any of you guys would be able to help me out with some refernces that could help me on my way to understanding what to do.
Cheers in advance for any and all suggestions.
Neil
hi !
i recommend that you take a look into the CCD-algorithm if you are looking for a fast and easy to implement method. i used it for my IK-system to animate a human. the following paper might help you :
http://64.233.183.104/search?q=cache:JZWI6DlyjPkJ:crpit.com/confpapers/CRPITV35Merrick.pdf+ccd+algorithm&hl=de
or just google it..
hope this helps and good luck
i recommend that you take a look into the CCD-algorithm if you are looking for a fast and easy to implement method. i used it for my IK-system to animate a human. the following paper might help you :
http://64.233.183.104/search?q=cache:JZWI6DlyjPkJ:crpit.com/confpapers/CRPITV35Merrick.pdf+ccd+algorithm&hl=de
or just google it..
hope this helps and good luck
cyclic coordinate descent is one relatively easy but imperfect way to go, you can also use lagrange multipliers and have joint weights to promote more natural poses.
alternatively, ls-dyna (a very high end piece of engineering software) uses spring constraints for dynamic simulations. if you do the same but apply an attractive force between the effector and the goal and iterate until the effector and goal positions are the same, you should be able to incorporate joint weights and still get natural poses with relatively simple and fast code. i think the basic explanation for this is on a website titles 'gaffer on games' or something similar.
james
alternatively, ls-dyna (a very high end piece of engineering software) uses spring constraints for dynamic simulations. if you do the same but apply an attractive force between the effector and the goal and iterate until the effector and goal positions are the same, you should be able to incorporate joint weights and still get natural poses with relatively simple and fast code. i think the basic explanation for this is on a website titles 'gaffer on games' or something similar.
james
Author
Cheers for the input guys i'll have a look at all those things suggested but as an utter newbie may well be back looking for an easier grounding in the subject at a later date!!!
Are you sure you are looking for the right thing? Your animated cricket character only needs IK if he needs to respond to an unpredictable environment (ie, lift his feet to climb stairs, or animate his elbows appropriately for incoming ball paths). If he simply needs to be animated, all you need is normal, forwards, kinematics, for which better search terms would be "skeletal animation".
they do need to catch the ball, dive, or bat it properly. If you have slo-mo, IK could be cool and give a much more realistic feel.
Here are 2 general articles on Inverse Kinematics(2 different methods written by the same person):
http://freespace.virgin.net/hugo.elias/models/m_ik.htm
http://freespace.virgin.net/hugo.elias/models/m_ik2.htm
http://freespace.virgin.net/hugo.elias/models/m_ik.htm
http://freespace.virgin.net/hugo.elias/models/m_ik2.htm
Quote:Original post by Saruman
IK is pretty much a requirement for any good sports game/simulation.
Really? How so?
All the sports games I have worked on or looked at do not use IK unless they have to. IK is widley uncontrollable, hard to tune and ultimately a frustrating process to work with. Blending hand or motion captured animations is a lot easier and the artists get control.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement